Activity
From 12/03/2015 to 01/01/2016
01/01/2016
- 06:17 PM Revision bc1cf95f (ceph): uniform the type format
- Signed-off-by: Ning Yao <zay11022@gmail.com>
- 06:09 PM Revision 56a964b7 (ceph): os/bluestore: write 'type' file
- Signed-off-by: Sage Weil <sage@redhat.com>
- 06:09 PM Revision 20dc14bf (ceph): os/bluestore: Don't forget close fd in _setup_block_symlink_or_file.
- Signed-off-by: Jianpeng Ma <jianpeng.ma@intel.com>
- 06:09 PM Revision e7a84fec (ceph): os/bluestore/BlueFS: remove deleted file from dirty list
- Signed-off-by: Sage Weil <sage@redhat.com>
- 06:09 PM Revision 37c6390a (ceph): qa/workunits/cephtool/test.sh: leave sortbitwise set
- New backends don't work if it's off.
Signed-off-by: Sage Weil <sage@redhat.com> - 06:09 PM Revision d6043d9a (ceph): fs: fix unclosed formatter session
- Fixes: #14210
Signed-off-by: xie.xingguo <xie.xingguo@zte.com.cn> - 06:09 PM Revision 757152f9 (ceph): os/bluestore/BlueStore: fix overlay keys
- Signed-off-by: YiQiang Chen <cyqsign@163.com>
- 06:09 PM Revision 7dddecb8 (ceph): ceph-objectstore-tool: get store type from 'type' file, if present
- Just like ceph-osd.
Signed-off-by: Sage Weil <sage@redhat.com> - 06:09 PM Revision ee01a7a6 (ceph): ceph-objectstore-tool: don't require journal path
- Assume journal symlink is present.
Signed-off-by: Sage Weil <sage@redhat.com> - 06:08 PM Revision 64b80f32 (ceph): ceph_test_objectstore: close cct on shutdown
- This lets us do leak checking.
Signed-off-by: Sage Weil <sage@redhat.com> - 06:08 PM Revision 7cb901c2 (ceph): os/bluestore/BlueFS: fix a few leaks
- Signed-off-by: Sage Weil <sage@redhat.com>
- 06:08 PM Revision 03362796 (ceph): os/bluestore/BlueFS: delay IOContext dtor until after io completes
- It's possible for the IO to be in flight when the caller closes the
writer handle (although dangerous of them). Queu... - 06:08 PM Revision 7172a7e2 (ceph): os/bluestore/BlockDevice: clean up aio completion a bit
- Signed-off-by: Sage Weil <sage@redhat.com>
- 06:08 PM Revision 198b156c (ceph): os/bluestore/BlueFS: wait for flush aios to complete before returning
- This makes all the IOs effectively synchronous (while holding the lock),
which isn't great, but at least it's correct... - 06:08 PM Revision 3fa09f02 (ceph): os/bluestore: be tolerant of collection_list bounds
- In particular, we may get a shard specified along with hobject_t min or
max (from PGBackend::objects_list_range()).
... - 06:08 PM Revision fc9d9522 (ceph): vstart.sh: k=2 for ec
- It didn't like k=1 with the default profile.
Signed-off-by: Sage Weil <sage@redhat.com> - 06:08 PM Revision a3ef0ec6 (ceph): vstart.sh: no need for 'quit' to ceph cli
- Signed-off-by: Sage Weil <sage@redhat.com>
- 06:08 PM Revision acc8217a (ceph): os/bluestore/BlockDevice: fix FS leak
- Signed-off-by: Sage Weil <sage@redhat.com>
- 06:08 PM Revision ef262055 (ceph): kv/RocksDBStore: fix leak of custom Env
- Signed-off-by: Sage Weil <sage@redhat.com>
- 06:08 PM Revision 3f4bd467 (ceph): os/bluestore/BlueFS: fix leak of iocs in writer
- Signed-off-by: Sage Weil <sage@redhat.com>
- 06:08 PM Revision 3f8e2bd8 (ceph): os/bluestore/BlueRocksEnv: fix several leaks
- Signed-off-by: Sage Weil <sage@redhat.com>
- 06:08 PM Revision 5950eae8 (ceph): ceph_test_objectstore: clean out test dir for each test
- Signed-off-by: Sage Weil <sage@redhat.com>
- 06:08 PM Revision b57fcf60 (ceph): os/kstore: fix bugs in read stripe logic
- Signed-off-by: Sage Weil <sage@redhat.com>
- 06:08 PM Revision e4e1a7f4 (ceph): os/kstore: debug nid
- Signed-off-by: Sage Weil <sage@redhat.com>
- 06:08 PM Revision 5c626908 (ceph): common/buffer: make hexdump look like 'hexdump -C ...'
- ...the better to diff them with.
Signed-off-by: Sage Weil <sage@redhat.com> - 06:08 PM Revision 645b5350 (ceph): os/kstore: cache in-flight stripe updates in memory
- The write process may do a read/modify/write on a stripe. In order to
allow multiple writes to coexist within the sa... - 06:08 PM Revision 491e2ce4 (ceph): os/Makefile: fix typo
- Signed-off-by: Sage Weil <sage@redhat.com>
- 06:08 PM Revision 88c3ee81 (ceph): CMakeLists: update for os/ reorg
- Signed-off-by: Sage Weil <sage@redhat.com>
- 06:08 PM Revision d2b5e926 (ceph): os/fs: move btrfs_ioctl.h to os/fs
- Signed-off-by: Sage Weil <sage@redhat.com>
- 06:08 PM Revision b16f591b (ceph): os/Makefile: cosmetic
- Signed-off-by: Sage Weil <sage@redhat.com>
- 06:08 PM Revision 893bfc7e (ceph): os/bluestore/bluestore_types: fix bluestore_extent_ref_map_t test inst
- Signed-off-by: Sage Weil <sage@redhat.com>
- 06:08 PM Revision 602b8de8 (ceph): unittest_bluestore_types: fix warnings
- Signed-off-by: Sage Weil <sage@redhat.com>
- 06:08 PM Revision 34a0f30b (ceph): os/bluestore/StupidAllocator: be less stupid about alloc_unit
- Make sure the extent we return is aligned to alloc_unit. If the first
extent isn't suitable, try the next one.
Keep... - 06:08 PM Revision 97373a77 (ceph): osd: do not catch mkfs exceptions
- We either want to take a clean error path, or dump core.
Signed-off-by: Sage Weil <sage@redhat.com> - 06:08 PM Revision 81243c05 (ceph): os/kstore: clear coll_map on umount
- Signed-off-by: Sage Weil <sage@redhat.com>
- 06:08 PM Revision ba2cc1eb (ceph): os/filestore: move FileStore to os/filestore/*
- Signed-off-by: Sage Weil <sage@redhat.com>
- 06:08 PM Revision 82cbc079 (ceph): os/memstore: move MemStore into os/memstore/*
- Signed-off-by: Sage Weil <sage@redhat.com>
- 06:08 PM Revision 98a0e107 (ceph): os/keyvaluestore: move KeyValueStore into os/keyvaluestore/*
- Signed-off-by: Sage Weil <sage@redhat.com>
- 06:08 PM Revision 8fc65e1b (ceph): os/ZFS: move to os/fs
- Signed-off-by: Sage Weil <sage@redhat.com>
- 06:07 PM Revision 193d25f7 (ceph): os/bluestore/StupidAllocator: add reservation assertions
- Signed-off-by: Sage Weil <sage@redhat.com>
- 06:07 PM Revision 2b2cc88d (ceph): os/bluestore/StupidAllocator: fix reservation release on allocate
- Use up the amount we allocated, not the amount we asked for.
Signed-off-by: Sage Weil <sage@redhat.com> - 06:07 PM Revision 6641482c (ceph): os/bluestore/Allocator: add unreserve()
- Signed-off-by: Sage Weil <sage@redhat.com>
- 06:07 PM Revision 5b57b098 (ceph): os/bluestore: allocate shouldn't fail if reserve succeeded
- Signed-off-by: Sage Weil <sage@redhat.com>
- 06:07 PM Revision f941642c (ceph): remove ceph_streamtest
- This is an ancient obsolete test that is never used.
Signed-off-by: Sage Weil <sage@redhat.com> - 06:07 PM Revision 6c1ef475 (ceph): ceph-objectstore-tool: drop unused FileSTore include
- Signed-off-by: Sage Weil <sage@redhat.com>
- 06:07 PM Revision e07ab3e2 (ceph): ceph:small_io_bench_dumb: drop unused include
- Signed-off-by: Sage Weil <sage@redhat.com>
- 06:07 PM Revision da9266d0 (ceph): os/ObjectStore: add fsck to interface
- Only bluestore and kstore implement this currently.
Signed-off-by: Sage Weil <sage@redhat.com> - 06:07 PM Revision 5f87bab2 (ceph): ceph-objectstore-tool: add fsck command
- Signed-off-by: Sage Weil <sage@redhat.com>
- 06:07 PM Revision 0cde5890 (ceph): ceph_test_objectstore: add omap iterator test
- Signed-off-by: Sage Weil <sage@redhat.com>
- 06:07 PM Revision 0f85d86c (ceph): osd/ReplicatedPG: print omap_digest in debug log
- Signed-off-by: Sage Weil <sage@redhat.com>
- 06:07 PM Revision b25a2428 (ceph): os/bluestore: do not include header in iterator result
- Signed-off-by: Sage Weil <sage@redhat.com>
- 06:07 PM Revision 624e9968 (ceph): os/kstore: do not include header in omap iterator result
- Signed-off-by: Sage Weil <sage@redhat.com>
- 06:07 PM Revision cff5dc5d (ceph): os/bluestore/BlockDevice: helpful error when aio cannot init
- Signed-off-by: Sage Weil <sage@redhat.com>
- 06:07 PM Revision 93964a72 (ceph): os/bluestore/BlueFS: do not flush metadata on flush()
- That's what fsync is for. Moreover, this can lead to some squirreliness
if we trigger this from _flush_log().
Signe... - 06:07 PM Revision 56375c59 (ceph): os: remove {get,set}_allow_sharded_objects from interface
- We've already forced everyone to upgrade through hammer, so everyone
supports this. Just unconditionally set the fea... - 06:07 PM Revision fd2659dd (ceph): osd/PGBackend: fix omap digest error message
- Print the *omap* digest, not *data* digest.
Signed-off-by: Sage Weil <sage@redhat.com> - 06:07 PM Revision 9aca4ad9 (ceph): os/bluestore: clear onode in _do_remove
- Signed-off-by: Sage Weil <sage@redhat.com>
- 06:07 PM Revision 789e99cf (ceph): os/kstore: clear onode on _do_remove
- Signed-off-by: Sage Weil <sage@redhat.com>
- 06:07 PM Revision de8a99d1 (ceph): ceph_test_objectstore: enable bluestore debug options
- Signed-off-by: Sage Weil <sage@redhat.com>
- 06:07 PM Revision 0a6f35dc (ceph): os/bluestore: optimize _dump_onode slightly
- Signed-off-by: Sage Weil <sage@redhat.com>
- 06:07 PM Revision e25cf208 (ceph): os/bluestore: debug msg on statfs
- Signed-off-by: Sage Weil <sage@redhat.com>
- 06:07 PM Revision 8e965db8 (ceph): os/bluestore: simplify _do_remove
- Signed-off-by: Sage Weil <sage@redhat.com>
- 06:07 PM Revision 225d3cdb (ceph): os/bluestore/BlockDevice: add invalidate_cache
- Signed-off-by: Sage Weil <sage@redhat.com>
- 06:07 PM Revision 5662bd8b (ceph): os/bluestore/BlueFS: implement invalidate_cache
- Signed-off-by: Sage Weil <sage@redhat.com>
- 06:07 PM Revision d48a2ba9 (ceph): os/bluestore/FreelistManager: audit
- Signed-off-by: Sage Weil <sage@redhat.com>
- 06:07 PM Revision fd75b4e4 (ceph): os/bluestore/bluestore_types: add extent FLAG_COW_{HEAD,TAIL}
- Signed-off-by: Sage Weil <sage@redhat.com>
- 06:07 PM Revision 64b4e2f6 (ceph): os/bluestore: Enode infrastructure
- Enodes will track extent ref counts for any extent
that is marked shared. There will be an enode for
any unique hash... - 06:07 PM Revision 8ebb3901 (ceph): os/bluestore/BlueStore: only allow clone if hash matches
- Signed-off-by: Sage Weil <sage@redhat.com>
- 06:07 PM Revision b758d9fb (ceph): os/bluestore: support copy-on-write clones
- Signed-off-by: Sage Weil <sage@redhat.com>
- 06:07 PM Revision 4aa5f025 (ceph): os/bluestore/BlueStore: fix _zero when previous extent partially unwritten
- Signed-off-by: Sage Weil <sage@redhat.com>
- 06:07 PM Revision a684a590 (ceph): os/bluestore/StupidAllocator: bluestore_debug_small_allocations
- Force small allocations for debugging purposes.
Signed-off-by: Sage Weil <sage@redhat.com> - 06:07 PM Revision 5534d2b5 (ceph): os/bluestore/BlueStore: fsck_on_umount
- Signed-off-by: Sage Weil <sage@redhat.com>
- 06:07 PM Revision a583e5de (ceph): ceph_test_objectstore: do Synthetic tests over larger objects
- 400k for objects.
Signed-off-by: Sage Weil <sage@redhat.com> - 06:07 PM Revision 2f9be0cc (ceph): os/bluestore: add bluestore_debug_no_reuse_blocks
- This makes debugging a bit easier because we never use the same
extent of the disk twice, leaving useful evidence beh... - 06:07 PM Revision d927919a (ceph): os/bluestore/BlockDevice: fix waiter wakeup use-after-free race
- Signed-off-by: Sage Weil <sage@redhat.com>
- 06:07 PM Revision 97bf3435 (ceph): os/bluestore/BlueStore: wal_op_t::OP_COPY
- Assume block-aligned.
Signed-off-by: Sage Weil <sage@redhat.com> - 06:07 PM Revision 45801ed4 (ceph): os/bluestore/bluestore_types: add contains(), clear(), empty() to exten...
- Signed-off-by: Sage Weil <sage@redhat.com>
- 06:07 PM Revision e93e9262 (ceph): unittest_bluefs, unittest_bluestore_types
- These should run during make check.
Signed-off-by: Sage Weil <sage@redhat.com> - 06:07 PM Revision 2162990a (ceph): ceph_test_objectstore: fix locking for a few ops
- Signed-off-by: Sage Weil <sage@redhat.com>
- 06:07 PM Revision c9659f99 (ceph): ceph_test_objectstore: save map lookups for a few ops
- Signed-off-by: Sage Weil <sage@redhat.com>
- 06:07 PM Revision a77363c9 (ceph): ceph_test_objectstore: debug enter/exit points
- Signed-off-by: Sage Weil <sage@redhat.com>
- 06:07 PM Revision ac08664b (ceph): ceph_test_objectstore: validate full object contents after writes
- Signed-off-by: Sage Weil <sage@redhat.com>
- 06:07 PM Revision f884edfa (ceph): ceph_test_objectstore: add many clone tests
- Signed-off-by: Sage Weil <sage@redhat.com>
- 06:07 PM Revision 82d21321 (ceph): ceph_test_objectstore: dump actual vs expected on read data mismatch
- Signed-off-by: Sage Weil <sage@redhat.com>
- 06:07 PM Revision eeb2251e (ceph): ceph_test_objectstore: use a few hash values for objects; clone between...
- We only guarantee support for clone between objects with the same hash.
Signed-off-by: Sage Weil <sage@redhat.com> - 06:07 PM Revision dc5d33b1 (ceph): os/bluestore/BlueStore: note wal releases in fsck
- Include these in used_blocks (they are about to be released but
not reflected in the onode).
Signed-off-by: Sage Wei... - 06:07 PM Revision 27e44dd8 (ceph): os/bluestore: clean up comments a bit
- Signed-off-by: Sage Weil <sage@redhat.com>
- 06:07 PM Revision 3133eca9 (ceph): os/bluestore: add some slow debug path
- Signed-off-by: Sage Weil <sage@redhat.com>
- 06:07 PM Revision 7e86bfd8 (ceph): ceph_test_objectstore: clone objects with same hash
- Signed-off-by: Sage Weil <sage@redhat.com>
- 06:07 PM Revision 48595ad4 (ceph): ceph_test_objectstore: clone non-empty objects, not empty ones
- This condition was backwards.
Signed-off-by: Sage Weil <sage@redhat.com> - 06:07 PM Revision 1769163e (ceph): ceph_test_objectstore: simplify object name generation
- The long names don't exercise useful code paths, and having
consistent naming makes it easier to grep through logs.
... - 06:07 PM Revision 549a335f (ceph): ceph_test_objectstore: fix clone
- Copy the buffer, in case other threads modify it in place.
Signed-off-by: Sage Weil <sage@redhat.com> - 06:07 PM Revision 6d41758c (ceph): os/bluestore/BlueStore: do WAL ops buffered to avoid RMW issues
- We may have multiple WAL ops that do read/modify/write covering
the same blocks. To avoid the complexity of identify... - 06:07 PM Revision e63dbdac (ceph): os/bluestore/BlockDevice: fix path
- Signed-off-by: Sage Weil <sage@redhat.com>
- 06:07 PM Revision 95b81e51 (ceph): os/bluestore/BlockDevice: adjust debug output
- 5 helpful (read/write offsets)
10 more, with aio completions
20 everything
30 fire hose
40 data hexdumps
Signed-off-... - 06:07 PM Revision bf1f02c1 (ceph): os/bluestore/BlueStore: remove unused OnodeMap::remove
- We install negative entries instead.
Signed-off-by: Sage Weil <sage@redhat.com> - 06:07 PM Revision c206832c (ceph): os/bluestore/BlueStore: fix rename
- Install a negative onode entry at the old name position.
Otherwise, a simple transaction like
rename a -> b
touch ... - 06:07 PM Revision 134a0e4c (ceph): os/kstore: fix rename
- Signed-off-by: Sage Weil <sage@redhat.com>
- 06:07 PM Revision 8941e325 (ceph): os/bluestore/BlueStore: fix read bug when there is a hole
- Signed-off-by: Sage Weil <sage@redhat.com>
- 06:07 PM Revision 1605e04d (ceph): os/bluestore/BlueFS: ignore flush when buffer is small
- Rocksdb does a flush after every append, each of which is often
less than a full block. This is very inefficient whe... - 06:07 PM Revision be598b7f (ceph): os/bluestore/BlueFS: do not dirty file when overwriting bytes
- The rocksdb log recycle option allows us to overwrite previously
allocated space in an old log file to avoid updating... - 06:07 PM Revision cf5d139a (ceph): os/bluestore/BlueFS: get_usage()
- Return (and log) usage for all bdevs.
Signed-off-by: Sage Weil <sage@redhat.com> - 06:07 PM Revision 3d2bb7a7 (ceph): os/bluestore/BlueStore: use BlueFS::get_usage()
- ...just so we log bdev utilization in the log.
Signed-off-by: Sage Weil <sage@redhat.com> - 06:07 PM Revision a3de5b68 (ceph): os/bluestore/BlockDevice: rename bdev options
- Signed-off-by: Sage Weil <sage@redhat.com>
- 06:07 PM Revision 9be0b537 (ceph): os/bluestore: handle both buffered and direct+async IO
- Prefer aio unless explicitly directed otherwise.
Signed-off-by: Sage Weil <sage@redhat.com> - 06:07 PM Revision e07fc756 (ceph): rocksdb: debug log writes/reads
- Signed-off-by: Sage Weil <sage@redhat.com>
- 06:07 PM Revision 510b4c78 (ceph): os/bluestore: don't create block.{db,wal} by default
- Signed-off-by: Sage Weil <sage@redhat.com>
- 06:07 PM Revision 4ebe78cc (ceph): os/bluestore/BlueStore: better error msg for bdev label check
- Signed-off-by: Sage Weil <sage@redhat.com>
- 06:07 PM Revision 6925a066 (ceph): os/fs/FS.h: fix aio_t::pread
- Allocate aligned buffer.
Signed-off-by: Sage Weil <sage@redhat.com> - 06:07 PM Revision 21583711 (ceph): os/bluestore: fail mount of fsck finds errors
- Signed-off-by: Sage Weil <sage@redhat.com>
- 06:07 PM Revision 9df3d967 (ceph): script/crash_bdev: simple script to inject bdev failures
- Signed-off-by: Sage Weil <sage@redhat.com>
- 06:07 PM Revision b1df8cbc (ceph): os/bluestore: better debugging on fsck alloc errors
- Signed-off-by: Sage Weil <sage@redhat.com>
- 06:07 PM Revision 5e4d3d29 (ceph): os/bluestore: update freelist in individual transactions
- We submit each operation's transaction individually to rocksdb,
and then since a final transction to flush them all. ... - 06:07 PM Revision aa844e58 (ceph): Makefile-rocksdb.am: update
- Signed-off-by: Sage Weil <sage@redhat.com>
- 06:07 PM Revision 3f8d46c6 (ceph): rocksdb: fix recycle replay
- Signed-off-by: Sage Weil <sage@redhat.com>
- 06:07 PM Revision 8c04a3be (ceph): os/bluestore/BlueStore: fix error path if label set fails
- Reported-by: David Zafman <dzafman@redhat.com>
Signed-off-by: Sage Weil <sage@redhat.com> - 06:07 PM Revision 06334688 (ceph): rpm, debian: package ceph-bluefs-tool
- Signed-off-by: Sage Weil <sage@redhat.com>
- 06:07 PM Revision 6b28449f (ceph): os/bluestore: bluestore bluefs = true
- Signed-off-by: Sage Weil <sage@redhat.com>
- 06:07 PM Revision 993c87fb (ceph): os/bluestore: fix fsck contains vs intersects
- Any overlap is an error.
Signed-off-by: Sage Weil <sage@redhat.com> - 06:07 PM Revision 1f05196e (ceph): vstart.sh: less noisy debug
- Signed-off-by: Sage Weil <sage@redhat.com>
- 06:07 PM Revision 02bd7d5e (ceph): ceph_objectstore_test: fix warning
- Signed-off-by: Sage Weil <sage@redhat.com>
- 06:07 PM Revision 196786d8 (ceph): os/bluestore/BlueStore: fix object key decode with key
- Signed-off-by: Sage Weil <sage@redhat.com>
- 06:07 PM Revision 499b40f5 (ceph): os/kstore/KStore: fix object key decode with key
- Signed-off-by: Sage Weil <sage@redhat.com>
- 06:07 PM Revision 197c5ae2 (ceph): os/bluestore: clear coll_map on umount, fsck finish
- Signed-off-by: Sage Weil <sage@redhat.com>
- 06:07 PM Revision 08f919eb (ceph): kv/RocksDBStore: behave if options string is empty
- Signed-off-by: Sage Weil <sage@redhat.com>
- 06:07 PM Revision 65e5ff4f (ceph): os/bluestore/BlueStore: fix _open_bdev() failure path
- Signed-off-by: Sage Weil <sage@redhat.com>
- 06:07 PM Revision 22073a08 (ceph): os/bluestore: default to 64k min_alloc_size
- Signed-off-by: Sage Weil <sage@redhat.com>
- 06:07 PM Revision 669bec7b (ceph): os/kstore: add new KStore backend
- This is based on BlueStore, but with all of the block-related code
and complexity ripped out, and a simple striping s... - 06:07 PM Revision e3affe6e (ceph): os/KeyValueStore: drop kinetic #include
- Signed-off-by: Sage Weil <sage@redhat.com>
- 06:06 PM Revision 5fb09a9a (ceph): os/bluestore/BlockDevice: inject block failures
- Signed-off-by: Sage Weil <sage@redhat.com>
- 06:06 PM Revision a17fe3e5 (ceph): os/bluestore: statfs
- Signed-off-by: Sage Weil <sage@redhat.com>
- 06:06 PM Revision 423f797d (ceph): os/bluestore: record kv backend
- Record kv backend at mkfs time instead of relying on current value
of config option.
Signed-off-by: Sage Weil <sage@... - 06:06 PM Revision 7664ffa6 (ceph): os/bluestore/FreelistManager: drop unused db ref
- Signed-off-by: Sage Weil <sage@redhat.com>
- 06:06 PM Revision 648f1328 (ceph): os/bluestore: add extent_ref_map_t
- This will be used to refcount extents for some subset
of the store (objects with same name or hash value?).
Signed-o... - 06:06 PM Revision bc0f2e66 (ceph): os/bluestore/bluestore_types: localize types
- Prefix with bluestore_
Signed-off-by: Sage Weil <sage@redhat.com> - 06:06 PM Revision 07a47a3b (ceph): rocksdb: pull up to master, include EnvMirror
- Signed-off-by: Sage Weil <sage@redhat.com>
- 06:06 PM Revision de300b1c (ceph): os/bluestore/BlueStore: drop internal EnvMirror
- Signed-off-by: Sage Weil <sage@redhat.com>
- 06:06 PM Revision f80b3359 (ceph): os/bluestore/BlueFS: all overwrites on open_for_write
- rocksdb will occasionally overwrite an existing file
if it is not present/valid in the manifest.
Signed-off-by: Sage... - 06:06 PM Revision 80943557 (ceph): os/bluestore: less debug noise
- Signed-off-by: Sage Weil <sage@redhat.com>
- 06:06 PM Revision 485557af (ceph): os/bluestore: block.db support
- Support a mid- to fast device that will preferentially
store the rocksdb data (and wal, if block.wal is not
present).... - 06:06 PM Revision c96e025a (ceph): ceph_test_objectstore: clean up synthetic collections
- Signed-off-by: Sage Weil <sage@redhat.com>
- 06:06 PM Revision b4879878 (ceph): os/bluestore/BlockDevice: restructure interface
- use atomics, do not track in-flight extents or magically cope
with racing ios (that is the users responsibility).
Si... - 06:06 PM Revision 9f114ac2 (ceph): os/bluestore: disable overlay for now
- Signed-off-by: Sage Weil <sage@redhat.com>
- 06:06 PM Revision 02605a66 (ceph): os/bluestore/BlueStore: fix zero gap bug
- Signed-off-by: Sage Weil <sage@redhat.com>
- 06:06 PM Revision 3745afb4 (ceph): os/bluestore: support second block.wal device
- Use this device for the bluefs log.
Signed-off-by: Sage Weil <sage@redhat.com> - 06:06 PM Revision 2e1edef3 (ceph): os/bluestore/BlueFS: fix replay of unlink
- Signed-off-by: Sage Weil <sage@redhat.com>
- 06:06 PM Revision 9565f0d8 (ceph): os/bluestore/BlueFS: flush log if needed
- If a file has dirty metadata (but no dirty data), we
still need to flush the log when it is flushed.
Signed-off-by: ... - 06:06 PM Revision 31307a5a (ceph): os/bluestore: label all block devices
- Label all of our block devices with a simple label
that includes the osd_uuid. Wire this into the
ObjectStore and OS... - 06:06 PM Revision db754e7d (ceph): os/bluestore/BlockDevice: check aio return values
- Signed-off-by: Sage Weil <sage@redhat.com>
- 06:06 PM Revision f6f4ed3d (ceph): os/bluestore/BlockDevice: fix alignment check
- Signed-off-by: Sage Weil <sage@redhat.com>
- 06:06 PM Revision e3fd2795 (ceph): os/bluestore/BlockDevice: debug read result
- Signed-off-by: Sage Weil <sage@redhat.com>
- 06:06 PM Revision ef06380b (ceph): os/bluestore/BlockDevice: lock device while open
- Signed-off-by: Sage Weil <sage@redhat.com>
- 06:06 PM Revision ccce793f (ceph): os/bluestore: reenable rocksdb recycling
- Signed-off-by: Sage Weil <sage@redhat.com>
- 06:06 PM Revision 13655fbb (ceph): os/bluestore/BlueFS: fix writes spanning extents
- Signed-off-by: Sage Weil <sage@redhat.com>
- 06:06 PM Revision 1727cebd (ceph): os/bluestore/BlueFS: fix overwrite
- Signed-off-by: Sage Weil <sage@redhat.com>
- 06:06 PM Revision ac05b4c1 (ceph): ceph_test_objectstore: trivial init fix
- Signed-off-by: Sage Weil <sage@redhat.com>
- 06:06 PM Revision b0730285 (ceph): os/bluestore/BlueFS: fix read
- Signed-off-by: Sage Weil <sage@redhat.com>
- 06:06 PM Revision dd901498 (ceph): os/bluestore/BlueFS: simplify extent list
- Merge contiguous extents.
Signed-off-by: Sage Weil <sage@redhat.com> - 06:06 PM Revision 73adec4c (ceph): os/bluestore/BlueFS: periodically compact log
- Rewrite only the current metadata in a fresh log
periodically to free log space.
Signed-off-by: Sage Weil <sage@redh... - 06:06 PM Revision 9785bc98 (ceph): vstart.sh: debug bluefs and rocksdb
- Signed-off-by: Sage Weil <sage@redhat.com>
- 06:06 PM Revision 05be4c6c (ceph): os/bluestore/BlueFS: prevent read+write sharing
- Signed-off-by: Sage Weil <sage@redhat.com>
- 06:06 PM Revision e7cce09c (ceph): os/bluestore/BlueFS: avoid lock during reads
- Signed-off-by: Sage Weil <sage@redhat.com>
- 06:06 PM Revision e4f6148c (ceph): os/bluestore/BlueStore: share space with BlueFS
- Signed-off-by: Sage Weil <sage@redhat.com>
- 06:06 PM Revision 2d053785 (ceph): os/bluestore/BlueFS: many fixes
- Signed-off-by: Sage Weil <sage@redhat.com>
- 06:06 PM Revision b8630ee4 (ceph): ceph-bluefs-tool: simple tool to export bluefs content
- Currently we just do a dump. We'll add more
functionality later.
Signed-off-by: Sage Weil <sage@redhat.com> - 06:06 PM Revision 98485dee (ceph): os/bluestore/BlueFS: readdir list dirs, too
- Signed-off-by: Sage Weil <sage@redhat.com>
- 06:06 PM Revision 3649a80a (ceph): os/bluestore/BlueFS: ref count BlueFS::File *
- There are FileWriters that exist when the file is
deleted.
Signed-off-by: Sage Weil <sage@redhat.com> - 06:06 PM Revision 9341eec5 (ceph): kv/RocksDBStore: rocksdb_separate_wal_dir option
- Signed-off-by: Sage Weil <sage@redhat.com>
- 06:06 PM Revision 1ffd5e69 (ceph): os/bluestore/StupidAllocator: fix locking
- Signed-off-by: Sage Weil <sage@redhat.com>
- 06:06 PM Revision 1b8d5b60 (ceph): os/bluestore/BlueStore: fix _do_read
- Signed-off-by: Sage Weil <sage@redhat.com>
- 06:06 PM Revision 226b3476 (ceph): ceph_test_objectstore: less verbose on hash collision test
- Signed-off-by: Sage Weil <sage@redhat.com>
- 06:06 PM Revision 6f5ac501 (ceph): ceph_test_objectstore: less verbose
- Signed-off-by: Sage Weil <sage@redhat.com>
- 06:06 PM Revision dd043917 (ceph): os/bluestore/BlueFS: simple file system to back rocksdb
- BlueFS is a simple file system that will back rocksdb.
BlueRocksEnv is the rocksdb::Env implementation that
glues the... - 06:06 PM Revision 653882c4 (ceph): os/bluestore/BlockDevice: move to simple mutex model
- Just for now, while we get the rest of this working.
Signed-off-by: Sage Weil <sage@redhat.com> - 06:06 PM Revision a62ffb0d (ceph): newstore -> bluestore
- Signed-off-by: Sage Weil <sage@redhat.com>
- 06:06 PM Revision 9d01b8df (ceph): os/bluestore: separate Allocator from freelist storage
- FreelistManager perists our freelist. Allocator is a policy that
allocates it.
Signed-off-by: Sage Weil <sage@redha... - 06:06 PM Revision d704628c (ceph): os/bluestore/BlockDevice: fix read return value
- Signed-off-by: Sage Weil <sage@redhat.com>
- 06:06 PM Revision a869f92f (ceph): os/bluestore: fix _do_read return value
- Signed-off-by: Sage Weil <sage@redhat.com>
- 06:06 PM Revision 65f720ae (ceph): kv/RocksDBStore: take custom Env
- Signed-off-by: Sage Weil <sage@redhat.com>
- 06:06 PM Revision 08a94d95 (ceph): os/bluestore/Allocator: init_rm_free
- Signed-off-by: Sage Weil <sage@redhat.com>
- 06:06 PM Revision 14460484 (ceph): os/bluestore/StupidAllocator: fix misc bugs
- Can't use invalid iterator; fix init_rm_free.
Signed-off-by: Sage Weil <sage@redhat.com> - 06:06 PM Revision 3a4d583f (ceph): os/newstore: always create db.wal
- Signed-off-by: Sage Weil <sage@redhat.com>
- 06:06 PM Revision 5658665c (ceph): os/newstore: consume a raw block device
- Signed-off-by: Sage Weil <sage@redhat.com>
- 06:06 PM Revision ad9f9fad (ceph): os/newstore: create db dir
- Signed-off-by: Sage Weil <sage@redhat.com>
- 06:05 PM Revision 84646ab1 (ceph): os/newstore: do not set/change frag_size if there are overlays
- Signed-off-by: Sage Weil <sage@redhat.com>
- 06:05 PM Revision 5e566dd7 (ceph): os/newstore: fix collection_list vs max entries
- Signed-off-by: Sage Weil <sage@redhat.com>
- 06:05 PM Revision 2dae3df8 (ceph): os/newstore: make key names more efficient
- - pack u32 and u64 in binary (instead of in hex)
- avoid duplicating the object name while making things still
sort... - 06:05 PM Revision 32e76839 (ceph): os/newstore: make collection_list tolerate sloppy start position
- Because of this change (#6076), the hobject_t will contain pool id, hence
the ghobject_t having this hobject_t will b... - 06:05 PM Revision be0528f4 (ceph): os/newstore: recycle rocksdb log files
- Signed-off-by: Sage Weil <sage@redhat.com>
- 06:05 PM Revision 29935046 (ceph): os/newstore: fixed fragment size
- Instead of a single, variable-length fragment for each object,
set a fixed size (newstore_min_frag_size = 1 MB) and s... - 06:05 PM Revision f0f815fb (ceph): os/newstore: dump onode contents
- Signed-off-by: Sage Weil <sage@redhat.com>
- 06:05 PM Revision 0af0dbdc (ceph): os/newstore: set alloc hint on new frags
- Signed-off-by: Sage Weil <sage@redhat.com>
- 06:05 PM Revision b2db842e (ceph): os/newstoer: add newstore types to ceph-dencoder
- Signed-off-by: Sage Weil <sage@redhat.com>
- 06:05 PM Revision 9291e168 (ceph): os/newstore: define a fid_backpointer_t type
- Signed-off-by: Sage Weil <sage@redhat.com>
fix wal_oP_t - 06:05 PM Revision feb2d3f6 (ceph): rocksdb: latest master
- Signed-off-by: Sage Weil <sage@redhat.com>
- 02:49 PM Revision 2694e117 (ceph): Merge pull request #6649 from majianpeng/filesstore-lfnunlink
- osd: FileStore:: optimize lfn_unlink
Reviewed-by: Kefu Chai <kchai@redhat.com> - 02:48 PM Revision 68a7c04b (ceph): Merge pull request #7017 from efirs/ef_atomic_ceph_tid
- osd: use atomic to generate ceph_tid
Reviewed-by: Sage Weil <sage@redhat.com>
Reviewed-by: Kefu Chai <kchai@redhat.com> - 02:47 PM Revision c6f88ecd (ceph): Merge pull request #7077 from XinzeChi/wip-fix-wip-perf
- osd: fix wip (l_osd_op_wip) perf counter and remove repop_map
Reviewed-by: Kefu Chai <kchai@redhat.com> - 02:37 PM Bug #14212 (Fix Under Review): osd/OSD.cc: 2469: FAILED assert(pg_stat_queue.empty()) on shutdown
- https://github.com/ceph/ceph/pull/7091
- 02:31 PM Bug #14212 (Resolved): osd/OSD.cc: 2469: FAILED assert(pg_stat_queue.empty()) on shutdown
- ...
- 02:35 PM Revision 04e41a60 (ceph): rados: 5m radosbench (not 10m)
- Signed-off-by: Sage Weil <sage@redhat.com>
- 02:32 PM Revision c25ff992 (ceph): osd: clear pg_stat_queue after stopping pgs
- Fixes: #14212
Signed-off-by: Sage Weil <sage@redhat.com> - 09:55 AM Bug #14209 (Resolved): fs: fix unintialized buffer during load fsid
- 01:34 AM Bug #14209: fs: fix unintialized buffer during load fsid
- Resolved by sage in wip-bluestore.
https://github.com/ceph/ceph/blob/wip-bluestore/src/os/kstore/KStore.cc
He added... - 09:54 AM Bug #14210 (Resolved): fs: fix unclosed formatter session
- 01:29 AM Bug #14210: fs: fix unclosed formatter session
- Resolved.
Sage applied to wip-bluestore mananually.
https://github.com/ceph/ceph/commit/4c7d09f375b917f3420adf6ed0b...
12/31/2015
- 02:33 PM Revision 44d8d638 (ceph): Merge pull request #6893 from kylinstorage/wip-osd_command
- librados: add c++ style osd/pg command interface
Reviewed-by: Josh Durgin <jdurgin@redhat.com> - 02:33 PM Revision cdc195a1 (ceph): Merge pull request #5630 from wonzhq/evict-after-flush
- osd: try evicting after flushing is done
Reviewed-by: Sage Weil <sage@redhat.com> - 02:33 PM Revision da581106 (ceph): Merge pull request #6639 from xiexingguo/xxg-wip-13822
- librados: potential null pointer access in list_(n)objects
Reviewed-by: Kefu Chai <kchai@redhat.com> - 02:32 PM Revision 5dd1cb0f (ceph): Merge pull request #6702 from liewegas/wip-fix-recency
- osd/ReplicatedPG: fix promotion recency logic
Reviewed-by: Samuel Just <sjust@redhat.com> - 02:32 PM Revision a67f873f (ceph): Merge pull request #6824 from Sandy4999/wip-crushtool-build
- crushtool: set type 0 name "device" for --build option
Reviewed-by: Sage Weil <sage@redhat.com> - 02:30 PM Revision d5b97676 (ceph): Merge pull request #6962 from liewegas/wip-buffer-lastp
- buffer: fix internal iterator invalidation on rebuild, get_contiguous
Reviewed-by: Samuel Just <sjust@redhat.com> - 02:30 PM Revision b5a2a76a (ceph): Merge pull request #6970 from aiicore/drop_removal_pg_type
- osd: drop deprecated removal pg type
Reviewed-by: Sage Weil <sage@redhat.com> - 09:33 AM Bug #14210: fs: fix unclosed formatter session
- https://github.com/ceph/ceph/pull/7089
- 08:50 AM Bug #14210 (Resolved): fs: fix unclosed formatter session
- See void onode_t::dump(Formatter *f) const
- 09:12 AM CephFS Bug #13977: ovh: sambatorture.scan-pipe_number fails on ENOMEM
- http://teuthology.ovh.sepia.ceph.com/teuthology/teuthology-2015-12-30_23:14:02-samba-master---basic-openstack/56621/t...
- 09:08 AM CephFS Bug #14211 (Resolved): test_version_splitting fails (tasks.cephfs.test_sessionmap.TestSessionMap)
- http://pulpito.ovh.sepia.ceph.com:8081/teuthology-2015-12-29_23:04:01-fs-infernalis---basic-openstack/55426/
- 08:07 AM Feature #6507: librados shouldn't block indefinitely when cluster doesn't respond
- I grep "client_mount_timeout" in source tree:
ack client_mount_timeout src/
src/client/Client.cc
4899: int r ... - 07:14 AM Bug #14209: fs: fix unintialized buffer during load fsid
- https://github.com/xiexingguo/ceph/commit/ffa6b467854e3152485f6b05f7c99f9518db6a4e
- 05:56 AM Bug #14209 (Resolved): fs: fix unintialized buffer during load fsid
- int NewStore::_read_fsid(uuid_d *uuid)
- 05:55 AM Bug #14188: fs: read_fsid may return undetermined result.
- https://github.com/xiexingguo/ceph/commit/bfaee93373938a88cfc2076690fcc850ff80e16b
- 01:00 AM Revision c01a314d (ceph): mon/PGMap: show rd/wr iops separately in status reports
- When check the iops, we always do not focuns on the total iops, but read iops and write iops. So the iops should be s...
12/30/2015
- 11:27 PM rgw Bug #14208 (Closed): rgw: 500 returned on Put Bucket ACL (ECANCELLED/-125)
- Put Bucket ACL is failing, returning ECANCELLED (-125)
<?xml version="1.0" encoding="UTF-8"?><Error><Code>UnknownE... - 11:05 PM Bug #14207 (Resolved): BlueStore compiling error without libaio support
- os/bluestore/BlockDevice.cc:413:2: warning: #warning fix discard (aio?) [-Wcpp]
#warning fix discard (aio?)
^
... - 10:50 PM Bug #14206 (Rejected): BlueStore Crashed
- -41> 2015-12-30 14:48:50.952056 7f0b3d7fa700 5 bdev(/mnt/nvmedevice/data/block) aio_write 2319581184~65536 aio 0x...
- 07:32 PM Revision 7404f857 (ceph): Remove unused SnapContext member snapc from MOSDSubOp message.
- Signed-off-by: Emile Snyder <emsnyder@ebay.com>
- 04:38 PM Feature #11692 (Fix Under Review): online change of mon_osd_full_ratio and mon_osd_nearfull_ratio...
- https://github.com/ceph/ceph/pull/7085
- 04:35 PM Revision 6a59aae0 (ceph): config: complains when a setting is not tracked
- * not all config items are tracked, so it does not take any effect after
we sucessfully changed them using "ceph te... - 03:19 PM Bug #14204: NewStore: potential null pointer assignment
- https://github.com/ceph/ceph/pull/7084
- 02:34 PM Bug #14204 (Resolved): NewStore: potential null pointer assignment
- See int NewStore::_truncate
OnodeRef o = c->get_onode(oid, false);
if (!o->exists)
{
r = -ENOENT... - 02:55 PM Bug #14205 (Can't reproduce): bad_alloc in DBObjectMap::set_keys
- ...
- 02:43 PM Revision 6110421b (ceph): rados: run radosbench for 10m (not 15m)
- Signed-off-by: Sage Weil <sage@redhat.com>
- 08:31 AM Bug #14203 (Closed): Expansion image,the client is unable to perceive
- 1)pool:LDTEST2,image:aa3,before the expansion,the image size is 200G
2)client: rbdmap and format the image,mount to ... - 06:52 AM Cleanup #14202 (In Progress): the method 'Message::decode_message' is too long
- hi, the method 'Message::decode_message' contains hundreds of switch-case statements, it's horribe.
i have found a ... - 06:41 AM Linux kernel client Feature #14201 (Resolved): cope with a pool being removed from under a mapped image
- 1. [root@c8 /]# uname -r
4.1.1
2. [root@ceph0 ~]# ceph -v
ceph version 0.94.5 (9764da52395923e0b32908d83a9f7... - 06:19 AM rbd Backport #14142 (In Progress): infernalis: Verify self-managed snapshot functionality on image cr...
- 06:16 AM Revision 83bba36a (ceph): tests: new integration test for validating new RBD pools
- Signed-off-by: Jason Dillaman <dillaman@redhat.com>
(cherry picked from commit 00cfe4efacd664032f700afe9701d41bacf8700a) - 06:16 AM Revision 96cb5e28 (ceph): librbd: optionally validate RBD pool configuration (snapshot support)
- Fixes: #13633
Signed-off-by: Jason Dillaman <dillaman@redhat.com>
(cherry picked from commit 1fea4dadc60e13518e9ee55d... - 06:13 AM rbd Backport #14199 (In Progress): infernalis: [ FAILED ] TestLibRBD.SnapRemoveViaLockOwner
- 05:54 AM rbd Backport #14199 (Resolved): infernalis: [ FAILED ] TestLibRBD.SnapRemoveViaLockOwner
- https://github.com/ceph/ceph/pull/7079
- 06:11 AM Revision 7a44714b (ceph): librbd: properly handle replay of snap remove RPC message
- Fixes: #14164
Backport: infernalis
Signed-off-by: Jason Dillaman <dillaman@redhat.com>
(cherry picked from commit bc3... - 06:05 AM rgw Backport #14140 (In Progress): infernalis: rgw: leak from RGWWatcher::handle_error
- 06:04 AM Linux kernel client Bug #14200 (Closed): The mounted image can not expansion
- Wheather the ceph linux kernal client support the expansion?
My linux kernal is 3.10.
Such as below after i finis... - 01:40 AM Linux kernel client Feature #14198 (New): rbd: renaming a pool does not update the pool names in /dev/rbd or /sys/bus...
- pool:LDTEST2,image:DE
client rbd map LDTEST2/DE; mkfs.ext4 /dev/rbd2;mount /dev/rbd2 /home/ld2;cp -r /root/centos.i... - 01:37 AM Bug #14197: Rados bench 4K write stops and reports pipe fault
- Re-ran same test on Hammer 0.94.2, was able to fill empty OSD to full with 4K writes and didn't see any pipe fault.
...
12/29/2015
- 07:07 PM Bug #14197 (Rejected): Rados bench 4K write stops and reports pipe fault
- Installed Infernalis 9.2.0 on one x64 Ubuntu 14.04 machine, setup one monitor and one OSD(480GB SSD) on same machine;...
- 02:48 PM Revision 2fd3f437 (ceph): osd: remove repop_map in osd
- If I do not misread, repop_map is useless.
Signed-off-by: Xinze Chi <xinze@xsky.com> - 02:32 PM CephFS Bug #14196 (Resolved): test_object_deletion fails (tasks.cephfs.test_damage.TestDamage)
http://teuthology.ovh.sepia.ceph.com/teuthology/teuthology-2015-12-21_23:04:02-fs-master---basic-openstack/48365/
...- 02:18 PM CephFS Bug #14195 (Won't Fix): test_full_fclose fails (tasks.cephfs.test_full.TestClusterFull)
- http://teuthology.ovh.sepia.ceph.com/teuthology/teuthology-2015-12-21_23:04:02-fs-master---basic-openstack/48408/teut...
- 02:00 PM Revision e29f55e0 (ceph): osd: fix wip (l_osd_op_wip) perf counter
- The l_osd_op_wip is for osd, so it should be the sum of all pgs in osd
Signed-off-by: Xinze Chi <xinze@xsky.com> - 09:26 AM Bug #14194: os: wrong return code of _do_copy_range
- https://github.com/ceph/ceph/pull/7059
- 06:17 AM Bug #14194 (Resolved): os: wrong return code of _do_copy_range
- Check int FileStore::_do_copy_range
if (pipe(pipefd) < 0)
{
r = errno;
derr <... - 09:16 AM Revision 1e120c88 (ceph): osd: Avoid osd_recovery_sleep cause recovery-thread suicide.
- Signed-off-by: Jianpeng Ma <jianpeng.ma@intel.com>
- 07:57 AM CephFS Feature #14146 (Fix Under Review): MDS: expose state of boot/replay to admins
- https://github.com/ceph/ceph/pull/7068
- 05:41 AM Revision 6f3f9d36 (ceph): osd: add cache hint when pushing raw clone during recovery
- When head or snapdir is missing on the primary, we push raw clone to the
replicas. Set cache hint also in this case.
... - 05:06 AM Messengers Bug #14193 (Won't Fix): XIO:OSD down when write data at fuse-client use iozone
- we recently work at ceph 9.0.3, and try to use XIO msg in infiniband environment.Firstly, we deploy 8 osds in our clu...
- 03:24 AM Bug #14192 (Resolved): os:lfn fd leak
- there are some cases which the lfn_fd is left unclosed.
12/28/2015
- 05:07 PM Bug #13990: Hammer (0.94.3) OSD does not delete old OSD Maps in a timely fashion (maybe at all?)
- I would say this probably will not fix the issue we're seeing as osd_pg_epoch_persisted_max_stale defaults to 200 (we...
- 08:50 AM Bug #13990: Hammer (0.94.3) OSD does not delete old OSD Maps in a timely fashion (maybe at all?)
- hey Tom,
> but there is at least one more we're hitting
the issue i am addressing is that the @last_persisted_o... - 11:23 AM Bug #14189: os: used space overflow
- https://github.com/ceph/ceph/pull/7061
- 08:17 AM Bug #14189 (Rejected): os: used space overflow
- See MemStore::_truncate
used_bytes += (o->get_size() - old_size);
The above stat may overflow. - 11:16 AM rgw Bug #14191: rgw: radosgw server abort when user passed bad parameters to set quota
- patches have been pushed:
https://github.com/ceph/ceph/pull/7063 - 10:53 AM rgw Bug #14191 (Resolved): rgw: radosgw server abort when user passed bad parameters to set quota
- ...
- 11:16 AM rgw Bug #14190: rgw: radosgw server abort when user passed bad parameters to set metadata
- patches have been pushed:
https://github.com/ceph/ceph/pull/7063 - 10:50 AM rgw Bug #14190 (Resolved): rgw: radosgw server abort when user passed bad parameters to set metadata
- ...
- 11:00 AM Revision 59f90584 (ceph): rgw: do not abort when user passed bad parameters to set quota
- when user passes bad parameter(e.g. pass "null" to bool parameter), the ceph_json.cc::decode_json_obj()
will raise a ... - 10:58 AM Revision 83503239 (ceph): rgw: do not abort when user passed bad parameters to set metadata
- Fixes: #14190
when user passes bad parameter(e.g. pass "null" to bool parameter), the ceph_json.cc::decode_json_obj(... - 08:40 AM Revision 4dd0d1bd (ceph): Merge pull request #6987 from H3C/wip-addr-bugfix
- common/address_help.cc: fix the leak in entity_addr_from_url()
Reviewed-by: Kefu Chai <kchai@redhat.com> - 08:09 AM CephFS Bug #14144 (Fix Under Review): standy-replay MDS does not cleanup finished replay threads
- https://github.com/ceph/ceph/pull/7060
- 03:14 AM CephFS Bug #14048 (Resolved): MDS heap stats standby threads
12/27/2015
- 02:44 AM Bug #14188 (Resolved): fs: read_fsid may return undetermined result.
- Check int FileStore::read_fsid(int fd, uuid_d *uuid)
- 02:29 AM Bug #14187 (Resolved): fs: fix return code of mkfs
- int FileStore::mkfs()
if (::fsync(fsid_fd) < 0)
{
ret = errno;
derr << "mkfs...
12/26/2015
- 11:29 PM Revision 9b686c33 (ceph): MonitorDBStore, LevelDBStore, RocksDBStore: make sure the provided bl i...
- If it's not, data will be appended to the end of bufferlist and difficult
to spot bugs might arise, so fail hard in t... - 10:54 AM Revision 580588ab (ceph): OSD: simplify wake_pg_waiters a little
- the input parameter pg is not needed any more.
Signed-off-by: xie xingguo <xie.xingguo@zte.com.cn> - 04:34 AM Bug #13990: Hammer (0.94.3) OSD does not delete old OSD Maps in a timely fashion (maybe at all?)
- Thanks for the updates Kefu. Is there something you would like me to attempt to gather additional info?
We create... - 03:24 AM Bug #14186 (Resolved): OS: thread leak during mount
- See int FileStore::mount()
{
stringstream err2;
if (g_conf->filestore_debug_omap_check && !o...
12/25/2015
- 02:51 PM Revision ecd87844 (ceph): ReplicatedPG: remove redundant result code checking
- It is going to fall through to fail lable anyway.
Signed-off-by: xie xingguo <xie.xingguo@zte.com.cn> - 02:42 PM Bug #14185: ReplicatedPG: result code not handled during omap operation.
- Please close due to https://github.com/xiexingguo/ceph/commit/04e3810c895733ecae9f54c7c73cd8b958c95577.
- 12:46 PM Bug #14185 (Fix Under Review): ReplicatedPG: result code not handled during omap operation.
- 10:12 AM Bug #14185: ReplicatedPG: result code not handled during omap operation.
- https://github.com/ceph/ceph/pull/7055
- 09:51 AM Bug #14185 (Won't Fix): ReplicatedPG: result code not handled during omap operation.
- See ReplicatedPG::do_osd_ops:
case CEPH_OSD_OP_OMAPGETVALSBYKEYS:
if (pool.info.supports_omap())
... - 08:06 AM Bug #14182 (Fix Under Review): Optimization the session_handle_reset function.
- 06:52 AM Bug #14182: Optimization the session_handle_reset function.
- https://github.com/ceph/ceph/pull/7054
- 06:32 AM Bug #14182 (Resolved): Optimization the session_handle_reset function.
- Optimization the session_handle_reset, delete useless loop, to run faster.
- 08:05 AM Feature #14183 (Fix Under Review): Show the totall pg on each osd.
- 06:49 AM Feature #14183: Show the totall pg on each osd.
- https://github.com/ceph/ceph/pull/7050
- 06:42 AM Feature #14183 (Fix Under Review): Show the totall pg on each osd.
- ceph pg dump osds
dumped osds in format plain
osdstat kbused kbavail kb hb in hb out pg sum
0 143304204 79623420 2... - 08:05 AM Bug #14184 (Fix Under Review): Optimize the json_command api
- 06:47 AM Bug #14184: Optimize the json_command api
- https://github.com/ceph/ceph/pull/6915
- 06:45 AM Bug #14184 (Rejected): Optimize the json_command api
- In json_command api, the prefix like 'osd tree' is ok, but 'osd tree' or ' osd tree ' will fail.
- 08:02 AM Bug #14181 (Fix Under Review): OSD flapping when Public/Cluster network down
- 03:25 AM Bug #14181: OSD flapping when Public/Cluster network down
- Fix proposed in https://github.com/ceph/ceph/pull/7053
- 03:20 AM Bug #14181 (Resolved): OSD flapping when Public/Cluster network down
- This is reproduceable by plug out(not ifdown) cable or using iptables dropping all package.
Bug was introduced by ... - 06:25 AM Bug #14175: clock skew report is incorrect by "ceph health detail" command
- I was using 0.94.5 version
How long it take the cluster recover fron clock skew reported depend on how large time of... - 03:07 AM Revision dd8221dd (ceph): osd/OSD.cc Check health state before pre_booting
- In previous code we forgot to check the health state before
going to STATE_PREBOOT, which will result OSD flapping wh...
12/24/2015
- 05:37 PM Revision 9af4d568 (ceph): test: enable changing subscribe level testing
- This test doesn't pass previously because of a bug. After fixing it, we
can now enable this test.
Signed-off-by: Zhi... - 05:37 PM Revision 74856683 (ceph): mon: don't send Mlog msg to client when there is no entries
- Signed-off-by: Zhiqiang Wang <wonzhq@hotmail.com>
- 05:35 PM Revision 9ea99fab (ceph): mon: remove conflicting subscribes when handle_subscribe
- For monitoring log, when changing between different levels, should
remove previous level in the subscribes. This is b... - 03:03 PM rgw Bug #14180 (Resolved): rgw: segv in RGWWatcher::~RGWWatcher
- ...
- 02:57 PM Revision 07232b1e (ceph): rados/singleton/all/pg-removal-interruption: ensure osd.0 is marked down
- We were waiting for teh dameon to fail, not for the mons to notice,
which meant that after we restart the daemon the ... - 01:21 PM Revision 4efa2145 (ceph): kv: implement value_as_ptr() and use it in .get()
- Current .get() methods return values as bufferlists, which are replacing
bufferlists provided by caller. This is part... - 01:03 PM rbd Bug #14127 (Resolved): librbd: wrongly update counter for readahead.
- 01:01 PM Bug #14175: clock skew report is incorrect by "ceph health detail" command
- Hopefully Joao will chime in with a deeper explanation, but until then I can say that I have run into a similar issue...
- 05:59 AM Bug #14175 (Resolved): clock skew report is incorrect by "ceph health detail" command
- I usr "ceph health detail" command to check my cluster health and found below warning:
mon.c7 addr 10.118.202.97:6... - 12:37 PM Revision 41fcd877 (ceph): doc: s/InvalidCap/InvalidCapability
- InvalidCapability is more readable
Signed-off-by: Dunrong Huang <riegamaths@gmail.com> - 12:29 PM Revision f302e058 (ceph): osd: avoid FORCE updating digest been overwritten by MAYBE
- Should set update to FORCE after MAYBE, to avoid it been overwritten.
Signed-off-by: Zhiqiang Wang <wonzhq@hotmail.com> - 12:26 PM Revision 8ea80a0e (ceph): rgw: add a new error code for invalid admin capability
- From http://docs.ceph.com/docs/master/radosgw/adminops/ when attempt to
grant invalid admin capability, rgw should re... - 10:10 AM Bug #14178 (Resolved): KeyValueStore: fd leak during mount
- See KeyValueStore::mount():
if (!store)
{
derr << "KeyValueStore::mount backend type "
... - 10:01 AM Revision 4275ee66 (ceph): rbd-fuse:image name can not include snap name
- It is legal to create a image name like "foo@s1" by the rbd-fuse method.However the "foo@s1" type image name is not a...
- 09:40 AM Bug #14177: osd:osd tree display in/out
- https://github.com/ceph/ceph/pull/7049
- 08:04 AM Bug #14177 (Closed): osd:osd tree display in/out
- In the use of ceph osd tree osd status display, the osd in/out status display for 1.00000 or 0
Example:... - 08:27 AM Revision 749c4242 (ceph): Merge pull request #6933 from jcsp/wip-14048
- mds: tear down connections from `tell` commands
- 08:00 AM Revision 9eefc430 (ceph): tests:delete the default order of RBD in test_rbd
- delete rbd'order = RBD_DEFAULT_OBJ_ORDER process, when rbd'order is set to zero
so,delete the default order of RBD in... - 07:54 AM Revision 7817a815 (ceph): ceph: wait unsafe requests in sync_fs()
- Signed-off-by: Yan, Zheng <zyan@redhat.com>
- 07:54 AM Revision 636a9841 (ceph): client: make sync_fs() flush dirty data in objectcacher
- Signed-off-by: Yan, Zheng <zyan@redhat.com>
- 07:54 AM Revision bd86c301 (ceph): objectcacher: introduce ObjectCacher::flush_all()
- Signed-off-by: Yan, Zheng <zyan@redhat.com>
- 07:49 AM Revision 8d2df89a (ceph): client: wait unsafe requests that created/modified the file in fsync()
- Signed-off-by: Yan, Zheng <zyan@redhat.com>
- 07:49 AM Revision a7501c9a (ceph): client: fix 'wait unsafe requests' code in _fsync()
- Unsafe requests in Inode::unsafe_ops are not sorted in tid order.
So checking first unsafe request's tid does not mak... - 07:49 AM Revision 428a1f31 (ceph): objectcacher: sort dirty_or_tx_bh in ObjectSet/Object/offset order
- so we can find dirty buffer heads in given a ObjectSet easily.
Signed-off-by: Yan, Zheng <zyan@redhat.com> - 07:49 AM Revision 5246e789 (ceph): objectcacher: make flush_set() only flushes dirty data in ObjectSet
- Buffer heads in dirty_or_tx_bh are sorted in ObjectSet/Object/offset
order. So we can find dirty buffer heads in the ... - 06:36 AM Bug #14176 (Resolved): KeyValueStore: wrongly placed dump_open_fds() logic
- Check KeyValueStore::_do_transaction:
assert(0 == "unexpected error");
if (r == -EMFILE) {
d... - 04:32 AM Revision c615feb7 (ceph): rgw: let radosgw-admin bucket stats return a standard josn
- Signed-off-by: Ruifeng Yang <yangruifeng.09209@h3c.com>
- 01:18 AM Bug #13381: osd/SnapMapper.cc: 282: FAILED assert(check(oid)) on hammer->jewel upgrade
This object has a hash and is in pool 0. Doesn't seem related to 13862. Don't know why check() failed.
1000a8962...- 01:02 AM devops Bug #14174 (Rejected): autobuild-ceph: gitbuilder_ceph_rpm mismanages 'dist'
- gitbuilder_ceph_rpm unconditionally does "sudo('echo centos6 > dists')". We build rpms for other dists now. However...
- 12:59 AM Revision 5864d565 (ceph): rbd: only test on xfs
- btrfs is not recommended for production
Signed-off-by: Josh Durgin <jdurgin@redhat.com> - 12:59 AM Revision 8bd0b9da (ceph): rbd: add some workloads with journalling enabled
- Separating features into its own facet would be good,
but requires changing some of the tests a bit.
Signed-off-by: ... - 12:59 AM Revision 888ff87d (ceph): rbd: add a couple new workunits
- for rbd-nbd and the journalling cli
Signed-off-by: Josh Durgin <jdurgin@redhat.com> - 12:59 AM Revision ef25ba39 (ceph): rbd: add helgrind to valgrind subsuite
- This checks locking problems. This can be a separate facet for
workunits, but for now it requires a manual 2nd config... - 12:59 AM Revision c7cd4f74 (ceph): rbd: increase length of non-valgrind fsx workloads
- New lab machines make these run much faster. These should take ~1-1.5 hours now.
Signed-off-by: Josh Durgin <jdurgin... - 12:32 AM Revision c485d29a (ceph): Merge remote-tracking branch 'origin/jewel'
12/23/2015
- 11:34 PM Revision 10b4a082 (ceph): osd/OSD: clear_temp_objects() include removal of Hammer temp objects
- Fixes: #13862
Signed-off-by: David Zafman <dzafman@redhat.com> - 11:34 PM Revision e85907fc (ceph): osd: Improve log message which isn't about a particular shard
- Remove redundant dout()
Signed-off-by: David Zafman <dzafman@redhat.com> - 11:25 PM rbd Bug #14127: librbd: wrongly update counter for readahead.
- @Nathan Cutler
Resolved - 10:30 PM Revision 508deb98 (ceph): common/address_help.cc: fix the leak in entity_addr_from_url()
- Fixes: #14132
Signed-off-by: Qiankun Zheng <zheng.qiankun@h3c.com> - 10:28 PM Revision f8c4d04c (ceph): Merge pull request #7026 from xdonghai/master
- rbd: must specify both of stripe-unit and stripe-count when specifying stripingv2 feature
Reviewed-by: Josh Durgin <... - 10:20 PM Bug #13381: osd/SnapMapper.cc: 282: FAILED assert(check(oid)) on hammer->jewel upgrade
- 2015-12-22 19:27:53.913613 7f64a689d700 20 snap_mapper.remove_oid -1/00000000/temp_4.40s0_0_34564_1/head
could thi... - 10:15 PM Revision ea131fe8 (ceph): Merge pull request #6998 from xiexingguo/xxg-wip-clsrbd
- stringify outputted error code and fix unmatched parentheses.
Reviewed-by: Josh Durgin <jdurgin@redhat.com> - 10:12 PM Revision 243146c6 (ceph): Merge pull request #6983 from xiexingguo/xxg-wip-14126
- librbd: small fixes for error messages and readahead counter
Reviewed-by: Josh Durgin <jdurgin@redhat.com> - 10:07 PM Revision 7f446f1a (ceph): Merge remote-tracking branch 'gh/jewel'
- 10:06 PM Revision 179ff642 (ceph): Merge pull request #755 from guangyy/wip-13441-test
- pg: add test cases for pg list_missing on EC pool
- 10:06 PM Feature #13441 (Resolved): change PG::do_command list_missing to iterate over missing_loc structu...
- 10:06 PM Revision 6953166a (ceph): Merge pull request #777 from ceph/wip-radosbench
- rados: run radosbench for 15m instead of 30m
Reviewed-by: Josh Durgin <jdurgin@redhat.com> - 10:05 PM Revision 459f2e8d (ceph): Merge pull request #6298 from guangyy/wip-13441
- osd: make list_missing query missing_loc.needs_recovery_map
Reviewed-by: Samuel Just <sjust@redhat.com> - 10:05 PM Revision 44a3dbb9 (ceph): Merge pull request #6572 from liewegas/wip-crush-chooseleaf-stable
- crush: add chooseleaf_stable tunable
Reviewed-by: Loic Dachary <ldachary@redhat.com> - 10:00 PM rgw Bug #14173 (Resolved): rgw: segv in RGWWatcher::handle_error
- ...
- 10:00 PM Revision e2f9cedd (ceph): rados: run radosbench for 15m instead of 30m
- The smithi disks fill after about 20m.
Signed-off-by: Sage Weil <sage@redhat.com> - 09:30 PM Revision a2976081 (ceph): Merge pull request #7043 from dillaman/wip-14170-jewel
- librbd: do not ignore self-managed snapshot release result
Reviewed-by: Josh Durgin <jdurgin@redhat.com> - 07:01 PM Bug #11969: chain_xattr: deal with 254 byte limit on inline xfs attrs
- Fix will ship in v0.94.6
- 07:00 PM rbd Bug #14164 (Pending Backport): [ FAILED ] TestLibRBD.SnapRemoveViaLockOwner
- 06:29 PM rbd Bug #14164 (Fix Under Review): [ FAILED ] TestLibRBD.SnapRemoveViaLockOwner
- *jewel PR*: https://github.com/ceph/ceph/pull/7042
- 05:22 PM rbd Bug #14164 (Resolved): [ FAILED ] TestLibRBD.SnapRemoveViaLockOwner
- http://qa-proxy.ceph.com/teuthology/joshd-2015-12-22_16:52:43-rbd-master---basic-smithi/2531/teuthology.log...
- 07:00 PM rbd Bug #14165 (Resolved): Failure in merge_diff.sh
- 05:57 PM rbd Bug #14165 (Fix Under Review): Failure in merge_diff.sh
- *master PR*: https://github.com/ceph/ceph/pull/7041
- 05:24 PM rbd Bug #14165 (Resolved): Failure in merge_diff.sh
- http://qa-proxy.ceph.com/teuthology/joshd-2015-12-22_16:52:43-rbd-master---basic-smithi/2428/teuthology.log...
- 06:59 PM rbd Bug #14170 (Fix Under Review): Failure in "rbd/singleton/{all/rbd-vs-unmanaged-snaps.yaml}"
- *jewel PR*: https://github.com/ceph/ceph/pull/7043
- 06:57 PM rbd Bug #14170 (In Progress): Failure in "rbd/singleton/{all/rbd-vs-unmanaged-snaps.yaml}"
- 05:30 PM rbd Bug #14170 (Resolved): Failure in "rbd/singleton/{all/rbd-vs-unmanaged-snaps.yaml}"
- http://qa-proxy.ceph.com/teuthology/joshd-2015-12-22_16:52:43-rbd-master---basic-smithi/2508/teuthology.log
- 06:57 PM Revision b4dbaa69 (ceph): librbd: do not ignore self-managed snapshot release result
- Fixes: #14170
Signed-off-by: Jason Dillaman <dillaman@redhat.com> - 06:52 PM Revision 1e1df754 (ceph): Merge pull request #7042 from dillaman/wip-14164-jewel
- librbd: properly handle replay of snap remove RPC message
Reviewed-by: Josh Durgin <jdurgin@redhat.com> - 06:51 PM Revision 5657f5bf (ceph): Merge pull request #7041 from dillaman/wip-14165
- qa/workunits: merge_diff shouldn't attempt to use striping
Reviewed-by: Josh Durgin <jdurgin@redhat.com> - 06:50 PM rbd Bug #14092 (Resolved): run-rbd-unit-tests.sh: various lock related asserts
- 05:14 PM rbd Bug #14092 (Fix Under Review): run-rbd-unit-tests.sh: various lock related asserts
- *jewel PR*: https://github.com/ceph/ceph/pull/7040
- 11:09 AM rbd Bug #14092 (In Progress): run-rbd-unit-tests.sh: various lock related asserts
- https://jenkins.ceph.com/job/ceph-pull-requests/42/console...
- 06:48 PM Revision c0c608b2 (ceph): Merge pull request #7040 from dillaman/wip-14092-jewel
- librbd: ensure librados callbacks are flushed prior to destroying
Reviewed-by: Josh Durgin <jdurgin@redhat.com> - 06:48 PM rbd Bug #14121 (Resolved): "[ FAILED ] TestLibRBD.RebuildObjectMap" in upgrade:client-upgrade-jewel...
- 03:34 PM rbd Bug #14121 (Fix Under Review): "[ FAILED ] TestLibRBD.RebuildObjectMap" in upgrade:client-upgra...
- *infernalis PR*: https://github.com/ceph/ceph/pull/7038
- 02:51 PM rbd Bug #14121 (In Progress): "[ FAILED ] TestLibRBD.RebuildObjectMap" in upgrade:client-upgrade-je...
- 06:47 PM Revision 64fab4f0 (ceph): Merge pull request #7038 from dillaman/wip-14121-infernalis
- tests: rebuild exclusive lock test should acquire exclusive lock
Reviewed-by: Josh Durgin <jdurgin@redhat.com> - 06:47 PM Revision 14933e7a (ceph): Merge pull request #7036 from xiexingguo/xxg-wip-kvstore
- KeyValueStore: fix return code of mkfs
Reviewed-by: David Zafman <dzafman@redhat.com> - 06:45 PM rbd Bug #14122 (Resolved): "import failed: (95) Operation not supported" in upgrade:client-upgrade-je...
- 02:49 PM rbd Bug #14122 (Fix Under Review): "import failed: (95) Operation not supported" in upgrade:client-up...
- *jewel PR*: https://github.com/ceph/ceph/pull/7035
- 01:45 PM rbd Bug #14122 (In Progress): "import failed: (95) Operation not supported" in upgrade:client-upgrade...
- 06:45 PM Revision 9bb9b854 (ceph): Merge pull request #7020 from chenyehua11692/master
- doc:adding "--allow-shrink" in decreasing the size of the rbd block to distinguish from the increasing option
Review... - 06:43 PM Revision f0767a74 (ceph): Merge pull request #7035 from dillaman/wip-14122-jewel
- librbd: clear error when older OSD doesn't support image flags
Reviewed-by: Josh Durgin <jdurgin@redhat.com> - 06:40 PM rbd Bug #14167 (In Progress): ceph_test_librbd_fsx 'read bad data' failure
- 05:27 PM rbd Bug #14167 (Resolved): ceph_test_librbd_fsx 'read bad data' failure
- http://qa-proxy.ceph.com/teuthology/joshd-2015-12-22_16:52:43-rbd-master---basic-smithi/2471/teuthology.log
http://q... - 06:32 PM Stable releases Tasks #13838 (Rejected): firefly v0.80.12
- firefly is EOL as of january 1st, 2016
- 06:29 PM rbd Bug #14169 (In Progress): ceph_test_librbd_fsx lock assertion failure
- 05:28 PM rbd Bug #14169 (Duplicate): ceph_test_librbd_fsx lock assertion failure
- http://qa-proxy.ceph.com/teuthology/joshd-2015-12-22_16:52:43-rbd-master---basic-smithi/2547/teuthology.log
- 06:26 PM Revision bc309d9d (ceph): librbd: properly handle replay of snap remove RPC message
- Fixes: #14164
Backport: infernalis
Signed-off-by: Jason Dillaman <dillaman@redhat.com> - 06:13 PM devops Bug #14172: https://jenkins.ceph.com/job/ceph-pull-requests/ aka make check fixes
- Setting to urgent because the absence of an automated make check has a noticeable daily impact on the work of Ceph de...
- 06:12 PM devops Bug #14172 (Resolved): https://jenkins.ceph.com/job/ceph-pull-requests/ aka make check fixes
- h3. Workaround
*run-make-check.sh ; sudo reboot* to cleanup lingering processes. This should allow the bot to run ... - 05:54 PM Revision 8e095ded (ceph): qa/workunits: merge_diff shouldn't attempt to use striping v2
- Fixes: #14165
Signed-off-by: Jason Dillaman <dillaman@redhat.com> - 05:31 PM rbd Bug #14171 (Resolved): rbd-nbd teuthology tests need to install rbd-nbd package
- http://qa-proxy.ceph.com/teuthology/joshd-2015-12-22_16:52:43-rbd-master---basic-smithi/2494/teuthology.log
http://q... - 05:28 PM rbd Bug #14168 (Resolved): ceph_test_librbd_fsx crashed
- http://qa-proxy.ceph.com/teuthology/joshd-2015-12-22_16:52:43-rbd-master---basic-smithi/2489/teuthology.log
- 05:25 PM rbd Bug #14166 (Can't reproduce): Failure in rbd/test_lock_fence.sh
- http://qa-proxy.ceph.com/teuthology/joshd-2015-12-22_16:52:43-rbd-master---basic-smithi/2437/teuthology.log
http://q... - 05:21 PM rbd Bug #14163 (Resolved): Helgrind errors reported during teuthology runs
- http://qa-proxy.ceph.com/teuthology/joshd-2015-12-22_16:52:43-rbd-master---basic-smithi/2409/teuthology.log
http://q... - 05:15 PM Revision bb296f83 (ceph): Suites/jewel in the smithi lab will be on centos 7.1
- Signed-off-by: yuriw <yuri.weinstein@gmail.com>
(cherry picked from commit d4d3787609f351b7efa0d52120858d090d1cf80e) - 05:06 PM Revision 98157ab3 (ceph): librbd: ensure librados callbacks are flushed prior to destroying image
- Fixes: #14092
Signed-off-by: Jason Dillaman <dillaman@redhat.com> - 03:32 PM rgw Bug #14090 (Fix Under Review): rgw: the response XML format of admin policy API is error
- 03:32 PM rgw Bug #14090 (Pending Backport): rgw: the response XML format of admin policy API is error
- 03:31 PM Revision 7676343c (ceph): tests: rebuild exclusive lock test should acquire exclusive lock
- Starting with Jewel, the object map will not be loaded until the
exclusive lock is acquired since it might be updated... - 03:23 PM Feature #13505: scrub/repair: persist scrub results.
- should not return the scrub result if the scrub is still in progress. we can
* check the status of pg before se... - 01:49 AM Feature #13505: scrub/repair: persist scrub results.
- Kefu Chai wrote:
>
> @dzafman, i found that @ReplicatedPG::_scrub()@ are repeating the check for missing/corrupted... - 03:17 PM rgw Bug #14162: rgw: return suitable error message for bad request when using Admin API
- pull request;
https://github.com/ceph/ceph/pull/7037 - 03:02 PM rgw Bug #14162 (Resolved): rgw: return suitable error message for bad request when using Admin API
- 1. use admin api to create two users, with same email address:
$ curl -s -X PUT http://ceph1:9090/admin/user?format=... - 02:49 PM Revision a069ea99 (ceph): KeyValueStore: fix return code of mkfs
- It shall return a negative result code instead.
Signed-off-by: xie xingguo <xie.xingguo@zte.com.cn> - 02:47 PM Revision b8d7aa2c (ceph): librbd: clear error when older OSD doesn't support image flags
- Fixes: #14122
Signed-off-by: Jason Dillaman <dillaman@redhat.com> - 02:39 PM Revision 65db4031 (ceph): rgw: return InvalidAccessKey for invalid access key specified
- Signed-off-by: Dunrong Huang <riegamaths@gmail.com>
- 02:36 PM Revision ad1e6489 (ceph): rgw: add a new error code for invalid key type
- From http://docs.ceph.com/docs/master/radosgw/adminops/ when invalid key type
specified, rgw should return InvalidKey... - 02:30 PM Revision df6d3239 (ceph): rgw: add a new error code for invalid secret key
- From http://docs.ceph.com/docs/master/radosgw/adminops/ when invalid secret
key specified, rgw should return 400 with... - 02:29 PM Revision e5d32b0f (ceph): rgw: add a new error code for key conflict
- From http://docs.ceph.com/docs/master/radosgw/adminops/ when provided key
exists, rgw should return KeyExists rather ... - 02:29 PM Revision c2273169 (ceph): rgw: add a new error code for email conflict
- From http://docs.ceph.com/docs/master/radosgw/adminops/ when provided email
address exists, rgw should return EmailEx... - 01:49 PM Revision 0c17b613 (ceph): Merge pull request #6996 from H3C/wip-mds-f11322
- mds: we should wait messenger when MDSDaemon suicide
Reviewed-by: John Spray <john.spray@redhat.com> - 01:39 PM Revision e0833f2c (ceph): Merge pull request #7015 from chipitsine/master
- ceph-fuse: fix double free of args
Reviewed-by: Kefu Chai <kchai@redhat.com> - 01:36 PM Revision d812e03b (ceph): Merge pull request #7004 from jazeltq/master
- doc: fix typo
Reviewed-by: Sage Weil <sage@redhat.com> - 01:28 PM Revision 8ca07f44 (ceph): Merge pull request #6870 from mslovy/wip-pgmeta-object
- osd: os: skip checking pg_meta object existance in FileStore
Reviewed-by: Sage Weil <sage@redhat.com> - 01:27 PM Revision 0b675268 (ceph): Merge pull request #6744 from majianpeng/scrub-fix
- osd: release related sources when scrub is interrupted
Reviewed-by: Kefu Chai <kchai@redhat.com> - 01:26 PM Revision 328a96f3 (ceph): Merge pull request #6961 from liewegas/wip-lockdep-max
- common/lockdep: increase max lock names
Reviewed-by: Kefu Chai <kchai@redhat.com> - 01:26 PM Bug #13949 (Need More Info): rados.py uses string_at, leaks string memory
- > It looks to me very much like ctypes.string_at() allocates new memory for the string; in the case of xattrs, at lea...
- 01:17 PM rbd Subtask #14161 (Resolved): Serialize snap create, snap rollback, and resize operations for journa...
- Since these three operations affect the data within an image, when journaling is enabled, these ops need to be serial...
- 12:48 PM Revision fff49412 (ceph): mds: we should wait messenger when MDSDaemon suicide
- Signed-off-by: Wei Feng <feng.wei@h3c.com>
- 12:47 PM rgw Bug #14160 (Triaged): rgw:add a new error message for access-key conflict when using Admin Ops
- ...
- 12:33 PM Feature #11692: online change of mon_osd_full_ratio and mon_osd_nearfull_ratio doesn't take effect.
- Right, that's a good idea. if no observers is tracking a config, changing it at runtime would lead to a warning.
- 12:18 PM Bug #13850: master fails to build when liblz4 is present (rocksdb)
- we'd better make sure our building env is a clean room, otherwise the configure script will detect the installed and ...
- 10:47 AM Bug #13850 (Fix Under Review): master fails to build when liblz4 is present (rocksdb)
- https://github.com/ceph/ceph/pull/7030
- 12:05 PM Bug #13843 (Pending Backport): ceph osd pool stats broken in hammer
- 12:04 PM Bug #13843: ceph osd pool stats broken in hammer
- https://github.com/ceph/ceph/pull/4350
- 10:59 AM Revision bea2e10a (ceph): Merge pull request #7025 from tchaikov/wip-ceph-detect-init-py3
- ceph-detect-init: fix py3 test
Reviewed-by: Loic Dachary <ldachary@redhat.com> - 10:47 AM Cleanup #13981: configure script doesn't check bzip2-devel presence
- https://github.com/ceph/ceph/pull/7030 also addresses this issue
- 09:40 AM Revision 0dc0d425 (ceph): Revert "Revert "makefiles: remove bz2-dev from dependencies""
- This reverts commit 35542eb6ed331ddedbc1406b5d49342c675af0f8.
- 09:39 AM Revision 911e7a02 (ceph): configure: link against bz2 and lz4 if detected
- if we configure --with-librocksdb-static, the rocksdb's Makefile
will detect the installed libbz2 and libz4 by its ow... - 08:48 AM Revision 6a560b51 (ceph): async: don't use shared_ptr to manage EventCallback
- Signed-off-by: Haomai Wang <haomai@xsky.com>
- 08:35 AM Revision 80c770e1 (ceph): ceph-detect-init: fix py3 test
- * in python3, None can not be compared with a str. and
we should always set codename with a non empty str, so updat... - 08:12 AM Revision 03f93077 (ceph): Merge pull request #7023 from tchaikov/wip-doc-cachemode
- doc: document "readforward" and "readproxy" cache mode
Reviewed-by: Loic Dachary <ldachary@redhat.com> - 07:59 AM Revision 54b12f53 (ceph): rbd:must specify both of stripe-unit and stripe-count when specify stri...
- when create a rbd image with striping features, Be sure specify both of stirpe-unit and stripe-count too, If not, ref...
- 07:57 AM Revision 29f9d10e (ceph): doc: add missing semicolon
- Signed-off-by: Dunrong Huang <riegamaths@gmail.com>
- 07:35 AM Revision be55a275 (ceph): doc: s/Bucket/BucketName
- from amazon s3 documentaion, the name shoule be BucketName rather than Bucket.
Signed-off-by: Dunrong Huang <riegama... - 07:24 AM Revision dd065d89 (ceph): doc: update SetACL API
- Since v2 and later version, SetACLRequest() and SetACL() were removed,
and replaced by PutACLRequest() and PutACL()
... - 07:15 AM Revision d46297e7 (ceph): doc: s/ListBucketResponse/ListBucketsResponse
- From following documentation:
- http://docs.aws.amazon.com/sdkfornet/v3/apidocs/items/S3/MS3ListBucketsResponsector... - 05:43 AM Revision d821acad (ceph): doc: document "readforward" and "readproxy" cache mode
- Signed-off-by: Kefu Chai <kchai@redhat.com>
- 05:23 AM Linux kernel client Bug #14086: Ceph File System logging warning about ceph_set_page_dirty
- Test has run for 2 days using the 4.4rc6 kernel from the Ubuntu kernel-ppa kernel site without error or warning. Loo...
- 05:13 AM Revision 777564c8 (ceph): doc:adding "--allow-shrink" in decreasing the size of the rbd block to ...
- In the original file, the increasing and decreaing of the size of the rbd block shares the same option:
"rbd resi... - 04:09 AM Revision fd6fca7f (ceph): doc: `Amazon.AWSClientFactory' is obsolete
- From http://blogs.aws.amazon.com/net/post/Tx2JT7AYCL2F6CB/ , Amazon.AWSClientFactory was removed
Signed-off-by: Dunr... - 03:23 AM Revision c2e39130 (ceph): man: document listwatchers cmd in "rados" manpage
- Signed-off-by: Kefu Chai <kchai@redhat.com>
- 02:56 AM Revision 877f332f (ceph): ceph doc fix slip of pen
- mon client hung interval -> mon client hunt interval
Signed-off-by: tianqing <tianqing@unitedstack.com> - 02:24 AM Bug #14134 (Resolved): FileJournal: create method shall return negative error code.
- 12:43 AM Bug #14134: FileJournal: create method shall return negative error code.
- @Nathan Cutler
Resolved.
12/22/2015
- 10:47 PM Bug #14158 (Can't reproduce): Segmentation fault in upgrade:hammer-x-infernalis-distro-basic-open...
- Run: http://pulpito.ovh.sepia.ceph.com:8081/teuthology-2015-12-22_17:02:01-upgrade:hammer-x-infernalis-distro-basic-o...
- 10:44 PM Bug #13381: osd/SnapMapper.cc: 282: FAILED assert(check(oid)) on hammer->jewel upgrade
- Also on hammer -> infernalis
Run: http://pulpito.ovh.sepia.ceph.com:8081/teuthology-2015-12-22_17:02:01-upgrade:hamm... - 10:02 PM Revision 0af73055 (ceph): Merge remote-tracking branch 'gh/jewel'
- 09:55 PM Revision 94aaecbd (ceph): Merge pull request #774 from ceph/wip-radostool
- suites/rados/singleton/all/radostool: whitelist quota error
Reviewed-by: Josh Durgin <jdurgin@redhat.com> - 09:49 PM Revision 2b4e9497 (ceph): suites/rados/singleton/all/radostool: whitelist quota error
- The test sets and hits a quota; whitelist the message
Signed-off-by: Sage Weil <sage@redhat.com> - 09:06 PM rbd Feature #14006 (Fix Under Review): Track IO dependencies in journal
- *master PR*: https://github.com/ceph/ceph/pull/7019
- 08:28 PM Revision eb8057c6 (ceph): ceph-fuse: fix double free of args
- Checking src/ceph_fuse.cc...
[src/ceph_fuse.cc:55]: (error) Memory pointed to by 'argv' is freed twice.
[src/ceph_fus... - 06:02 PM Revision f14fe56d (ceph): Merge pull request #6988 from xiexingguo/xxg-wip-14134
- FileJournal: fix return code of create method
Reviewed-by: David Zafman <dzafman@redhat.com> - 05:52 PM Feature #12038 (Fix Under Review): make rgw_fronends more forgiving of whitespace
- PR for master is https://github.com/ceph/ceph/pull/6944
- 05:06 PM Feature #12038: make rgw_fronends more forgiving of whitespace
- I've updated PR with what I recall were your original verbal comments:
1. don't warn just because "port" missing
2.... - 05:21 PM Bug #13990 (New): Hammer (0.94.3) OSD does not delete old OSD Maps in a timely fashion (maybe at ...
- 01:37 PM Bug #13990: Hammer (0.94.3) OSD does not delete old OSD Maps in a timely fashion (maybe at all?)
- https://github.com/ceph/ceph/pull/7007 is posted to address the potential issue caused by the reference (last_persist...
- 11:49 AM Bug #13990: Hammer (0.94.3) OSD does not delete old OSD Maps in a timely fashion (maybe at all?)
- we send NullEvt when a new osdmap is consumed, which also updates all PGs in the pg_map, if a PG's persisted osd map ...
- 11:48 AM Bug #13990 (Need More Info): Hammer (0.94.3) OSD does not delete old OSD Maps in a timely fashion...
- hi Tom,
> It really looks to me like nothing ever moves the cache window forward (by setting the map_cache lower b... - 05:09 PM Bug #12956: ceph-disk prepare fails for lvm logical volumes
- Misremembering. Last I noticed was https://github.com/ceph/ceph/pull/6021#issuecomment-144956928 and didn't see the r...
- 04:28 PM Revision 93497e57 (ceph): rgw: warn on suspicious civetweb frontend parameters
- Warns about incorrect spacing near 'port' (e.g., 'port = 80').
Fixes issue #12038.
Signed-off-by: Matt Benjamin <mb... - 04:25 PM Bug #14150 (Fix Under Review): OSD segmentation fault when network failue
- 08:51 AM Bug #14150: OSD segmentation fault when network failue
- Fix proposed in https://github.com/ceph/ceph/pull/7013
- 08:26 AM Bug #14150 (Resolved): OSD segmentation fault when network failue
Reproduce:
1. setup a Ceph cluster with 2+ Hosts
2. Plug out the Nic of a Host to make network_error = true. (h...- 04:24 PM rbd Bug #14129 (Resolved): librbd: snap_exists API overflow!
- 05:38 AM rbd Bug #14129: librbd: snap_exists API overflow!
- xie xingguo wrote:
> https://github.com/ceph/ceph/pull/6986
@Nathan Cutler
Resolved.
- 03:54 PM Feature #13942 (In Progress): ceph-disk: support bluestore
- 03:54 PM Feature #13942: ceph-disk: support bluestore
- encryption support will need to extend to *block* as well as *osd-data* since the data is no longer in the *osd-data*...
- 03:20 PM Feature #13942: ceph-disk: support bluestore
- For the 'ceph-disk prepare' part, I think we should keep it simple initially:
ceph-disk --osd-objectstore bluesto... - 03:17 PM Feature #13942 (In Progress): ceph-disk: support bluestore
- 03:26 PM Revision cacc38e9 (ceph): Merge pull request #7005 from YankunLi/patch-4
- delete default zone
- 03:18 PM Revision cc1edf3a (ceph): Merge pull request #7006 from YankunLi/patch-5
- correct radosgw-admin command
- 03:11 PM Revision eace76d9 (ceph): osd: Avoid osd_op_thread suicide because osd_scrub_sleep >
- tp_suicide_timeout.
Signed-off-by: Jianpeng Ma <jianpeng.ma@intel.com> - 02:01 PM Revision 714c9e3e (ceph): Merge pull request #6997 from zhouyuan/evict_check_range
- osd: cache tier: add config option for eviction check list size
Reviewed-by: Sage Weil <sage@redhat.com> - 01:59 PM Bug #13213: stuck recovering apparently due to pg_query not sent by async messenger?
- I ran rados/thrash suite here(http://pulpito.ceph.com/haomai-2015-12-20_00:32:30-rados:thrash-master-distro-basic-mul...
- 01:55 PM Revision 7b3e5bfe (ceph): Merge pull request #6990 from linuxbox2/master-fixes
- Fixes some small issues
Reviewed-by: Kefu Chai <kchai@redhat.com> - 01:42 PM Bug #14154 (Can't reproduce): OSD keep crashing
- we have several osds that keeps crashing.
here are the last bits of logs from the one of the OSDs:
-4> 2015-1... - 01:35 PM Revision 8978cd6f (ceph): osd: do not keep ref of old osdmap in pg
- do not hold a strong reference to last_persisted_osdmap in PG. as an
OSD tries to trim previously persisted osdmaps, ... - 01:13 PM Documentation #14153 (Resolved): Document forward, readonly, and readforward cache modes
- As now we have two new cache modes :
1. read-forward
2. read-proxy
http://tracker.ceph.com/projects/ceph/wik... - 11:53 AM Revision b5b17362 (ceph): osd: skip promote for writefull w/ FADVISE_DONTNEED/NOCACHE.
- Signed-off-by: Jianpeng Ma <jianpeng.ma@intel.com>
- 09:05 AM Revision b3aa29e3 (ceph): ReplicatedPG: fix sparse-read result code checking logic
- Move ahead the result code checking logic before we continue to verify the trailing hole, otherwise
the real result o... - 08:48 AM Bug #14151 (Resolved): ReplicatedPG: wrong result code checking logic during sparse_read
- See ReplicatedPG::do_osd_ops:
// verify trailing hole?
if (cct->_conf->osd_verify_s... - 07:59 AM Revision b6d0fc9e (ceph): filter lo addr when bind osd addr
- Occasionally found osd will bind a wrong ip addr
which appear in the loopback eth device. So this
ip addr need be fil... - 06:22 AM Bug #14149 (Won't Fix): ceph-mon failed to be started, if configure the source code with "enable-...
- software environment: centos7+ceph-0.94.5
prolbem: I want to run unit test and get the code coverage, so I add "--en... - 05:56 AM Revision 4dd86ca9 (ceph): correct radosgw-admin command
- the command to disable users quota should be 'radosgw-admin quota disable --quota-scope=<user | bucket> --uid=<uid>'
- 05:46 AM Revision b1df12c0 (ceph): delete default zone
- if the default zone exist, delete the default zone from both the east and west pools.
- 12:59 AM Revision 09a3f69e (ceph): Merge remote-tracking branch 'origin/jewel'
- 12:57 AM Bug #13862: pgs stuck inconsistent after infernalis upgrade
- Yes, I just wanted to confirm. So only "temp\u33.e02s0\u0\u93385203\u2__head_00000000__none_ffffffffffffffff_0" is a...
- 12:42 AM Revision 1665f818 (ceph): log: Log.cc: Assign LOG_DEBUG priority to syslog calls
- Fixes: #13993
Signed-off-by: Brad Hubbard <bhubbard@redhat.com>
(cherry picked from commit 8e93f3f45db681f82633ca695a... - 12:20 AM Revision 4d0b9a1b (ceph): log: Log.cc: Assign LOG_DEBUG priority to syslog calls
- Fixes: #13993
Signed-off-by: Brad Hubbard <bhubbard@redhat.com>
(cherry picked from commit 8e93f3f45db681f82633ca695a...
12/21/2015
- 10:51 PM rbd Bug #14122: "import failed: (95) Operation not supported" in upgrade:client-upgrade-jewel-distro-...
- Same in http://pulpito.ceph.com/teuthology-2015-12-19_09:01:14-upgrade:client-upgrade-jewel-distro-basic-mira/
- 09:44 PM rbd Bug #14147 (Closed): coredumps in rbd-jewel-distro-basic-mira
- Run: http://pulpito.ceph.com/teuthology-2015-12-19_08:59:28-rbd-jewel-distro-basic-mira/
Jobs: ['359', '366']
Logs:... - 07:17 PM Feature #12038: make rgw_fronends more forgiving of whitespace
- I did comment on the PR, or on the commit. But I can't find my comments now. Could it be that it got lost due to re-p...
- 07:03 PM Feature #12038: make rgw_fronends more forgiving of whitespace
- I've pushed branch wip-civetwarn to git@github.com:linuxbox2/ceph.git
Yehuda intended to suggest some changes, I don... - 05:58 PM Revision e8dccaca (ceph): Merge pull request #6926 from dachary/wip-14080-ceph-disk-udevadm
- ceph-disk: fix failures when preparing disks with udev > 214
On CentOS 7.1 and other operating systems with a versio... - 03:35 PM rbd Bug #14092 (Resolved): run-rbd-unit-tests.sh: various lock related asserts
- 02:08 PM rbd Bug #14092 (Fix Under Review): run-rbd-unit-tests.sh: various lock related asserts
- *jewel PR*: https://github.com/ceph/ceph/pull/7002
- 09:31 AM rbd Bug #14092: run-rbd-unit-tests.sh: various lock related asserts
- http://jenkins.ceph.dachary.org/job/ceph/LABELS=centos-7&&x86_64/10362/console...
- 09:26 AM rbd Bug #14092: run-rbd-unit-tests.sh: various lock related asserts
- 09:25 AM rbd Bug #14092: run-rbd-unit-tests.sh: various lock related asserts
- http://jenkins.ceph.dachary.org/job/ceph/LABELS=centos-7&&x86_64/10366/console...
- 03:31 PM Revision df9ff252 (ceph): Merge pull request #7002 from dillaman/wip-14092-jewel
- tests: flush op work queue prior to destroying MockImageCtx
Reviewed-by: Josh Durgin <jdurgin@redhat.com> - 03:29 PM Revision b8726972 (ceph): Merge pull request #6986 from xiexingguo/xxg-wip-14129
- librbd: fix snap_exists API return code overflow
Reviewed-by: Jason Dillaman <dillaman@redhat.com> - 03:13 PM Revision 6a45bbc8 (ceph): test: update test cases with the new snap_exists API
- Signed-off-by: xie xingguo <xie.xingguo@zte.com.cn>
- 02:55 PM Revision fbac8172 (ceph): Merge pull request #6938 from wuxiangwei/wip-wxw-rbdfuserename
- rbd-fuse: implement mv operation
Reviewed-by: Jason Dillaman <dillaman@redhat.com> - 02:42 PM CephFS Feature #14146 (Resolved): MDS: expose state of boot/replay to admins
- Right now, about all the information you can pull out is that the MDS is in standby/replay/rejoin/whatever state. We ...
- 02:41 PM Bug #14132 (Fix Under Review): common/address_help.cc: the freeaddrinfo() isn't called, leading t...
- 02:18 PM Revision 6850d47e (ceph): ECBackend: fix unmatched parenttheses
- Signed-off-by: xie xingguo <xie.xingguo@zte.com.cn>
- 02:18 PM Revision e8abb83c (ceph): rbd: stringify outputted error code
- As it is more human-readable.
Signed-off-by: xie.xingguo <xie.xingguo@zte.com.cn> - 02:03 PM Revision e8911cfe (ceph): tests: flush op work queue prior to destroying MockImageCtx
- Fixes: #14092
Signed-off-by: Jason Dillaman <dillaman@redhat.com> - 01:09 PM Bug #13990: Hammer (0.94.3) OSD does not delete old OSD Maps in a timely fashion (maybe at all?)
- by repeating adding a pool and deleting it, we can see 8 PGs in the pool:...
- 12:20 PM Bug #13990: Hammer (0.94.3) OSD does not delete old OSD Maps in a timely fashion (maybe at all?)
- ...
- 12:25 PM Bug #14094: ceph-disk list: OSError: [Errno 2] No such file or directory: '/dev/vdb1'
- The disappearance of /dev/vdb1 is a side effect of sgdisk -i indirectly triggering the equivalent of partprobe (which...
- 12:24 PM Bug #14094 (Duplicate): ceph-disk list: OSError: [Errno 2] No such file or directory: '/dev/vdb1'
- 11:16 AM Backport #14145 (In Progress): infernalis: ceph-disk: use blkid instead of sgdisk -i
- 10:51 AM Backport #14145 (Resolved): infernalis: ceph-disk: use blkid instead of sgdisk -i
- https://github.com/ceph/ceph/pull/7001
- 10:51 AM Bug #14080 (Fix Under Review): ceph-disk: use blkid instead of sgdisk -i
- 10:49 AM Bug #14080 (Pending Backport): ceph-disk: use blkid instead of sgdisk -i
- 10:37 AM Revision 9dce05a8 (ceph): ceph-disk: use blkid instead of sgdisk -i
- sgdisk -i 1 /dev/vdb opens /dev/vdb in write mode which indirectly
triggers a BLKRRPART ioctl from udev (starting ver... - 10:37 AM Revision 6395bf85 (ceph): ceph-disk: protect deactivate with activate lock
- When ceph-disk prepares the disk, it triggers udev events and each of
them ceph-disk activate. If systemctl stop ceph... - 10:31 AM Revision 5fa35ba1 (ceph): ceph-disk: do not discard stderr
- Signed-off-by: Loic Dachary <loic@dachary.org>
- 10:31 AM Revision f5d36b9a (ceph): ceph-disk: log parted output
- Should parted output fail to parse, it is useful to get the full output
when running in verbose mode.
Signed-off-by:... - 10:31 AM Revision b271a06b (ceph): ceph-disk: fix typo
- Signed-off-by: Loic Dachary <loic@dachary.org>
- 10:31 AM Revision fd7fe8c4 (ceph): tests: ceph-disk workunit increase verbosity
- So that reading the teuthology log is enough in most cases to figure out
the cause of the error.
Signed-off-by: Loic... - 10:31 AM Revision 730b5d62 (ceph): ceph-disk: increase partprobe / udevadm settle timeouts
- The default of 120 seconds may be exceeded when the disk is very slow
which can happen in cloud environments. Increas... - 10:31 AM Revision fe71647b (ceph): ceph-disk: dereference symlinks in destroy and zap
- The behavior of partprobe or sgdisk may be subtly different if given a
symbolic link to a device instead of an actual... - 09:46 AM CephFS Bug #14144 (Resolved): standy-replay MDS does not cleanup finished replay threads
- ...
- 08:27 AM Revision 0b951fde (ceph): Doc: fix misleading configuration guide on cache tiering
- * The flush/evict won't work if target_max_bytes/objects are not configured.
* All client requests will be blocked o... - 07:33 AM Revision 845f4a1c (ceph): cache-tier: Allow to config eviction check max size
- This patch adds the option for eviction check size in cache-tier.
On a busy setup, it's better to check bigger number... - 06:08 AM Bug #14139: test_rbd:delete rbd order
- https://github.com/ceph/ceph/pull/6995
- 03:32 AM Bug #14139 (Resolved): test_rbd:delete rbd order
- Delete rbd'order = RBD_DEFAULT_OBJ_ORDER process, when rbd'order is set to zero
so,delete rbd order test in test_rbd... - 06:06 AM rbd Backport #14143 (Resolved): hammer: Verify self-managed snapshot functionality on image create
- https://github.com/ceph/ceph/pull/7182
- 06:05 AM rbd Backport #14142 (Resolved): infernalis: Verify self-managed snapshot functionality on image create
- https://github.com/ceph/ceph/pull/7080
- 06:05 AM rgw Backport #14141 (Rejected): hammer: rgw: leak from RGWWatcher::handle_error
- 06:04 AM rgw Backport #14140 (Rejected): infernalis: rgw: leak from RGWWatcher::handle_error
- https://github.com/ceph/ceph/pull/7078
- 02:39 AM Linux kernel client Bug #14086: Ceph File System logging warning about ceph_set_page_dirty
- The latest test with 4.4rc5 with CONFIG_DEBUG_VM=y has ran for over 36 hours with no ERRORS or WARNINGS. My plan is ...
- 01:33 AM Backport #14137 (Fix Under Review): infernalis: Use of syslog results in all log messages at prio...
- 01:24 AM Backport #14137: infernalis: Use of syslog results in all log messages at priority "emerg"
- https://github.com/ceph/ceph/pull/6993
- 01:10 AM Backport #14137 (In Progress): infernalis: Use of syslog results in all log messages at priority ...
- 01:06 AM Backport #14137 (Resolved): infernalis: Use of syslog results in all log messages at priority "em...
- https://github.com/ceph/ceph/pull/6993
- 01:33 AM Backport #14138 (Fix Under Review): hammer: Use of syslog results in all log messages at priority...
- 01:32 AM Backport #14138 (Resolved): hammer: Use of syslog results in all log messages at priority "emerg"
- https://github.com/ceph/ceph/pull/6994
12/20/2015
- 09:10 PM Bug #14133 (Duplicate): FileJournal: create method shall return negative error code.
- 02:02 PM Bug #14133: FileJournal: create method shall return negative error code.
- Sorry, this is a duplicated request with http://tracker.ceph.com/issues/14134.
Please close this one and keep http:/... - 01:24 PM Bug #14133 (Duplicate): FileJournal: create method shall return negative error code.
- See int FileJournal::create().
- 09:10 PM Bug #14134 (Fix Under Review): FileJournal: create method shall return negative error code.
- 01:28 PM Bug #14134 (Resolved): FileJournal: create method shall return negative error code.
- See int FileJournal::create()
https://github.com/ceph/ceph/pull/6988 - 06:32 PM Revision cd2947d7 (ceph): Dispatcher.h: include assert.h
- The inline ms_fast_dispatch implementation calls assert, which
puts asset.h in the interface.
The fact that many Dis... - 06:14 PM Revision c2a24992 (ceph): xio: avoid conversion warning w/xio_queue_depth
- Signed-off-by: Matt Benjamin <mbenjamin@redhat.com>
- 05:42 PM Revision 34bb892a (ceph): xio: remove static declspec on buffer::create_msg
- Likely an update missed from PR #6686.
Signed-off-by: Matt Benjamin <mbenjamin@redhat.com> - 04:59 PM Bug #14136 (Won't Fix): "Error EINVAL: crushtool check failed with -22" in upgrade:infernalis-inf...
- Run: http://pulpito.ovh.sepia.ceph.com:8081/teuthology-2015-12-20_13:18:02-upgrade:infernalis-infernalis-distro-basic...
- 02:13 PM Bug #14135 (Can't reproduce): osd: leak of Session from ms_verify_authorizor
- <kind>Leak_DefinitelyLost</kind>
<xwhat>
<text>1,022 (408 direct, 614 indirect) bytes in 1 blocks are defin... - 01:41 PM Revision 9287e980 (ceph): FileJournal: fix return code of create method
- Shall return negative error code instead.
Fixes: #14134
Signed-off-by: xie xingguo <xie.xingguo@zte.com.cn> - 12:44 PM Revision eca5cc92 (ceph): librbd: fix wrongly reported error code
- Signed-off-by: xie xingguo <xie.xingguo@zte.com.cn>
- 09:50 AM Revision 1908d326 (ceph): Test: add tests for corrupted list/nlist process
- Signed-off-by: xie xingguo <xie.xingguo@zte.com.cn>
- 09:35 AM Revision e259d717 (ceph): Objecter: potential null pointer access in list_(n)objects.
- In list_objects and list_nobjects, we are possibly access a null returned pointer from the osdmap->get_pg_pool() call...
- 09:14 AM Revision 71eb4f9f (ceph): rbd-fuse: discard space restriction for mv operation
- discard space restriction for destination image name of
the mv operation.
Signed-off-by: Xiangwei Wu wuxiangwei@h3c.com - 09:05 AM rbd Bug #14092: run-rbd-unit-tests.sh: various lock related asserts
- http://jenkins.ceph.dachary.org/job/ceph/LABELS=centos-7&&x86_64/10346/console...
- 08:49 AM Revision 4a4048b1 (ceph): librbd: string standard error number
- It's more human-readable.
Signed-off-by: xie xingguo <xie.xingguo@zte.com.cn> - 08:49 AM Revision a7526499 (ceph): librbd: fix wrong tip message
- As equivalent size is ok for copy.
Signed-off-by: xie xingguo <xie.xingguo@zte.com.cn> - 08:23 AM Revision c97b1a68 (ceph): librbd: fix snap_exists API overflow issue
- The original one may overflow and thus not be safe.
Fixes: #14129
Signed-off-by: xie xingguo <xie.xingguo@zte.com.cn> - 08:23 AM Revision 084f22ce (ceph): tools: replace snap_exists with a new safer version
- Signed-off-by: xie xingguo <xie.xingguo@zte.com.cn>
- 07:02 AM Revision 0c31a6e1 (ceph): librbd: fix readahead counter update logic
- Fixes: #14127
Signed-off-by: xie xingguo <xie.xingguo@zte.com.cn>
12/19/2015
- 07:58 PM rbd Bug #14106 (Resolved): rbd: 'rbd create' crashed when --order option given
- 07:58 PM rbd Bug #14113 (Resolved): qa: workunit mon/rbd_snaps_ops.sh needs updating
- 07:02 PM Revision 7dca87c6 (ceph): Merge pull request #6766 from xiexingguo/xxg-wip-cleanup
- osd: fix wrongly placed assert and some cleanups
Reviewed-by: Loic Dachary <ldachary@redhat.com> - 07:01 PM Revision a96ee933 (ceph): Merge pull request #6708 from xiaoxichen/shutdown_when_flipping
- osd: shut down if we flap too many times in a short period
Reviewed-by: Sage Weil <sage@redhat.com> - 07:00 PM Revision 64cb5f6a (ceph): Merge pull request #5782 from ceph/wip-cxx11time
- common: new timekeeping common code, and Objecter conversion
Reviewed-by: Sage Weil <sage@redhat.com> - 06:59 PM Revision ae0dc478 (ceph): Merge pull request #6796 from dx9/wip-12406-sys_siglist
- common: signal_handler: added support for using reentrant strsignal() implementations vs. sys_siglist[]
Reviewed-by:... - 06:59 PM Revision 5f9f986f (ceph): Merge pull request #6814 from stiopaa1/rem_inc
- msg: removed unneeded includes from Dispatcher
Reviewed-by: Sage Weil <sage@redhat.com> - 06:58 PM Bug #13993 (Pending Backport): Use of syslog results in all log messages at priority "emerg"
- 06:58 PM Revision 0a4b7ab2 (ceph): Merge pull request #6815 from badone/wip-13993
- common: log: Assign LOG_DEBUG priority to syslog calls
Reviewed-by: Sage Weil <sage@redhat.com> - 06:58 PM Revision 2fe81803 (ceph): Merge pull request #6881 from wido/mon-codestyle-fix
- mon: fix coding-style on PG related Monitor files
Reviewed-by: Sage Weil <sage@redhat.com> - 06:57 PM Revision 6bb34fef (ceph): Merge pull request #6921 from liewegas/wip-kv-log
- osd,mon: log leveldb and rocksdb to ceph log
Reviewed-by: Kefu Chai <kchai@redhat.com> - 06:57 PM Revision 0114cb53 (ceph): Merge pull request #6950 from mslovy/wip-optimaze-map
- osd: misc optimization for map utilization
Reviewed-by: Sage Weil <sage@redhat.com> - 06:56 PM Revision ce995d7d (ceph): Merge pull request #6963 from liewegas/wip-memstore
- osd: memstore: fix two bugs
Reviewed-by: Casey Bodley <cbodley@redhat.com> - 06:46 PM Bug #13862: pgs stuck inconsistent after infernalis upgrade
- Looks like 33.f62 is no longer showing up as inconsistent. Maybe some operation performed on the pg over the last few...
- 02:02 PM rbd Bug #14126 (Won't Fix): librbd: return error if we are unable to remove the previously created se...
- This was an intentional choice -- no way to recover the snap id (not that you would need to anyway). Returning a fai...
- 08:14 AM rbd Bug #14126 (Fix Under Review): librbd: return error if we are unable to remove the previously cre...
- 02:21 AM rbd Bug #14126: librbd: return error if we are unable to remove the previously created selfmanaged_snap.
- https://github.com/ceph/ceph/pull/6983
- 01:52 AM rbd Bug #14126 (Won't Fix): librbd: return error if we are unable to remove the previously created se...
- See Internal.cc:
int validate_pool(IoCtx &io_ctx, CephContext *cct)
r = io_ctx.selfmanaged_snap_remove(snap_id);
... - 09:23 AM Bug #14132: common/address_help.cc: the freeaddrinfo() isn't called, leading to memory leaks
- https://github.com/ceph/ceph/pull/6987
- 09:00 AM Bug #14132 (Resolved): common/address_help.cc: the freeaddrinfo() isn't called, leading to memory...
- the freeaddrinfo() should be called to free resources
- 08:20 AM rbd Bug #14129 (Fix Under Review): librbd: snap_exists API overflow!
- 06:28 AM rbd Bug #14129: librbd: snap_exists API overflow!
- https://github.com/ceph/ceph/pull/6986
- 04:35 AM rbd Bug #14129 (Resolved): librbd: snap_exists API overflow!
- See Internal.cc
bool snap_exists(ImageCtx *ictx, const char *snap_name)
{
ldout(ictx->cct, 20) << "snap_ex... - 08:19 AM Bug #14130 (Fix Under Review): mon: monmap created is 0.000000 use ceph-deploy installed
- 04:58 AM Bug #14130: mon: monmap created is 0.000000 use ceph-deploy installed
- ...
- 04:55 AM Bug #14130 (Resolved): mon: monmap created is 0.000000 use ceph-deploy installed
- ...
- 08:15 AM rbd Bug #14118 (Resolved): librbd: exit if parent's snap is gone
- 02:32 AM rbd Bug #14118: librbd: exit if parent's snap is gone
- Resolved. Original PR:
https://github.com/ceph/ceph/pull/6968 - 08:14 AM rbd Bug #14127 (Fix Under Review): librbd: wrongly update counter for readahead.
- 02:25 AM rbd Bug #14127: librbd: wrongly update counter for readahead.
- https://github.com/ceph/ceph/pull/6983
- 02:25 AM rbd Bug #14127 (Resolved): librbd: wrongly update counter for readahead.
- See internal.cc
void readahead(ImageCtx *ictx,
const vector<pair<uint64_t,uint64_t> >& image_exte... - 08:14 AM Revision 70b09620 (ceph): osd: use atomic to generate ceph_tid
- Signed-off-by: Evgeniy Firsov <evgeniy.firsov@sandisk.com>
- 05:15 AM Revision 253d45bd (ceph): common/OpQueue: Add base class for Op Queues. Introduce the new Weighte...
- Signed-Off-By: Robert LeBlanc <robert.leblanc@endurance.com>
- 02:43 AM rgw Bug #14128 (Rejected): browser direct Connect radosgw
- Unable to respond to the request sent by the browser OPTIONS
2015-12-19 10:39:30.611534 7f992dffb700 20 enqueu...
12/18/2015
- 11:57 PM rbd Bug #14092: run-rbd-unit-tests.sh: various lock related asserts
- http://jenkins.ceph.dachary.org/job/ceph/LABELS=centos-7&&x86_64/10316/console...
- 07:56 PM rbd Bug #14092: run-rbd-unit-tests.sh: various lock related asserts
- http://jenkins.ceph.dachary.org/job/ceph/LABELS=centos-7&&x86_64/10307/console...
- 07:55 PM rbd Bug #14092: run-rbd-unit-tests.sh: various lock related asserts
- http://jenkins.ceph.dachary.org/job/ceph/LABELS=centos-7&&x86_64/10308/console...
- 11:44 PM Bug #13862 (Fix Under Review): pgs stuck inconsistent after infernalis upgrade
- 06:18 PM Bug #13862 (In Progress): pgs stuck inconsistent after infernalis upgrade
- I'm creating a pull request for Jewel release which will clean out these stray Hammer temp files on OSD startup. Thi...
- 11:19 PM Revision a2befde2 (ceph): Merge pull request #6982 from dillaman/wip-librbd-valgrind
- librbd: uninitialized state in snap remove state machine
Reviewed-by: Josh Durgin <jdurgin@redhat.com> - 10:23 PM Revision 6c902c30 (ceph): Merge pull request #773 from ceph/wip-14124-hammer
- Fixes issue 14124
- 10:19 PM Revision 96563fb5 (ceph): msg/Message.h:remove unneeded inline
- Member functions defined within class are implicitly inline.
Signed-off-by: Michal Jarzabek <stiopa@gmail.com> - 10:00 PM Revision e095c668 (ceph): Fixes issue 14124
- Signed-off-by: Yuri Weinstein <yweinste@redhat.com>
- 09:41 PM Revision 6d9da6a6 (ceph): librbd: uninitialized state in snap remove state machine
- Signed-off-by: Jason Dillaman <dillaman@redhat.com>
- 09:26 PM rbd Bug #14123 (Resolved): Failure in TestLibRBD.CacheMayCopyOnWrite when cache disabled
- 08:25 PM rbd Bug #14123 (Fix Under Review): Failure in TestLibRBD.CacheMayCopyOnWrite when cache disabled
- *jewel PR*: https://github.com/ceph/ceph/pull/6979
- 06:22 PM rbd Bug #14123 (Resolved): Failure in TestLibRBD.CacheMayCopyOnWrite when cache disabled
- ...
- 09:25 PM rbd Bug #14108 (Resolved): librbd: hang in api tests
- 06:17 PM rbd Bug #14108 (Fix Under Review): librbd: hang in api tests
- *jewel PR*: https://github.com/ceph/ceph/pull/6974
- 04:19 PM rbd Bug #14108 (In Progress): librbd: hang in api tests
- 09:25 PM rbd Bug #14107 (Resolved): librbd: fsx crash during clone
- 06:17 PM rbd Bug #14107 (Fix Under Review): librbd: fsx crash during clone
- *jewel PR*: https://github.com/ceph/ceph/pull/6974
- 03:23 PM rbd Bug #14107 (In Progress): librbd: fsx crash during clone
- 09:25 PM Revision d7581cda (ceph): Merge pull request #6979 from dillaman/wip-14123-jewel
- librbd: correct issues discovered when cache is disabled
Reviewed-by: Josh Durgin <jdurgin@redhat.com> - 09:13 PM Revision 73358df6 (ceph): Merge pull request #6974 from dillaman/wip-14108-jewel
- librbd: correct issues discovered during teuthology testing
Reviewed-by: Josh Durgin <jdurgin@redhat.com> - 08:42 PM Revision 28081d77 (ceph): Merge pull request #6978 from trociny/fix-rbd-init-order
- rbd: fix static initialization ordering issues
Reviewed-by: Jason Dillaman <dillaman@redhat.com> - 08:35 PM rbd Backport #14062 (Fix Under Review): infernalis: rbd merge-diff doesn't properly handle >2GB diffs
- 08:30 PM Revision d790a60d (ceph): common/PrioritizedQueue.h: Style clean up and remove uneeded includes.
- Signed-Off-By: Robert LeBlanc <robert.leblanc@endurance.com>
- 08:23 PM rbd Backport #14063 (Fix Under Review): hammer: rbd merge-diff doesn't properly handle >2GB diffs
- 08:16 PM rbd Backport #14063 (In Progress): hammer: rbd merge-diff doesn't properly handle >2GB diffs
- 07:46 PM rbd Backport #14063: hammer: rbd merge-diff doesn't properly handle >2GB diffs
- FYI https://github.com/ceph/ceph/pull/6889 does not cherry-pick cleanly to Hammer due to the RBD refactor. It looks l...
- 08:22 PM Revision 26e832e7 (ceph): librbd: fix merge-diff for >2GB diff-files
- Fixes: #14063
Signed-off-by: Jason Dillaman <dillaman@redhat.com> - 08:16 PM Revision ac964259 (ceph): osdc: Whitespace
- Since I'm making a pervasive change anyway, knock off end-of-line spaces
and wrap overly long lines.
Signed-off-by: ... - 08:16 PM Revision ecf2bebe (ceph): time: Update OSDC for C++11 Time
- Signed-off-by: Adam C. Emerson <aemerson@redhat.com>
- 08:15 PM Revision 4367eb18 (ceph): perfcounters: Allow C++11 time increment/decrement
- Signed-off-by: Adam C. Emerson <aemerson@redhat.com>
- 08:15 PM Revision 84dcfbd8 (ceph): time: Add basic support for std::chrono based time
- Signed-off-by: Adam C. Emerson <aemerson@redhat.com>
- 08:15 PM Revision 08f26766 (ceph): time: Overload for std:chrono in Cond::TimedWait
- Signed-off-by: Adam C. Emerson <aemerson@redhat.com>
- 08:15 PM Revision a3a7c037 (ceph): timer: Another timer using new time/lock standards
- Signed-off-by: Adam C. Emerson <aemerson@redhat.com>
- 08:14 PM Revision bfc9690d (ceph): librbd: skip journal event if write/discard was no-op
- Signed-off-by: Jason Dillaman <dillaman@redhat.com>
- 08:14 PM Revision f6659cc7 (ceph): librbd: object map ENOENT optimizations should complete asynchronously
- This ensures a consistent lock ordering between the optimized path and
a librados callback. Remove the cache writebac... - 08:09 PM Revision 04e29088 (ceph): Merge pull request #6701 from XinzeChi/wip-batch-peek-journal
- osd: FileJournal: support batch peak and pop from writeq
Reviewed-by: Sage Weil <sage@redhat.com> - 08:08 PM Revision 14f5d3ba (ceph): Merge pull request #6804 from ceph/putting-the-fail-in-assertion-failed
- common: assert: abort() rather than throw
Reviewed-by: Sage Weil <sage@redhat.com> - 08:08 PM Bug #13988 (Resolved): new OSD re-using old OSD id fails to boot
- 08:07 PM Revision 3aa9eb35 (ceph): Merge pull request #6900 from liewegas/wip-13988
- mon: fix reuse of osd ids (clear osd info on osd deletion)
Reviewed-by: Loic Dachary <ldachary@redhat.com> - 08:07 PM Revision 2dbb19d9 (ceph): Merge pull request #6902 from XinzeChi/wip-pg_stat_t-encoding
- osd: utime_t, eversion_t, osd_stat_sum_t encoding optimization
Reviewed-by: Sage Weil <sage@redhat.com> - 08:06 PM Revision 471164d6 (ceph): Merge pull request #6909 from mslovy/wip-shared-cache
- common: improve shared_cache and simple_cache efficiency with hash table
Reviewed-by: Sage Weil <sage@redhat.com> - 08:03 PM Bug #9218: FAILED assert(peering_queue.empty()) during OSD shutdown
- /a/sage-2015-12-17_20:33:40-rados-wip-sage-testing---basic-openstack/42535
- 07:40 PM Revision b21152a2 (ceph): rbd: fix "static initialization order fiasco"
- The order in which static objects in different compilation units are
initialized is undefined. Use static functions a... - 07:26 PM Revision 277b353e (ceph): msg/Connection::add override to virutal methods
- Signed-off-by: Michal Jarzabek <stiopa@gmail.com>
- 07:15 PM Revision 913339f7 (ceph): librbd: update exclusive lock state when shutting down
- Signed-off-by: Jason Dillaman <dillaman@redhat.com>
- 06:24 PM Revision 744a29a3 (ceph): rgw: dump_status() uses integer
- so that we can avoid atoi() later
Signed-off-by: Yehuda Sadeh <yehuda@redhat.com> - 06:21 PM Revision 4e5921db (ceph): rgw: Do not send a Content-Length header on status 204
- RFC7230 says:
A server MUST NOT send a Content-Length header field in any response with a
status code of 1xx (Inf... - 06:21 PM Revision 2a12ffc3 (ceph): rgw: move status_num initialization into constructor
- Signed-off-by: Yehuda Sadeh <yehuda@redhat.com>
- 06:14 PM Revision 200d3106 (ceph): librbd: extra request required if journal enabled w/ caching
- Fixes: #14108
Signed-off-by: Jason Dillaman <dillaman@redhat.com> - 06:09 PM Revision bab16bbb (ceph): librbd: stop the copyup thread during shutdown
- Fixes: #14107, #14108
Signed-off-by: Jason Dillaman <dillaman@redhat.com> - 05:28 PM rbd Bug #14122 (Resolved): "import failed: (95) Operation not supported" in upgrade:client-upgrade-je...
- Run: http://pulpito.ovh.sepia.ceph.com:8081/teuthology-2015-12-16_23:20:01-upgrade:client-upgrade-jewel-distro-basic-...
- 05:25 PM rbd Bug #14121 (Resolved): "[ FAILED ] TestLibRBD.RebuildObjectMap" in upgrade:client-upgrade-jewel...
- Run: http://pulpito.ovh.sepia.ceph.com:8081/teuthology-2015-12-16_23:20:01-upgrade:client-upgrade-jewel-distro-basic-...
- 05:00 PM Revision 2340fbdb (ceph): Merge pull request #6968 from xiexingguo/xxg-wip-14118
- librbd: exit if parent's snap is gone during clone
Reviewed-by: Jason Dillaman <dillaman@redhat.com>
Reviewed-by: Jo... - 04:46 PM Revision ae18bcb7 (ceph): Merge pull request #6972 from dillaman/wip-14106-jewel
- rbd: merge_diff test should use new --object-size parameter instead of --order
Reviewed-by: Josh Durgin <jdurgin@red... - 04:44 PM Revision b78dd035 (ceph): Merge pull request #772 from ceph/wip-14113-jewel
- suites/rbd: disable pool validation for pool snaps test
Reviewed-by: Josh Durgin <jdurgin@redhat.com> - 04:15 PM rbd Bug #14089 (Duplicate): ceph rbd api is not thread safe
- 04:13 PM rbd Bug #14089: ceph rbd api is not thread safe
- I believe this is a duplicate of #13636 (#13758 is the backport ticket for Hammer). It should be included in the for...
- 02:10 AM rbd Bug #14089: ceph rbd api is not thread safe
- (gdb) bt
#0 0x00007f39abea35f9 in raise () from /lib64/libc.so.6
#1 0x00007f39abea5068 in abort () from /lib64/li... - 02:01 AM rbd Bug #14089: ceph rbd api is not thread safe
- Could you open the core file that was dumped with gdb and get a backtrace?
i.e. run 'gdb python /path/to/core/file... - 12:45 AM rbd Bug #14089: ceph rbd api is not thread safe
- The core dump print is as below
pure virtual method called
terminate called without an active exception
Aborted ... - 03:56 PM rbd Bug #14106 (Fix Under Review): rbd: 'rbd create' crashed when --order option given
- *jewel PR*: https://github.com/ceph/ceph/pull/6972
- 03:15 PM rbd Bug #14106 (In Progress): rbd: 'rbd create' crashed when --order option given
- 03:52 PM Linux kernel client Bug #14086: Ceph File System logging warning about ceph_set_page_dirty
- With the 4.4rc4 kernel and the cephfs_new.patch and CONFIG_DEBUG_VM=y I hit a BUG in mm/filemap.c. ...
- 07:21 AM Linux kernel client Bug #14086: Ceph File System logging warning about ceph_set_page_dirty
- https://lkml.org/lkml/2015/12/12/113
- 03:48 PM Revision 7794bc4d (ceph): osd/ReplicatedPG: check obs.exists for CEPH_OSD_OP_CMPXATTR/SRC_CMPXATTR.
- We may use those ops(CMPXATTR, WRITE, SET_XATTR) to do a write. Using
CMPXATTR to indiate whether to write. But if ob... - 03:25 PM rbd Bug #14098 (Resolved): librbd: return error if fail to delete "object_map" head object.
- 03:23 PM rbd Bug #14113 (Fix Under Review): qa: workunit mon/rbd_snaps_ops.sh needs updating
- *jewel PR*: https://github.com/ceph/ceph-qa-suite/pull/772
- 03:21 PM rbd Bug #14113 (In Progress): qa: workunit mon/rbd_snaps_ops.sh needs updating
- 12:36 AM rbd Bug #14113 (Resolved): qa: workunit mon/rbd_snaps_ops.sh needs updating
- This fails with the new selfmanaged-snaps mode fix. From
http://teuthology.ovh.sepia.ceph.com/teuthology/teutholo... - 03:01 PM Revision 395bf8a8 (ceph): librbd: exit if parent snap is gone during clone
- Fixes: #14118
Signed-off-by: xie xingguo <xie.xingguo@zte.com.cn> - 02:49 PM Messengers Bug #14120 (Fix Under Review): Pipe::do_recv() may loop infinitely
- https://github.com/ceph/ceph/pull/6971
- 02:42 PM Messengers Bug #14120 (Resolved): Pipe::do_recv() may loop infinitely
- After cutting down network between osd and client, I saw one OSD thread use 100% CPU, The thread loops Pipe::do_recv
... - 02:43 PM Revision c93732db (ceph): msg/simple/Pipe: avoid infinite loop in Pipe::do_recv()
- I found that, sometimes an OSD thread uses 100% CPU after cutting network
between OSD and client. recv(2) in Pipe::do... - 02:14 PM Revision 6eacdc53 (ceph): Merge pull request #6914 from x11562/x11562
- rbd: additional validation for striping parameters
Reviewed-by: Jason Dillaman <dillaman@redhat.com> - 02:11 PM Revision 0f0c76b7 (ceph): osd/ReplicatedPG: for getattr_maybe_cache, it never return -EEXIST.
- Signed-off-by: Jianpeng Ma <jianpeng.ma@intel.com>
- 02:06 PM Revision deaf2c1b (ceph): Merge pull request #6958 from xiexingguo/patch-1
- librbd: return error if we fail to delete object_map head object
Reviewed-by: Jason Dillaman <dillaman@redhat.com> - 02:03 PM Revision 9caf3caa (ceph): Merge pull request #6965 from kylinstorage/wip-rbd-default-order
- rbd: use default order from configuration when not specified
Reviewed-by: Jason Dillaman <dillaman@redhat.com> - 01:47 PM Revision 0de959f9 (ceph): Drop deprecated removal pg type.
- FORREMOVAL_ pg type is deprecated and can be removed.
Signed-off-by: Igor Podoski <igor.podoski@ts.fujitsu.com> - 01:31 PM Bug #14094: ceph-disk list: OSError: [Errno 2] No such file or directory: '/dev/vdb1'
- It turns out that sgdisk 0.8.6 -i 2 /dev/vdb removes partitions and re-adds them on CentOS 7 with a 3.10.0-229.11.1.e...
- 01:23 PM Feature #7150: mon: hunt for multiple mons in parallel
- When the MonClient is in "hunting" mode, we should open connections to multiple random monitors at once, and then use...
- 01:21 PM Feature #4440 (Resolved): permanent log of all operator actions
- 12:15 PM Linux kernel client Bug #14117: KRBD : Performance issues with kernel 4.1 series
- Thanks Ilya for your feedback.
I am pasting the discussion : http://lists.ceph.com/pipermail/ceph-users-ceph.com/2... - 11:54 AM Linux kernel client Bug #14117: KRBD : Performance issues with kernel 4.1 series
- Alex posted on ceph-users and I replied there. Quoting:
"This is due to an old regression in blk-mq. rbd was swi... - 10:04 AM Linux kernel client Bug #14117: KRBD : Performance issues with kernel 4.1 series
- Ubuntu 14.04.3
- 10:04 AM Linux kernel client Bug #14117 (Won't Fix): KRBD : Performance issues with kernel 4.1 series
- I am finding that it is a specific client that is experiencing this issue. Doing an RBD write from another client ha...
- 12:14 PM Bug #13990: Hammer (0.94.3) OSD does not delete old OSD Maps in a timely fashion (maybe at all?)
- Kefu, shouldn't the backport be "infernalis, hammer" ? I don't think we're backporting to jewel yet.
- 10:33 AM Bug #13990: Hammer (0.94.3) OSD does not delete old OSD Maps in a timely fashion (maybe at all?)
- tested on master, this also applies to master.
- 09:47 AM Bug #13990: Hammer (0.94.3) OSD does not delete old OSD Maps in a timely fashion (maybe at all?)
- seems we are leaking an osdmap somewhere:...
- 10:28 AM CephFS Feature #13974: MDS: mon: expose state of log flushing
- We do expose the read_pos in a perf counter (not that friendly but it's there)
- 12:40 AM CephFS Feature #13974: MDS: mon: expose state of log flushing
- We also don't expose log replay state. Put that in at the same time.
- 10:27 AM rbd Bug #14118 (Resolved): librbd: exit if parent's snap is gone
- In internal.cc
int clone(IoCtx& p_ioctx, const char *p_name, const char *p_snap_name,
IoCtx& c_ioctx, c... - 09:15 AM Revision d4cf79b8 (ceph): mon: modify the level of a log about OSD's condition in OSDMonitor.cc
- In actual use, we replace old hard disk by new or do performance testing regularly and need to down OSDs manually, bu...
- 08:25 AM Revision 0bf7c0e0 (ceph): Merge pull request #6928 from HeyoQiang/heyongqiang1215
- Monitor: modify a dout level in OSDMonitor.cc
Reviewed-by: Joao Eduardo Luis <joao@suse.de>
Reviewed-by: Loic Dachar... - 08:03 AM Revision 69bcac51 (ceph): qa/workunits/rbd: use --object-size instead of --order
- Fixes: #14106
Signed-off-by: Jason Dillaman <dillaman@redhat.com> - 07:52 AM Revision 94cd34b6 (ceph): rbd: process crushed, rbd creating with striping parameters
- Signed-off-by: Na Xie <xie.na@h3c.com>
- 07:31 AM Revision 46dddfc5 (ceph): suites/rbd: disable pool validation for pool snaps test
- Fixes: #14113
Signed-off-by: Jason Dillaman <dillaman@redhat.com> - 07:16 AM Revision 127cd929 (ceph): rbd: use default order value from g_conf
- Signed-off-by: Yunchuan Wen <yunchuan.wen@kylin-cloud.com>
- 02:00 AM RADOS Bug #14115 (Can't reproduce): crypto: race in nss init
- From running the script from #14089 in a loop:...
12/17/2015
- 11:58 PM Bug #14080: ceph-disk: use blkid instead of sgdisk -i
- Loic,
I've emailed you directly! :)
Matt. - 11:51 AM Bug #14080: ceph-disk: use blkid instead of sgdisk -i
- In the following we see that partprobe fails immediately the first time. Since it happens right after a udevadm settl...
- 07:46 AM Bug #14080: ceph-disk: use blkid instead of sgdisk -i
- @Matthew could you please try with the updated https://raw.githubusercontent.com/dachary/ceph/f4cd0d45be68900c223f779...
- 07:42 AM Bug #14080: ceph-disk: use blkid instead of sgdisk -i
- ...
- 07:36 AM Bug #14080: ceph-disk: use blkid instead of sgdisk -i
- @Matthew the log is very useful, thank you. It would be great if you could repeat the same operation without ceph-dep...
- 07:22 AM Bug #14080: ceph-disk: use blkid instead of sgdisk -i
- I filed http://tracker.ceph.com/issues/14099 so ceph-deploy stops calling partx / partprobe (I noticed from your log ...
- 12:28 AM Bug #14080: ceph-disk: use blkid instead of sgdisk -i
- I looped through 9 disks on one of our Ceph servers using the ceph-disk script above, 8 failed and 1 completed.
Lo... - 12:05 AM Bug #14080: ceph-disk: use blkid instead of sgdisk -i
- Tried using https://raw.githubusercontent.com/dachary/ceph/ff273c9654b27b629c3e9e8f06c868ce225eb458/src/ceph-disk (su...
- 11:35 PM rbd Bug #14092: run-rbd-unit-tests.sh: various lock related asserts
- http://jenkins.ceph.dachary.org/job/ceph/LABELS=centos-7&&x86_64/10278/console...
- 10:52 PM Revision 8e93f3f4 (ceph): log: Log.cc: Assign LOG_DEBUG priority to syslog calls
- Fixes: #13993
Signed-off-by: Brad Hubbard <bhubbard@redhat.com> - 09:41 PM Revision af44fe64 (ceph): Merge pull request #6847 from oritwas/wip-fix-warning
- build: fix a few warnings
Reviewed-by: Sage Weil <sage@redhat.com> - 09:39 PM Revision 8a3e55f6 (ceph): Merge pull request #6924 from stiopaa1/removefriend
- msg/simple: remove unneeded friend declarations
Reviewed-by: Sage Weil <sage@redhat.com> - 08:38 PM Revision a32dd5e3 (ceph): Merge pull request #6405 from jcsp/wip-9963
- librados: new style (sharded) object listing
Reviewed-by: Sage Weil <sage@redhat.com> - 07:49 PM Linux kernel client Bug #14086: Ceph File System logging warning about ceph_set_page_dirty
- I have applied both cephfs.patch and cephfs1.patch provided on the dev mail list and re-ran the test. I am now seein...
- 08:54 AM Linux kernel client Bug #14086: Ceph File System logging warning about ceph_set_page_dirty
- I patched the Linux 4.4rc4 kernel with the patch supplied by Yan, Zheng on the dev mail list, and shortly after reboo...
- 07:39 PM rbd Bug #14108 (Resolved): librbd: hang in api tests
- In this run:
http://teuthology.ovh.sepia.ceph.com/teuthology/teuthology-2015-12-16_23:00:03-rbd-jewel-distro-basic... - 07:31 PM rbd Bug #14107 (Resolved): librbd: fsx crash during clone
- From http://teuthology.ovh.sepia.ceph.com/teuthology/teuthology-2015-12-16_23:00:03-rbd-jewel-distro-basic-openstack/...
- 07:06 PM rbd Bug #14106 (Resolved): rbd: 'rbd create' crashed when --order option given
- From http://teuthology.ovh.sepia.ceph.com/teuthology/teuthology-2015-12-16_23:00:03-rbd-jewel-distro-basic-openstack/...
- 07:05 PM rgw Bug #14105 (Need More Info): adminops subuser specify S3 keys
- The subuser create interface in the adminops api does not allow you to to specify a secret key and access key for s3....
- 06:26 PM Feature #14104 (New): add backfill targets to osd stats and a way to dump all osd stats to json
- 04:57 PM Bug #14096 (Duplicate): s3tests failed in upgrade:hammer-hammer-distro-basic-openstack
- 04:32 PM Bug #14102 (Resolved): "Syntax error .. Cannot load .../mod_fastcgi.so" in upgrade:hammer-x-jewel...
- We had disabled fastcgi repos and package installs immediately after the move; reverted that change as we have the or...
- 04:14 PM Bug #14102 (Resolved): "Syntax error .. Cannot load .../mod_fastcgi.so" in upgrade:hammer-x-jewel...
- Run: http://pulpito.ovh.sepia.ceph.com:8081/teuthology-2015-12-16_17:02:01-upgrade:hammer-x-jewel-distro-basic-openst...
- 04:30 PM Revision fc39e6fe (ceph): os/MemStore: fix use-after-free on iterator
- Signed-off-by: Sage Weil <sage@redhat.com>
- 04:30 PM Revision 750e4127 (ceph): memstore: fix loop in PageSetObject::clone()
- Signed-off-by: Casey Bodley <cbodley@redhat.com>
- 04:28 PM Revision 0e77e3b1 (ceph): common/buffer: reset iterator on rebuild
- When we rebuild we implicitly invalidate any iterator into the
old list of bufferptrs. Reset the internal last_p ite... - 04:28 PM Revision b88561a1 (ceph): common/buffer: reset last_p on get_contiguous
- This may modify _buffers.
Signed-off-by: Sage Weil <sage@redhat.com> - 04:18 PM Bug #14103 (Closed): "FileStore.cc: 2757: FAILED assert(0 == "unexpected error")" in upgrade:hamm...
- Run: http://pulpito.ovh.sepia.ceph.com:8081/teuthology-2015-12-16_17:02:01-upgrade:hammer-x-jewel-distro-basic-openst...
- 04:07 PM Bug #13350: "Error cancelling global ust listener thread" errors
- More runs:
http://teuthology.ovh.sepia.ceph.com/teuthology/teuthology-2015-12-16_17:02:01-upgrade:hammer-x-jewel-dis... - 03:16 PM Revision 9739d4de (ceph): Merge pull request #6544 from liewegas/wip-smaller-object-info
- osd: make encoded object_info_t smaller to fit inside the XFS inode
Reviewed-by: Samuel Just <sjust@redhat.com> - 02:54 PM Bug #13827: OSD takes suicide timeout during deep_scrub on pg after upgrade to Hammer 94.5
- Previous version of Firefly prior to Hammer 94.5 upgrade:
"common": "0.80.9-1precise"
"osd": "0.80.9-2-g62645d3-1... - 02:51 PM Bug #13827: OSD takes suicide timeout during deep_scrub on pg after upgrade to Hammer 94.5
- Deep scrub was disabled for several days and saw no further crashes. On 12/14 deep scrub was re-enabled and within th...
- 02:17 PM Revision 9d874f1d (ceph): Merge pull request #6945 from jcsp/wip-blacklist-clear
- mon: add `osd blacklist clear`
Reviewed-by: Joao Eduardo Luis <joao@suse.de> - 02:09 PM Bug #14095: FreeBSD testsuite generates traps on os/FileJournal.cc:1043
- Just to make sure I tested the 2 expressions in the if().
2015-12-17 13:45:54.550389 805217400 -1 journal bl.is_ali... - 12:02 PM Bug #14095: FreeBSD testsuite generates traps on os/FileJournal.cc:1043
- Added the debug code, and this is from running
test-ceph-helpers.sh
running /usr/srcs/Ceph/wip-freebsd-wjw/ceph/... - 01:09 AM Bug #14095: FreeBSD testsuite generates traps on os/FileJournal.cc:1043
- diff --git a/src/os/FileJournal.cc b/src/os/FileJournal.cc
index 34c2f44..3ae10f6 100644
--- a/src/os/FileJournal.c... - 02:04 PM Revision 3a951650 (ceph): Merge pull request #6959 from oritwas/wip-fix-run-cmake-check
- cmake: no need to run configure from run-cmake-check.sh
Reviewed-by: Loic Dachary <ldachary@redhat.com> - 01:59 PM Revision 70240f7f (ceph): Merge pull request #6564 from branch-predictor/bp-fix-mon-kv-store
- mon: fixes related to mondbstore->get() changes
Reviewed-by: Sage Weil <sage@redhat.com> - 01:58 PM Revision 4060d21e (ceph): Merge pull request #6202 from cxwshawn/wip-ful-fix
- rados: implement rm --force option to force remove when full
Reviewed-by: Kefu Chai <kchai@redhat.com> - 01:52 PM Revision 66346607 (ceph): common/lockdep: increase max lock names
- Recently hit this limit with
/a/sage-2015-12-16_21:46:23-rados-wip-sage-testing---basic-openstack/40275
Signed-off... - 01:38 PM Revision 3167608b (ceph): rgw: use unsigned long to print to log
- Signed-off-by: Orit Wasserman <owasserm@redhat.com>
- 01:38 PM Revision 95afcad9 (ceph): newstore: fix was hidden warning
- Signed-off-by: Orit Wasserman <owasserm@redhat.com>
- 01:38 PM Revision 3d6e51c4 (ceph): rbd: fix "was hidden" warning
- Signed-off-by: Orit Wasserman <owasserm@redhat.com>
- 01:35 PM Revision 03f9c6f2 (ceph): Merge pull request #6956 from somnathr/wip-conditional-drive-metadata-c...
- osd: FileStore: conditional collection of drive metadata
Reviewed-by: Sage Weil <sage@redhat.com> - 01:27 PM Revision cf5da012 (ceph): Merge pull request #6929 from clever215/mon-log
- mon: degrade a log message to level 2
Reviewed-by: Joao Eduardo Luis <joao@suse.de> - 01:24 PM RADOS Bug #14100: the semantics of CEPH_OSD_OP_ZERO.
- Perhaps a simple fix is to add a flag argument to the operation like FLAG_EXTEND_SIZE to cover the non-hole-punchy ca...
- 09:05 AM RADOS Bug #14100: the semantics of CEPH_OSD_OP_ZERO.
- jianpeng ma wrote:
> Now in ReplicatedPG::do_osd_ops, for CEPH_OSD_OP_ZERO, there are the following handle:
>
> a... - 07:23 AM RADOS Bug #14100 (New): the semantics of CEPH_OSD_OP_ZERO.
- Now in ReplicatedPG::do_osd_ops, for CEPH_OSD_OP_ZERO, there are the following handle:
a)if object din't exist, th... - 01:23 PM Bug #14101 (Can't reproduce): newstore/NewStore.cc: 1591: FAILED assert(k >= start_key && k < end...
- Hit this asset when doing scrub.
-1> 2015-12-15 17:02:33.729448 7fc48a08b700 5 -- op tracker -- seq: 177, time:... - 01:03 PM Revision 432d26cb (ceph): cmake: no need to run configure from run-cmake-check.sh
- Signed-off-by: Orit Wasserman <owasserm@redhat.com>
- 12:17 PM Bug #14094: ceph-disk list: OSError: [Errno 2] No such file or directory: '/dev/vdb1'
- ...
- 11:02 AM Linux kernel client Bug #12763: rbd: unmap failed: (16) Device or resource busy
- There was a similar issue reported on the mailing list - http://www.spinics.net/lists/ceph-devel/msg27435.html. The ...
- 10:42 AM Linux kernel client Bug #12763: rbd: unmap failed: (16) Device or resource busy
- Ilya Dryomov wrote:
> Yeah, sorry - rbd unmap returns -EBUSY, not hangs. So there is nothing in /sys/kernel/debug/c... - 09:50 AM Linux kernel client Bug #12763: rbd: unmap failed: (16) Device or resource busy
- Yeah, sorry - rbd unmap returns -EBUSY, not hangs. So there is nothing in /sys/kernel/debug/ceph/<id>/osdc after it ...
- 09:37 AM Linux kernel client Bug #12763: rbd: unmap failed: (16) Device or resource busy
- ceph zte wrote:
> Thank you for your answer. It is too long.The prolem above only in that vmware machine can show.
... - 09:29 AM Linux kernel client Bug #12763: rbd: unmap failed: (16) Device or resource busy
- Ilya Dryomov wrote:
> If you post a reproducer - one that starts with the installation of multipathd package, then s... - 09:25 AM Linux kernel client Bug #12763: rbd: unmap failed: (16) Device or resource busy
- Ilya Dryomov wrote:
> If you post a reproducer - one that starts with the installation of multipathd package, then s... - 09:18 AM Linux kernel client Bug #12763: rbd: unmap failed: (16) Device or resource busy
- Thank you for your answer. It is too long.The prolem above only in that vmware machine can show.
In our oter machine... - 09:14 AM Linux kernel client Bug #12763: rbd: unmap failed: (16) Device or resource busy
- If you post a reproducer - one that starts with the installation of multipathd package, then sets up multipath and hi...
- 07:17 AM Linux kernel client Bug #12763: rbd: unmap failed: (16) Device or resource busy
- when doing rbd map with a process of multipathd, this issue will come up, with no doubt. The weird thing is both mult...
- 09:39 AM Revision df45ba35 (ceph): Merge pull request #771 from dachary/wip-buildpackages-centos-7.1-hammer
- buildpackages: support CentOS 7.1 and CentOS 7.2
- 09:38 AM Revision 2cfe7ceb (ceph): buildpackages: support CentOS 7.1 and CentOS 7.2
- Signed-off-by: Loic Dachary <loic@dachary.org>
(cherry picked from commit 834a38cf90774802c92ba34ae28aab6987a5d4d6) - 09:35 AM Revision 686d6e9c (ceph): Merge pull request #769 from dachary/wip-buildpackages-centos-7.1
- buildpackages: support CentOS 7.1 and CentOS 7.2
- 09:31 AM Revision 834a38cf (ceph): buildpackages: support CentOS 7.1 and CentOS 7.2
- Signed-off-by: Loic Dachary <loic@dachary.org>
- 09:30 AM Revision 42692630 (ceph): crush: fix the default type 0 name if it's not specified when decompiled
- Change the type 0 name from "device" to "osd", just to make it consistent
Signed-off-by: Sangdi Xu <xu.sangdi@h3c.com> - 09:28 AM Revision 895bf8b8 (ceph): crushtool: set type 0 name "osd" for --build option
- The crushmap built by crushtool cannot be used directly as the name of type 0 is not specified.
We may decompile the ... - 08:28 AM CephFS Bug #14048: MDS heap stats standby threads
- Vitaliy Dutchak wrote:
> Sorry for question. It's I am need to review or CephFs team? )
We will review it, If you... - 06:16 AM CephFS Bug #14048: MDS heap stats standby threads
- Sorry for question. It's I am need to review or CephFs team? )
- 07:10 AM Revision d9144481 (ceph): OSD: fix null pointer access and race condition
- We are risking of accessing a null osdmap pointer when we failed to handle an osdmap message due to network error.
Al... - 06:52 AM Revision 54d6b47a (ceph): Merge pull request #6864 from scienceluo/wip-cleanup
- osdmap: remove unused local variables
Reviewed-by: Loic Dachary <ldachary@redhat.com> - 06:51 AM Feature #13505: scrub/repair: persist scrub results.
- note to myself, in the last discussion with david, he advised that we should not overwrite the scrub result of deep s...
- 06:43 AM Feature #13505: scrub/repair: persist scrub results.
- wondering how can we fix the snapset inconsistencies like
# snap missing in snapset,clone_overlap
# snapset.clone... - 06:13 AM Revision 7daa1ccf (ceph): librbd: return error if we fail to delete object_map head object
- Fixes: #14098
Signed-off-by: xie xingguo <xie.xingguo@zte.com.cn> - 06:02 AM rbd Bug #14098 (Resolved): librbd: return error if fail to delete "object_map" head object.
- See remove_object_map:
r = ictx->md_ctx.remove(ObjectMap::object_map_name(ictx->id, CEPH_NOSNAP));
if (r < ... - 03:56 AM Revision 2e17c34b (ceph): osdmap: remove unused local variables
- Signed-off-by: luo kexue <luo.kexue@zte.com.cn>
- 02:40 AM Bug #13862: pgs stuck inconsistent after infernalis upgrade
- Logan can you find the temp_33.f62s0_0_93345312_70 in the filestore and tell me the path inside the filestore and act...
- 01:38 AM rbd Bug #14089: ceph rbd api is not thread safe
- I run the below python script in Multithreading.Such as 100 threads.Sometimes it usually has the below errors like "E...
- 01:21 AM Revision 6e02d328 (ceph): Merge pull request #6904 from wuxiangwei/wip-wxw-rbdadminsocket
- rbd: add pool name to disambiguate rbd admin socket commands
Reviewed-by: Josh Durgin <jdurgin@redhat.com> - 01:18 AM Revision 8acfddc9 (ceph): Merge pull request #6942 from x11562/stripe
- rbd: striping parameters should support 64bit integers
Reviewed-by: Jason Dillaman <dillaman@redhat.com>
Reviewed-by...
12/16/2015
- 11:54 PM Bug #14080: ceph-disk: use blkid instead of sgdisk -i
- Loic,
15 seconds sleep (failed): http://pastebin.com/g5J5G5Wq
25 second sleep (failed): http://pastebin.com/mjMfs... - 11:48 PM Bug #14080: ceph-disk: use blkid instead of sgdisk -i
- @Matthew : using partx is appealing (and that's what ceph-disk used in previous versions of Ceph). The reason partpro...
- 11:09 PM Bug #14080: ceph-disk: use blkid instead of sgdisk -i
- As I had mentioned in ceph-users (http://lists.ceph.com/pipermail/ceph-users-ceph.com/2015-December/006780.html), I w...
- 03:40 PM Bug #14080: ceph-disk: use blkid instead of sgdisk -i
- CentOS Linux 7.1.1503 Core
sdb is a spinning drive (HGST HUS724020ALA640)
solved by putting time.sleep(5) befor... - 12:11 PM Bug #14080: ceph-disk: use blkid instead of sgdisk -i
- Immediately after a disk was zapped, udevadm settle waited for ~2mn30 on an OpenStack instance: this is more than the...
- 11:53 PM Bug #14097 (Closed): "rcu_sched self-detected stall on CPU" failed jobs in upgrade:hammer-hammer-...
- Run: http://pulpito.ovh.sepia.ceph.com:8081/teuthology-2015-12-16_16:05:02-upgrade:hammer-hammer-distro-basic-opensta...
- 11:49 PM Bug #14096 (Duplicate): s3tests failed in upgrade:hammer-hammer-distro-basic-openstack
- Run: http://pulpito.ovh.sepia.ceph.com:8081/teuthology-2015-12-16_16:05:02-upgrade:hammer-hammer-distro-basic-opensta...
- 10:02 PM Revision 3296905f (ceph): FileStore: Conditional collection of drive metadata
- get_device_by_uuid->blkid_find_dev_with_tag() call from
FileStore::collect_metadata() is hanging for ~3min before ret... - 08:11 PM Bug #14094: ceph-disk list: OSError: [Errno 2] No such file or directory: '/dev/vdb1'
- Or just that a partprobe is ongoing and /dev/sdb1 is re-created while ceph-disk list. Maybe protecting ceph-disk list...
- 08:06 PM Bug #14094: ceph-disk list: OSError: [Errno 2] No such file or directory: '/dev/vdb1'
- ...
- 07:02 PM Bug #14094 (Duplicate): ceph-disk list: OSError: [Errno 2] No such file or directory: '/dev/vdb1'
- After a failed run of the ceph-disk suite on CentOS 7.1, ceph-disk list tries to stat /dev/vdb1 although it does not ...
- 08:05 PM Bug #14095 (Resolved): FreeBSD testsuite generates traps on os/FileJournal.cc:1043
- After patching the src tree so that just about everything compiles
Running 'gmake check' some of the tests fail.
On... - 06:07 PM Bug #14093 (New): scripts/ceph-release-notes does not handle merge description
- The merge description may be different from the pull request description. When that's the case and the merge descript...
- 06:04 PM Revision 0dba0e19 (ceph): Merge remote-tracking branch 'gh/jewel'
- 06:03 PM Revision fbd5959f (ceph): doc/release-notes: v10.0.1
- Signed-off-by: Sage Weil <sage@redhat.com>
- 03:48 PM Revision 91bf56e2 (ceph): Merge remote-tracking branch 'gh/jewel'
- 03:01 PM rbd Bug #14092 (Resolved): run-rbd-unit-tests.sh: various lock related asserts
- http://jenkins.ceph.dachary.org/job/ceph/LABELS=centos-7&&x86_64/10235/console...
- 02:40 PM Bug #13862: pgs stuck inconsistent after infernalis upgrade
- Interesting.. thanks for the info. In advance of a fix, should I be able to fix these pgs by removing the obsolete te...
- 04:20 AM Bug #13862: pgs stuck inconsistent after infernalis upgrade
This is a hammer temp object which no longer applies in Infernalis.
-1/00000000/temp_33.f62s0_0_93345312_70/head...- 02:08 PM rbd Bug #14089 (Need More Info): ceph rbd api is not thread safe
- What is the exact issue you are encountering?
- 09:15 AM rbd Bug #14089 (Duplicate): ceph rbd api is not thread safe
- My ceph version is 0.94.When i use the Multithreading run rbd api like below.
The librbd will core dump.Why the li... - 01:46 PM Revision 4fc41028 (ceph): Merge pull request #6932 from aiicore/test_fix
- tests: allow object corpus readable test to skip specific incompat instances
Reviewed-by: Sage Weil <sage@redhat.com> - 12:10 PM CephFS Bug #14029 (Resolved): "[ FAILED ] LibCephFS.OpenLayout" in smoke-master-distro-basic-openstack
- 12:02 PM Revision 42389742 (ceph): mon: add `osd blacklist clear`
- This is just like 'blacklist rm' except it removes
everything. Useful if you've got a whole bunch of
things in your ... - 11:55 AM Revision db8bf2dc (ceph): Merge pull request #6941 from jcsp/wip-14029
- mds: fix setvxattr (broken in a536d114)
- 10:56 AM rgw Bug #14090: rgw: the response XML format of admin policy API is error
- the fix is:
https://github.com/ceph/ceph/pull/6951 - 10:35 AM rgw Bug #14090 (Resolved): rgw: the response XML format of admin policy API is error
- * reproduce step:
$ curl -s -X GET http://ceph1:9090/admin/bucket?format=xml&policy&bucket=m8x | xmllint --format ... - 10:55 AM rgw Fix #14091: rgw: support json format for admin policy API
- the fix is:
https://github.com/ceph/ceph/pull/6951 - 10:42 AM rgw Fix #14091 (New): rgw: support json format for admin policy API
- when use admin policy API with format=json parameter,
$ curl -s -X GET http://ceph1:9090/admin/bucket?format=json&p... - 10:33 AM Bug #13850: master fails to build when liblz4 is present (rocksdb)
- I got into this bug trying to build ceph on a machine that runs Arch. Since arch does not have different packages for...
- 09:57 AM rbd Bug #14084 (Fix Under Review): rbd:when ctx_budget is empty,the logical assertion may arise in th...
- 07:43 AM rbd Bug #14084: rbd:when ctx_budget is empty,the logical assertion may arise in the function Objecter...
- https://github.com/ceph/ceph/pull/6935
- 09:52 AM Fix #10443: compressing ceph-test-dbg takes a long time
- I stumbled upon this while building test packages. It's single-threaded, too. Takes 26 minutes for me.
Perhaps it ... - 08:24 AM Revision 4276fb3f (ceph): Delete rbd'order define
- Delete rbd'order define
Fixes: #14047
Signed-off-by: huanwen ren <ren.huanwen@zte.com.cn> - 08:21 AM Revision 16d50459 (ceph): Delete rbd'order setup process
- Delete rbd'order = RBD_DEFAULT_OBJ_ORDER process, when rbd'order is set to zero
Fixes: #14047
Signed-off-by: huanwen... - 08:02 AM Bug #14028: PG stats are not refresh within osd_pg_stat_report_interval_max
- Samuel Just wrote:
> I'm not sure this is a bug. The OSDs need to send a stat message so the mons don't assume they... - 06:31 AM Revision 51e956bf (ceph): fix OSD utilization is abnormal after data disk lost
- in function "OSDService::update_osd_stat", local variable "stbuf"" isn't initialized and it's still used to
update th... - 06:13 AM Revision ec42514f (ceph): osd: use unordered_map for repop_map
- With a prediction of expected per pg maximum ops to initialize the number of hash bucket
Signed-off-by: Ning Yao <za... - 03:47 AM Revision d13f6024 (ceph): rbd:add destination image name validation for rbd-fuse mv operation
- add validation for destination image name to
avoid image name contains invalid characters, such as "/" and "@".
Sign... - 01:48 AM Revision 82841656 (ceph): rbd: process crushed, rbd creating with striping parameters
- Signed-off-by: Na Xie <xie.na@h3c.com>
- 01:47 AM Revision 1d7b53f4 (ceph): Merge pull request #6948 from ceph/revert-6939-bp-remove-bz2dev
- Revert "makefiles: remove bz2-dev from dependencies"
Reviewed-by: Dan Mick <dmick@redhat.com> - 01:43 AM Revision a204f6f9 (ceph): repair_test: Add whitelist item for new stat error message
- Signed-off-by: David Zafman <dzafman@redhat.com>
- 01:09 AM Revision e7b7e1af (ceph): rgw: add a method to purge all associate keys when removing a subuser
- Fixes: #12890
When removing a subuser, make sure all of its keys, including the swift key and possible s3 keys, are ...
12/15/2015
- 11:56 PM Cleanup #13981 (In Progress): configure script doesn't check bzip2-devel presence
- Turns out the new patch has broken gitbuilders: http://gitbuilder.sepia.ceph.com/gitbuilder-ceph-rpm-centos7-amd64-ba...
- 02:47 PM Cleanup #13981 (Resolved): configure script doesn't check bzip2-devel presence
- 01:16 PM Cleanup #13981 (In Progress): configure script doesn't check bzip2-devel presence
- 01:15 PM Cleanup #13981 (New): configure script doesn't check bzip2-devel presence
- 01:08 PM Cleanup #13981: configure script doesn't check bzip2-devel presence
- https://github.com/ceph/ceph/pull/6939
- 11:12 PM Revision 35542eb6 (ceph): Revert "makefiles: remove bz2-dev from dependencies"
- 08:10 PM RADOS Bug #14088 (Resolved): mon: nothing logged when ENOSPC encountered during start up
- For example, ceph-mon process was dying and reporting only the
following to dmesg, nothing to /var/log/ceph/mon.log
... - 07:39 PM rgw Bug #13764: Radosgw incomplete files
- The manifest is normal. The object stripe seem to be missing. There's not much else that I can see in here. Do you ha...
- 07:00 PM rgw Bug #14046: When I get AWS JS API to create buckets, using S3 Browser delete buckets, radosgw cra...
- also if you could add 'debug ms = 1' it would help.
- 06:58 PM rgw Bug #14046: When I get AWS JS API to create buckets, using S3 Browser delete buckets, radosgw cra...
- can you reproduce the issue using the debug packages? The backtrace here doesn't provide any interesting info
- 06:49 PM rgw Bug #13680 (Resolved): New keystone expire dates cannot be parsed
- 06:39 PM rgw Bug #13529 (Pending Backport): rgw: leak from RGWWatcher::handle_error
- Merged in commit:0857da4c213992feadd6a944f1c2ec6dfe1cf98c
- 05:39 PM Backport #14087 (Fix Under Review): FAILED assert(old_size == total_chunk_size) in 0.94.5 / stran...
- 05:19 PM Backport #14087 (Resolved): FAILED assert(old_size == total_chunk_size) in 0.94.5 / strange file ...
- https://github.com/ceph/ceph/pull/6946
- 05:36 PM Revision f04e0075 (ceph): osd: log inconsistent shard sizes
- When an OSD asserts because it cannot recover from an unexpected erasure
coded shard size, the object needs to be man... - 04:58 PM Bug #14009: FAILED assert(old_size == total_chunk_size) in 0.94.5 / strange file size
- I think the problem is fixed in infernalis via https://github.com/ceph/ceph/pull/5173 . It also is my understanding t...
- 04:15 PM Bug #14009: FAILED assert(old_size == total_chunk_size) in 0.94.5 / strange file size
- The bug is about the OSDs that shouldn't crash in this case where only one shard in a EC 8+2 setting is corrupted in ...
- 03:38 PM Bug #14009 (Can't reproduce): FAILED assert(old_size == total_chunk_size) in 0.94.5 / strange fil...
- Reopen if we get more information about how these objects came to be the wrong size.
- 04:50 PM Linux kernel client Bug #14022 (Need More Info): map_sem for read + request_mutex are held indefinitely
- 04:46 PM rbd Bug #14082 (Rejected): test_JournalPlayer.cc:380: Failure
- issue is present in pending PR only
- 07:49 AM rbd Bug #14082 (Rejected): test_JournalPlayer.cc:380: Failure
- http://jenkins.ceph.dachary.org/job/ceph/LABELS=centos-7&&x86_64/10166/console...
- 04:45 PM rbd Bug #14085: TestInternal.ShrinkFlushesCache librbd/test_internal.cc:686: Value of: c->is_complete()
- ...
- 09:18 AM rbd Bug #14085 (Can't reproduce): TestInternal.ShrinkFlushesCache librbd/test_internal.cc:686: Value ...
- http://jenkins.ceph.dachary.org/job/ceph/LABELS=centos-7&&x86_64/10183/console...
- 04:37 PM rbd Bug #13549 (Duplicate): Locking assertion failure in librbd/LibrbdWriteback.cc
- 04:21 PM devops Bug #13860 (Resolved): openSUSE Tumbleweed build cannot find systemd-tmpfiles
- Only jewel will go into openSUSE Tumbleweed, so there's no point in backporting this to infernalis.
- 04:20 PM devops Bug #13860 (Pending Backport): openSUSE Tumbleweed build cannot find systemd-tmpfiles
- 04:19 PM Linux kernel client Bug #14086 (Closed): Ceph File System logging warning about ceph_set_page_dirty
- While testing Linux Target SCSI, LIO, with a Ceph File System file backstore, that is kernel mounted, I am seeing thi...
- 03:43 PM Bug #13991 (Won't Fix): osd: fix ScrubJob compare function
- 03:43 PM Bug #13991 (Can't reproduce): osd: fix ScrubJob compare function
- 03:43 PM Bug #13991: osd: fix ScrubJob compare function
- in practice, it does not happen. as the chance that A.sched_time == B.sched_time is rare.
- 03:38 PM Revision 79c7ccb9 (ceph): osd: combine map.count() and map.find() into one searching operation
- Signed-off-by: Ning Yao <zay11022@gmail.com>
- 03:37 PM Bug #14021: osd coredump on rhel7.1
- This looks like a thread timed out. You'll want to reproduce with logging and determine which thread.
- 03:36 PM Bug #14028: PG stats are not refresh within osd_pg_stat_report_interval_max
- I'm not sure this is a bug. The OSDs need to send a stat message so the mons don't assume they died, but the PGs don...
- 03:32 PM Bug #13989 (Duplicate): OSD boot fails with os/FileJournal.cc: 1907: FAILED assert(0)
- 03:28 PM Revision ff9712b2 (ceph): Merge pull request #6692 from SUSE/wip-13860
- ceph.spec.in: add BuildRequires: systemd
Reviewed-by: Ken Dreyer <kdreyer@redhat.com> - 02:43 PM Revision 34ceddf0 (ceph): Merge pull request #6939 from branch-predictor/bp-remove-bz2dev
- makefiles: remove bz2-dev from dependencies
Reviewed-by: Kefu Chai <kchai@redhat.com> - 02:38 PM Bug #14072 (Fix Under Review): OSD: null pointer access when do handle_osd_map.
- https://github.com/ceph/ceph/pull/6916
- 01:34 PM Bug #14072: OSD: null pointer access when do handle_osd_map.
- Could you please include the snippet of code you believe to be problematic ?
- 02:33 PM Revision 09c57ff5 (ceph): Merge pull request #6943 from smithfarm/wip-doc-dev-typo
- doc: fix typo in developer guide
Reviewed-by: Loic Dachary <ldachary@redhat.com> - 02:32 PM Bug #14074 (Need More Info): OSD daemon can not start after cluster deployment
- ...
- 02:29 PM Bug #12956 (Won't Fix): ceph-disk prepare fails for lvm logical volumes
- @joejulian I think to remember we concluded that won't be fixed. Am I misremembering ?
- 02:25 PM Revision fec54583 (ceph): doc: fix typo in developer guide
- Signed-off-by: Nathan Cutler <ncutler@suse.com>
- 02:25 PM Bug #13403 (Rejected): Error running "mkfs.xfs" in upgrade:hammer-x-infernalis-distro-basic-multi...
- I'm closing this as an environmental issue since the disk was busy. If you think it's not environmental, it probably ...
- 02:23 PM Feature #13419: On osd startup we should log all configuration settings
- +1 :-)
- 02:23 PM Bug #13421 (Won't Fix): ceph-osd Segmentation fault in pthread_join
- I'm closing this because giant is no longer supported. If you think this bug is not giant specific, please let me kno...
- 02:22 PM Bug #13428 (Rejected): multiple OSDs crashed on same node
- The only way ceph::buffer::create_aligned can fail is if memory allocation fails. I think the failures you saw was be...
- 02:17 PM Bug #13433 (Resolved): "SELinuxError: SELinux denials found "
- Feel free to re-open if this is still a problem. I think to remember selinux related problems got fixed.
- 02:16 PM rbd Documentation #14070 (Resolved): admin/build-doc fails to build because of -lrbd
- 11:55 AM rbd Documentation #14070: admin/build-doc fails to build because of -lrbd
- *master PR*: https://github.com/ceph/ceph/pull/6934
- 11:38 AM rbd Documentation #14070: admin/build-doc fails to build because of -lrbd
- Note that this isn't a problem with ...
- 11:33 AM rbd Documentation #14070 (In Progress): admin/build-doc fails to build because of -lrbd
- Compared to what is documented at http://docs.ceph.com/docs/master/dev/generatedocs/#install-the-required-tools we no...
- 02:15 PM Bug #13444 (Resolved): FS::zero -> "The FALLOC_FL_PUNCH_HOLE flag must be ORed with FALLOC_FL_KEE...
- 02:12 PM Bug #13503 (Resolved): CMakeList:Cannot find source file
- a lot of cmake work has been done since, it's likely fixed :-)
- 02:12 PM CephFS Bug #13903: Failure in TestStrays.test_ops_throttle
- Greg: that linked result is from TestDamage, was there something in the logs that indicated it had a common cause wit...
- 02:11 PM Bug #13527 (Duplicate): moniter segmentation fault
- 02:08 PM Bug #13555 (Won't Fix): Assert occured in PG::RecoveryState::Active::react processing AdvMap message
- Closing because giant is no longer supported. Feel free to re-open if you think the same problem happens on hammer or...
- 02:06 PM Bug #13398 (Need More Info): leveldb compaction error
- Could you please add more information about how to reproduce the problem ? Thanks :-)
- 02:03 PM Bug #13773 (Duplicate): pid files on some OSD's are not created correctly (empty)
- 02:01 PM Bug #13787 (Can't reproduce): test-erasure-code.sh: Error EINVAL: The option value '2147483647' s...
- Did not show up again.
- 02:00 PM Bug #13238: duplicate start of ceph-osd daemon
- 01:59 PM Bug #13848 (Fix Under Review): Removed objects may stay in cache forever
- 01:58 PM Bug #13848 (Duplicate): Removed objects may stay in cache forever
- 01:52 PM rbd Bug #14058 (Resolved): Can't build Ceph with --without-rbd and --with-debug
- 01:49 PM Bug #12449 (Duplicate): ceph-osd core dumped when writing data to the backing storage pool which ...
- 09:55 AM Bug #12449: ceph-osd core dumped when writing data to the backing storage pool which has a quota ...
- Looks like a duplicate of #13098 (or rather #13098 is a duplicate of this bug)
- 01:43 PM Bug #13532 (Need More Info): osd flapping
- Could you please add the steps to reproduce the problem ?
- 01:38 PM rbd Bug #14059: Can't build Ceph with --with-rbd (Cython)
- On CentOS 7:
@$ yum list installed | grep python-devel
python-devel.x86_64 2.7.5-18.el7_1.1 ... - 01:34 PM rbd Bug #14059: Can't build Ceph with --with-rbd (Cython)
- Just FYI: on CentOS 6:
@$ yum list installed | grep python-devel
python-devel.x86_64 2.6.6-64.el6 @base
@ - 01:32 PM rbd Bug #14059: Can't build Ceph with --with-rbd (Cython)
- Here's what I get on CentOS 6 with Cython 0.24 any python-devel package installed:
@
CXXLD ceph_test_librbd_fs... - 01:28 PM Revision 4bf6fbf2 (ceph): rbd: specify pool name for rbd admin socket commands
- Add the pool name for a given rbd imgae when executing rbd admin socket
commands in case there are more than one imag... - 01:28 PM Bug #13128: mixed client firefly osd with hammer client - rados api test failure
- @Vasu, did you have time to test this some more ?
- 01:22 PM CephFS Bug #14029 (Fix Under Review): "[ FAILED ] LibCephFS.OpenLayout" in smoke-master-distro-basic-o...
- https://github.com/ceph/ceph/pull/6941
- 01:05 PM CephFS Bug #14029 (In Progress): "[ FAILED ] LibCephFS.OpenLayout" in smoke-master-distro-basic-openstack
- 03:36 AM CephFS Bug #14029: "[ FAILED ] LibCephFS.OpenLayout" in smoke-master-distro-basic-openstack
- Can you look at this, John? We got this some more:
http://pulpito.ovh.sepia.ceph.com:8081/gregf-2015-12-11_09:34:0... - 01:22 PM Revision d4658301 (ceph): mds: fix setvxattr (broken in a536d114)
- There was an accidental move of this line
when adding the MAY_SET_POOL check, which
was causing setxattr to proceed b... - 01:22 PM Bug #13843 (Fix Under Review): ceph osd pool stats broken in hammer
- 01:08 PM Bug #13843: ceph osd pool stats broken in hammer
- ...
- 01:16 PM CephFS Bug #12175 (Resolved): Fix ceph-fuse --help
- ...
- 01:15 PM Revision d2781342 (ceph): Merge pull request #6821 from H3C/wip-cephfs-f11322
- ceph-fuse: add process to ceph-fuse --help
Reviewed-by: John Spray <john.spray@redhat.com> - 01:01 PM Revision a8b59202 (ceph): makefiles: remove bz2-dev from dependencies
- The only thing that uses bzip2-devel is RocksDB, and it's optional, not
requirement. Drop the bzip2-devel/libbz2-dev ... - 12:59 PM Bug #13897: (jewel) rados/objectstore/objectstore.yaml common/Mutex.cc: 105: FAILED assert(r == 0)
- I did not keep the URL of the run where it happened.
- 12:45 PM Revision 9409e96a (ceph): ceph-fuse: add process to ceph-fuse --help
- Signed-off-by: Wei Feng <feng.wei@h3c.com>
- 12:42 PM Revision 228aedff (ceph): rbd: implement mv operation for rbd-fuse
- implement mv operation for rbd-fuse to rename image.
Signed-off-by: Xiangwei Wu wuxiangwei@h3c.com - 12:35 PM Revision f4a077c7 (ceph): doc: Use 'ceph auth get-or-create' for creating RGW keyring
- This saves us a lot of steps which the old documentation took. The
result is however the same. - 12:24 PM Revision 736532ca (ceph): Merge pull request #6934 from smithfarm/wip-14070
- doc: Update list of admin/build-doc dependencies
Reviewed-by: Loic Dachary <ldachary@redhat.com> - 12:17 PM rbd Bug #14084: rbd:when ctx_budget is empty,the logical assertion may arise in the function Objecter...
- Assertion is not going to happen, but ctx_budget = NULL, run "int op_budget = _take_op_budget (op);" Is superfluous, ...
- 09:05 AM rbd Bug #14084 (Won't Fix): rbd:when ctx_budget is empty,the logical assertion may arise in the funct...
- Although "ctx_budget = NULL" && "initialize the op->ctx_budgeted = false" assertions will not trigger calls * ctx_bud...
- 11:41 AM Revision eadc771d (ceph): doc: Update list of admin/build-doc dependencies
- http://tracker.ceph.com/issues/14070 Fixes: #14070
Signed-off-by: Nathan Cutler <ncutler@suse.com> - 11:23 AM CephFS Bug #14048 (Fix Under Review): MDS heap stats standby threads
- https://github.com/ceph/ceph/pull/6933
https://github.com/ceph/ceph-qa-suite/pull/767 - 11:20 AM Revision b1429aa7 (ceph): mds: tear down connections from `tell` commands
- We can identify sessions that were never really
opened, just created when a client sent in
a 'tell' message. When we... - 11:08 AM Revision 8d68d023 (ceph): test/encoding/readable.sh: add non-whole type skip
- With this modification there will be possibility to skip only one/few
object/s of particular type, not whole type as ... - 10:21 AM Revision 9c34d0c4 (ceph): Merge pull request #6910 from dachary/wip-dev-guide-ceph-qa-suite
- doc: dev: introduction to tests
Reviewed-by: Nathan Cutler <ncutler@suse.com> - 09:13 AM Revision d0ffeb84 (ceph): osd/ReplicatedPG: when delete pg, don't forget free scrub resources.
- Signed-off-by: Jianpeng Ma <jianpeng.ma@intel.com>
- 08:34 AM Fix #14083 (New): degrade a log message to a default level
- When leader election was done successfully, it will ask peon to collect info. When failed, it will cause monitor to d...
- 07:57 AM Revision eb3b647b (ceph): check for striping parameters
- stripe_unit is same to object size or stripe_count is equal to 1,the remainder for it is misleading.
Signed-off-by: ... - 07:52 AM Revision 575d2380 (ceph): Merge pull request #6927 from tchaikov/wip-fix-ftbfs
- rgw: fix the build failure
Reviewed-by: Loic Dachary <ldachary@redhat.com> - 06:46 AM Revision 34a6c2ae (ceph): rgw: remove unused variable
- to fix the compiler warning of
rgw/rgw_formats.cc: In member function 'virtual void
RGWFormatter_Plain::dump_forma... - 06:17 AM Revision a57784f7 (ceph): rgw: fix the build failure
- * s/bucket_name_str/bucket_name/
* the member variable name of `req_state` was changed to `bucket_name`
in f7ca00a.... - 05:33 AM rgw Bug #14081 (Closed): The radosgw-agent cannot unlock, and error code is 404
- When the "rgw_num_rados_handles" is set as 16, the radosgw-agent cannot unlock, the error code is 404. the log is as...
- 05:21 AM rgw Bug #13817: RGW does not send HTTP header
- I noticed the same thing while diagnosing https://github.com/kahing/goofys/issues/46 . Seems like ceph doesn't send C...
- 04:14 AM Revision 265e18fc (ceph): Merge pull request #6859 from dillaman/wip-librbd-async-refresh
- librbd: image refresh code paths converted to async state machines
Reviewed-by: Josh Durgin <jdurgin@redhat.com> - 02:13 AM Revision d6c097dd (ceph): osd: fix wrong checking for uuid_d::parse() return
- The type of return value from uuid_d::parse() is bool, it never returns
a negative value.
Signed-off-by: Dunrong Hua... - 02:07 AM Revision fe1c28de (ceph): osd/ReplicatedPG: do not set omap digest if there is no omap
- We want to avoid encoding it if we can. And if the FLAG_OMAP is not set
we don't need to *also* store an empty crc.
... - 02:07 AM Revision 98bdb09b (ceph): osd/PGBackend: use mtime for digest decision if local_mtime is empty
- If we don't have a local_mtime value, use mtime instead, for the purposes
of deciding if we should record a digest af... - 02:07 AM Revision 1548a3fd (ceph): osd/ReplicatedPG: do not set local_mtime on non-tiered pool
- If a pool isn't tiered, don't bother with setting local_mtime. The only
users are the tiering agent (which isn't nee... - 02:07 AM Revision a9d3f078 (ceph): osd/osd_types: skip encoding newer object_info_t fields if they are unused
- This reduces the size of the encoded object_info_t in most cases,
enough to get us under the 255 byte limit for a sin... - 02:03 AM Revision dda6c67e (ceph): kv/RocksDBStore: log to ceph log
- Signed-off-by: Sage Weil <sage@redhat.com>
- 02:03 AM Revision 69842395 (ceph): kv/LevelDBStore: log leveldb to ceph log
- This greatly eases debugging, and avoids the unbounded growth of
the leveldb LOG file.
Signed-off-by: Sage Weil <sag... - 01:34 AM rbd Feature #13633 (Pending Backport): Verify self-managed snapshot functionality on image create
- 01:31 AM Revision 8aabb8cb (ceph): librbd: separate exclusive lock handling into async state machines
- Signed-off-by: Jason Dillaman <dillaman@redhat.com>
- 01:31 AM Revision 22eb25bc (ceph): tests: exclusive lock state machines test cases
- Signed-off-by: Jason Dillaman <dillaman@redhat.com>
- 01:31 AM Revision 6e1718c7 (ceph): librbd: separate image refresh into async state machines
- Signed-off-by: Jason Dillaman <dillaman@redhat.com>
- 01:31 AM Revision ccdeb151 (ceph): librbd: decouple ImageWatcher from exclusive lock management
- Use new ExclusiveLock state machine to handle all the proper
transitions between lock states.
Signed-off-by: Jason D... - 01:31 AM Revision 16733323 (ceph): librbd: object map lifespan now controlled by state machines
- Signed-off-by: Jason Dillaman <dillaman@redhat.com>
- 01:31 AM Revision 3fd24eb2 (ceph): librbd: rename ImageRefresh to ImageState
- This class handles more than just refreshing the image from
disk.
Signed-off-by: Jason Dillaman <dillaman@redhat.com> - 01:31 AM Revision c00b28b0 (ceph): librbd: implement image state tracker for open/close/refresh/etc
- Signed-off-by: Jason Dillaman <dillaman@redhat.com>
- 01:31 AM Revision 8740ddf0 (ceph): librbd: ImageWatcher should release lock upon request
- Signed-off-by: Jason Dillaman <dillaman@redhat.com>
- 01:31 AM Revision f1973b1d (ceph): librbd: integrate new async image state machines
- Signed-off-by: Jason Dillaman <dillaman@redhat.com>
- 01:31 AM Revision bad46415 (ceph): librbd: add shut down support to the AIO work queue
- Signed-off-by: Jason Dillaman <dillaman@redhat.com>
- 01:31 AM Revision 18f4f69f (ceph): librbd: flush all in-flight IO on block write and shut down
- Signed-off-by: Jason Dillaman <dillaman@redhat.com>
- 01:31 AM Revision 01bc5156 (ceph): WorkQueue: PointerWQ now supports out-of-band processing
- Signed-off-by: Jason Dillaman <dillaman@redhat.com>
- 01:31 AM Revision 8ef37b02 (ceph): librbd: IO ops will now asynchronously refresh the image if needed
- This avoids blocking the AIO path with the previous synchronous
refresh path.
Signed-off-by: Jason Dillaman <dillama... - 01:31 AM Revision 7e3a0322 (ceph): librbd: all object map methods are now async
- Signed-off-by: Jason Dillaman <dillaman@redhat.com>
- 01:31 AM Revision 2a4aad5f (ceph): tests: adjust test cases to refactored librbd
- Signed-off-by: Jason Dillaman <dillaman@redhat.com>
- 01:31 AM Revision 16547c9b (ceph): librbd: the journal state machine is now asynchronous
- Signed-off-by: Jason Dillaman <dillaman@redhat.com>
- 01:31 AM Revision db2d495c (ceph): librbd: refresh image asynchronously from watch/notify path
- The AIO path already ensures asynchronous image refreshes, but the
watch/notify path for maintenance ops was still us... - 01:31 AM Revision d2d4bb42 (ceph): librbd: refresh object map after rollback
- An image refresh will no longer trigger an automatic object map
reload, therefore explicitly reload the object map af... - 01:30 AM Revision e2eb27fb (ceph): librbd: fix missing header dependencies for parent_types.h
- Signed-off-by: Jason Dillaman <dillaman@redhat.com>
- 01:30 AM Revision fc4370ea (ceph): cls_rbd: async version of class helpers added
- New async versions of get_id, get_immutable_metadata,
get_mutable_metadata, get_flags, get_stripe_unit_count,
snapsho... - 01:30 AM Revision 8a4c1558 (ceph): librbd: support updated async cls_rbd API
- Signed-off-by: Jason Dillaman <dillaman@redhat.com>
- 01:30 AM Revision dd253af8 (ceph): librbd: convert object_map::InvalidateRequest to template
- Signed-off-by: Jason Dillaman <dillaman@redhat.com>
- 01:30 AM Revision 700a0def (ceph): librados_test_stub: additional mocked IoCtx methods
- Signed-off-by: Jason Dillaman <dillaman@redhat.com>
- 01:30 AM Revision 7248b93c (ceph): librbd: async object map lock/unlock/refresh state machines
- Creating async versions to support an async image refresh
Signed-off-by: Jason Dillaman <dillaman@redhat.com> - 01:30 AM Revision 19127742 (ceph): tests: mock test cases for new async object map lock/unlock/refresh ops
- Signed-off-by: Jason Dillaman <dillaman@redhat.com>
- 01:30 AM Revision 03b79abd (ceph): librados_test_stub: additional definitions and mock functions
- Signed-off-by: Jason Dillaman <dillaman@redhat.com>
- 01:30 AM Revision 3d25f514 (ceph): librbd: moved object name helpers to librbd::util namespace
- Signed-off-by: Jason Dillaman <dillaman@redhat.com>
- 01:30 AM Revision 8a33a897 (ceph): tests: restrict rebuild object map test case
- It should only run when the object map is enabled for the image.
Signed-off-by: Jason Dillaman <dillaman@redhat.com> - 01:30 AM Revision 3f3a0400 (ceph): librados_test_stub: implemented ObjectReadOperation::stat
- Signed-off-by: Jason Dillaman <dillaman@redhat.com>
- 01:30 AM Revision 06f2eb29 (ceph): tests: integrate cleaned up librbd AIO callback adapter
- Signed-off-by: Jason Dillaman <dillaman@redhat.com>
- 01:30 AM Revision c67725b7 (ceph): librbd: cleanup Context and librados AIO completion callbacks
- Signed-off-by: Jason Dillaman <dillaman@redhat.com>
- 01:30 AM Revision 004090a6 (ceph): librbd: cleanup librbd AIO completion callbacks
- Signed-off-by: Jason Dillaman <dillaman@redhat.com>
- 01:30 AM Bug #8885: SIGABRT in TrackedOp::dump() via dump_ops_in_flight()
- Seeing this on 0.94.3.3...
- 01:00 AM Revision f1c02a38 (ceph): Merge pull request #6643 from rzarzynski/wip-rgw-slo-rzarzynski
- - rgw: swift bulk delete (Radoslaw Zarzynski)
- rgw: static large objects (Radoslaw Zarzynski, Yehuda Sadeh)
Reviewe... - 12:01 AM Bug #14080 (Fix Under Review): ceph-disk: use blkid instead of sgdisk -i
- https://github.com/ceph/ceph/pull/6926
12/14/2015
- 11:48 PM Revision 9180a926 (ceph): 10.0.1
- 11:47 PM Revision 0a720017 (ceph): Merge pull request #6925 from dillaman/wip-13633
- librbd: optionally validate new RBD pools for snapshot support
Reviewed-by: Josh Durgin <jdurgin@redhat.com> - 11:23 PM Bug #14080: ceph-disk: use blkid instead of sgdisk -i
- Could be that it takes more than 120 seconds....
- 11:21 PM Bug #14080: ceph-disk: use blkid instead of sgdisk -i
- Nothing conclusive on dmesg. After a while (a few minutes at most), it comes back to normal.
- 11:18 PM Bug #14080 (Resolved): ceph-disk: use blkid instead of sgdisk -i
- On CentOS 7, on OpenStack (udevadm settle may take 45 sec or more). This blocks testing....
- 11:00 PM rbd Feature #13633 (Fix Under Review): Verify self-managed snapshot functionality on image create
- *master PR*: https://github.com/ceph/ceph/pull/6925
- 09:09 PM rbd Feature #13633 (In Progress): Verify self-managed snapshot functionality on image create
- 10:49 PM Revision 00cfe4ef (ceph): tests: new integration test for validating new RBD pools
- Signed-off-by: Jason Dillaman <dillaman@redhat.com>
- 10:46 PM Revision 42b23869 (ceph): msg/simple/Pipe.h:remove unneeded friend decl
- Removed unneeded friend declarations. Nested class already has access
to all the names(private, protected) to which e... - 10:42 PM rbd Documentation #14070: admin/build-doc fails to build because of -lrbd
- Odd, this worked before on gitbuilder-doc, which I had to upgrade to trusty to get a recent cython. I'll check what's...
- 10:41 PM Revision 1fea4dad (ceph): librbd: optionally validate RBD pool configuration (snapshot support)
- Fixes: #13633
Signed-off-by: Jason Dillaman <dillaman@redhat.com> - 10:39 PM rbd Bug #14059: Can't build Ceph with --with-rbd (Cython)
- Do you have everything that install-deps.sh would install? Cython 0.20.1 was the earliest version I tried on, but I d...
- 10:16 PM Revision b85775bb (ceph): Merge pull request #6917 from majianpeng/journal-iohint
- Journal iohint for librbd
Reviewed-by: Jason Dillaman <dillaman@redhat.com>
Reviewed-by: Josh Durgin <jdurgin@redhat... - 09:44 PM rbd Feature #14006 (In Progress): Track IO dependencies in journal
- 09:44 PM rbd Bug #14001 (Resolved): TestMockOperationSnapshotRemoveRequest.Success common/RWLock.h: 58: FAILED...
- 09:43 PM rbd Bug #14060 (Resolved): TestMockObjectMapResizeRequest.UpdateHeadOnDisk librbd/Journal.cc: 652: FA...
- 09:08 PM rbd Bug #14060 (Fix Under Review): TestMockObjectMapResizeRequest.UpdateHeadOnDisk librbd/Journal.cc:...
- *master PR*: https://github.com/ceph/ceph/pull/6923
- 04:46 PM rbd Bug #14060 (In Progress): TestMockObjectMapResizeRequest.UpdateHeadOnDisk librbd/Journal.cc: 652:...
- 08:14 AM rbd Bug #14060: TestMockObjectMapResizeRequest.UpdateHeadOnDisk librbd/Journal.cc: 652: FAILED assert...
- http://jenkins.ceph.dachary.org/job/ceph/LABELS=centos-7&&x86_64/10127/console...
- 07:54 AM rbd Bug #14060: TestMockObjectMapResizeRequest.UpdateHeadOnDisk librbd/Journal.cc: 652: FAILED assert...
- http://jenkins.ceph.dachary.org/job/ceph/LABELS=centos-7&&x86_64/10124/console...
- 09:41 PM Revision 13b80577 (ceph): Merge pull request #6923 from dillaman/wip-14060
- librbd: correct race conditions discovered during unit testing
Reviewed-by: Josh Durgin <jdurgin@redhat.com> - 08:58 PM Feature #11692: online change of mon_osd_full_ratio and mon_osd_nearfull_ratio doesn't take effect.
- Yeah, I don't envision any good way to make it possible to change these values by updating the runtime config.
Whe... - 07:40 AM Feature #11692: online change of mon_osd_full_ratio and mon_osd_nearfull_ratio doesn't take effect.
- after a second thought: before the new setting applies to the new osdmap, and be sent to OSD nodes. it needs to be ag...
- 07:37 AM Feature #11692: online change of mon_osd_full_ratio and mon_osd_nearfull_ratio doesn't take effect.
- 08:28 PM Revision 664903fb (ceph): journal: avoid race between flush request and in-flight flushes
- Previously, if the journal object was empty, a flush request could
complete instantly even if there is an in-flight '... - 08:27 PM Revision 3e6c9903 (ceph): librbd: fix test case race condition for maintenance requests
- Fixes: #14060
Signed-off-by: Jason Dillaman <dillaman@redhat.com> - 06:12 PM Revision 03f719cb (ceph): Merge pull request #6922 from liewegas/wip-cephfs-posix
- doc/cephfs/posix: update
- 05:49 PM Revision 16ba8261 (ceph): Merge pull request #6899 from branch-predictor/bp-fix-no-rbd-build
- rbd: fix build with "--without-rbd"
Reviewed-by: Josh Durgin <jdurgin@redhat.com> - 05:20 PM Revision 15c840c1 (ceph): rbd: fix build with "--without-rbd"
- It's not possible to build Ceph configured with "--without-rbd", as
tests and FileStore depends on components provide... - 04:29 PM Revision 3b085001 (ceph): common: eversion_t encoding optimization
- There is no padding between version and epoch, no matter 32-bit
or 64-bit machine.
Signed-off-by: Xinze Chi <xinze@x... - 04:29 PM Revision 8ad665d6 (ceph): osd: object_stat_sum_t encoding optimization
- Signed-off-by: Xinze Chi <xinze@xsky.com>
- 04:29 PM Revision a977e39f (ceph): common: utime_t encoding optimization
- there is no padding between tv_sec and tv_nsec, no matter 32-bit
or 64-bit machine.
Signed-off-by: Xinze Chi <xinze@... - 03:45 PM Revision d5cee59b (ceph): doc/cephfs/posix: update
- Signed-off-by: Sage Weil <sage@redhat.com>
- 02:54 PM Feature #13198 (Resolved): mon: include min_last_epoch_clean as part of PGMap::print_summary and ...
- 02:53 PM rgw Bug #12913 (Resolved): rgw: bucket listing hangs on versioned buckets
- 02:52 PM rgw Bug #13458 (Resolved): rgw: get bucket location returns region name, not region api name
- 02:52 PM Bug #13214 (Resolved): ceph upstart script rbdmap.conf incorrectly processes parameters
- 02:52 PM Bug #13208 (Resolved): small probability sigabrt when setting rados_osd_op_timeout
- 02:52 PM Bug #13116 (Resolved): osd: pg stuck in replay
- 02:52 PM Bug #13114 (Resolved): LibRadosWatchNotify.WatchNotify2Timeout
- 02:52 PM Bug #13112 (Resolved): should recalc the min_last_epoch_clean when decode PGMap
- 02:52 PM devops Fix #12994 (Resolved): rbd-replay* should ship in ceph-common
- 02:52 PM Bug #12990 (Resolved): osd: we do not ignore notify from down osds
- 02:52 PM rgw Bug #12958 (Resolved): rgw: orphan tool should be careful about removing head objects
- 02:52 PM rgw Fix #12939 (Resolved): Content-Type header should have correct initial capitals
- 02:52 PM devops Bug #12924 (Resolved): ceph.spec.in: rgw placeholder dirs are not packaged
- 02:52 PM Bug #12828 (Resolved): ReplicatedBackend: populate recovery_info.size for clone (bug symptom is s...
- 02:52 PM Bug #12748 (Resolved): bug with cache/tiering and snapshot reads [merged, needs a test in ceph-qa...
- 02:52 PM rgw Bug #12735 (Resolved): rgw: missing handling of encoding-type=url when listing keys in bucket
- 02:52 PM Bug #12565 (Resolved): osd: do not cache unused memory in attrs
- 02:52 PM CephFS Bug #12426 (Resolved): "ceph mds add_data_pool" check for EC pool is wrong
- 02:52 PM CephFS Bug #11504 (Resolved): CephFS restriction on removing cache tiers is overly strict
- 02:51 PM Backport #14077 (Resolved): hammer: scrub bogus results when missing a clone
- https://github.com/ceph/ceph/pull/7702
- 02:51 PM Backport #14076 (Resolved): hammer: ceph-objetstore-tool: add operations to remove clone ids from...
- https://github.com/ceph/ceph/pull/7702
- 02:46 PM Revision beacff5c (ceph): doc/rados/operations/crush-map: update docs on crush tunables
- Signed-off-by: Sage Weil <sage@redhat.com>
- 02:17 PM Revision 425c6448 (ceph): crushtool: add mapping test for jewel tunables
- Signed-off-by: Sage Weil <sage@redhat.com>
- 02:07 PM rbd Backport #13045 (Resolved): rbd export-diff crashes in librbd::simple_diff_cb
- 02:03 PM rbd Backport #13045 (In Progress): rbd export-diff crashes in librbd::simple_diff_cb
- 02:07 PM Backport #13210 (Resolved): tests : Fixed broken Makefiles after integration of lttng into rados
- 02:03 PM Backport #13210 (In Progress): tests : Fixed broken Makefiles after integration of lttng into rados
- 02:07 PM rbd Backport #13460 (Resolved): rbd-replay-prep: crashes on incorrect trace file
- 02:03 PM rbd Backport #13460 (In Progress): rbd-replay-prep: crashes on incorrect trace file
- 02:07 PM rbd Backport #13461 (Resolved): librbd: object map invalid but not flagged as such
- 02:03 PM rbd Backport #13461 (In Progress): librbd: object map invalid but not flagged as such
- 02:07 PM Backport #13637 (Resolved): FileStore: potential memory leak if getattrs fails.
- 02:03 PM Backport #13637 (In Progress): FileStore: potential memory leak if getattrs fails.
- 02:07 PM Backport #13654 (Resolved): crush: crash if we see CRUSH_ITEM_NONE in early rule step
- 02:03 PM Backport #13654 (In Progress): crush: crash if we see CRUSH_ITEM_NONE in early rule step
- 12:24 PM Bug #13916 (Need More Info): erasure code profile name
- 12:24 PM Bug #13916: erasure code profile name
- I tried the following and the problem does not show. It would be great if you could provide me with a sequence of com...
- 11:52 AM Revision 98e82463 (ceph): Merge pull request #6873 from javacruft/wip/debian-infernalis-patches
- ceph-detect-init: Ubuntu >= 15.04 uses systemd
Reviewed-by: Loic Dachary <ldachary@redhat.com> - 11:49 AM Revision d314362c (ceph): Merge pull request #6787 from oritwas/wip-run-cmake-check
- Wip run cmake check
Reviewed-by: Loic Dachary <ldachary@redhat.com> - 11:49 AM Revision c287fea0 (ceph): ajdust mstart and mstop script to run with cmake build
- Signed-off-by: Orit Wasserman <owasserm@redhat.com>
- 10:51 AM CephFS Bug #14048 (In Progress): MDS heap stats standby threads
- 10:28 AM CephFS Bug #14048: MDS heap stats standby threads
- Reproduced on master.
- 10:22 AM Bug #14075 (Resolved): OSD: wrongly placed assert during handle_pg_peering_evt.
- "assert(service.splitting(pgid))" will definitely fire if PG is currently unavailable and _try_resurrect_pg returns R...
- 10:14 AM Linux kernel client Bug #14071 (Rejected): Ceph FS not working as backstore for Linux SCSI target
- CephFS imposes a maximum file size. It's 1<<40 (aka one terabyte) by default. You can adjust it with "ceph mds set ...
- 09:27 AM Feature #13505: scrub/repair: persist scrub results.
- Draft implementation at https://github.com/ceph/ceph/pull/6898
- 08:56 AM Bug #14074 (Duplicate): OSD daemon can not start after cluster deployment
- I cloned the latest code and compile packaged into deb, then to the respective host DEB install a new package, versio...
- 08:47 AM Cleanup #13981 (Can't reproduce): configure script doesn't check bzip2-devel presence
- see https://github.com/ceph/ceph/pull/6885#issuecomment-164378681
- 08:32 AM Bug #14073 (Can't reproduce): osd: hammer: fail to start due to stray pglog object after firefly-...
- We have a large object storage cluster upgrade from firefly(0.80.10) to hammer(0.94.5). We found exists probability t...
- 07:52 AM Revision 2245a0cb (ceph): erasure-code/TestErasureCodeIsa.cc: add const
- Added const to k and m variable, so the later declared array
buffer::ptr enc[k + m] doesn't have to use variable-leng... - 07:28 AM Bug #14072 (Resolved): OSD: null pointer access when do handle_osd_map.
- We are risking of accessing null pointer of osdmap when OSD try to reconnect and encounters a network error.
- 05:32 AM Revision df0c1f5d (ceph): journal: add iohint flags for journal replayer.
- Signed-off-by: Jianpeng Ma <jianpeng.ma@intel.com>
- 05:32 AM Revision 84310c53 (ceph): journal: Add iohint flags for journal write ops.
- Signed-off-by: Jianpeng Ma <jianpeng.ma@intel.com>
- 05:11 AM Bug #14026 (Fix Under Review): OSD utilization is abnormal after data disk lost
- 04:15 AM Revision 0063ee24 (ceph): Merge pull request #6790 from kylinstorage/wip-subprocess
- SubProcess: fix multiple definition bug
Reviewed-by: Kefu Chai <kchai@redhat.com> - 04:13 AM Revision 3cbb824d (ceph): Merge pull request #5921 from caibo2014/add-tool-feature
- tools:support printing part cluster map in readable fashion
Reviewed-by: Kefu Chai <kchai@redhat.com>
12/13/2015
- 09:12 PM Linux kernel client Bug #14071 (Rejected): Ceph FS not working as backstore for Linux SCSI target
- I have been testing native Linux SCSI target (LIO) with the backstore file being a file on a Ceph File System, kernel...
- 09:02 PM Revision 5440334d (ceph): doc: dev: introduction to tests
- Signed-off-by: Loic Dachary <loic@dachary.org>
- 08:57 PM Revision e1b3b903 (ceph): Merge pull request #6906 from HeyoQiang/heyongqiang1212
- client: modify a word in log
Reviewed-by: Loic Dachary <ldachary@redhat.com> - 07:38 PM rbd Documentation #14070 (Resolved): admin/build-doc fails to build because of -lrbd
- On a virgin ubuntu 14.04 the hack to get the fake librbd does not appear to be working....
- 03:54 PM Revision 74c249d7 (ceph): common: improve cache efficiency
- if (weak_refs.count(key))
weak_refs[key]
it traverses the b-tree twice
replace it with weak_refs.find(key), which t... - 03:54 PM Revision c3a383c3 (ceph): common: improve cache, replace to unordered map with initializing hash ...
- Based on the PR https://github.com/ceph/ceph/pull/4441
Signed-off-by: Ning Yao <zay11022@gmail.com> - 01:51 PM Revision ef8fa8b9 (ceph): pybind: Implementation of Ioctx.set_read to allow read from snapshots
- Signed-off-by: Florent Manens <florent@beezim.fr>
- 12:47 PM Revision 09d7a60c (ceph): Merge pull request #6866 from xiaoxichen/umount_before_mount
- init-ceph: do umount when the path exists.
Reviewed-by: Loic Dachary <ldachary@redhat.com> - 12:45 PM rbd Bug #14060: TestMockObjectMapResizeRequest.UpdateHeadOnDisk librbd/Journal.cc: 652: FAILED assert...
- http://jenkins.ceph.dachary.org/job/ceph/LABELS=centos-7&&x86_64/10114/console...
- 11:44 AM rbd Bug #14064 (Duplicate): TestMockObjectMapSnapshotCreateRequest.ReadMapError librbd/Journal.cc: 71...
- marking as duplicate. There is another false negative today in the same area but with a different message. I'll clust...
- 08:34 AM rbd Bug #14030: rbd merge-diff doesn't properly handle >2GB diffs
- Thanks Josh. I have assigned http://tracker.ceph.com/issues/14062 to you for now. I would look out for rbd backports ...
- 08:17 AM Revision 09e21b49 (ceph): init-ceph: do umount when the path exists.
- If the specified mount point is in use, umount it instead
of skipping mounting the fs.
In previous code , if we forg... - 04:54 AM Bug #14066 (Resolved): mount/mtab.cc: memory leaks
- since the mem leak happens in the mount.ceph CLI tool, the tool is not a long-running command. and the leak only happ...
- 04:52 AM Revision d7a17900 (ceph): Merge pull request #6905 from H3C/wip-bugfix-mount1
- mount.ceph: memory leaks
Reviewed-by: Kefu Chai <kchai@redhat.com> - 01:10 AM Revision cd4b4cf4 (ceph): Merge pull request #6908 from nilamdyuti/wip-doc-rbdsnapshot
- doc: Modified a note section in rbd-snapshot doc.
Reviewed-by: Loic Dachary <ldachary@redhat.com> - 01:03 AM Revision 14e9d292 (ceph): mount/mtab.cc: memory leaks
- the free() should be called to free resources, in order to avoid memory leaks
Fixes:#14066
Signed-off-by: Qiankun Zh...
12/12/2015
- 09:49 PM rbd Bug #14030: rbd merge-diff doesn't properly handle >2GB diffs
- No other backports are needed for this - the same change should just be applied to src/rbd.cc in infernalis and hamme...
- 01:30 PM rbd Bug #14030: rbd merge-diff doesn't properly handle >2GB diffs
- src/tools/rbd/action/MergeDiff.cc doesn't exist for hammer and infernalis. This may require backporting fa4e00f8c8560...
- 09:47 PM rbd Bug #13924 (Resolved): TestJournalPlayer.PrefetchSkippedObject test/journal/test_JournalPlayer.cc...
- 08:22 PM Revision 58bad43e (ceph): client: modify a word in log
- This modification changed the word "comit" in line 7704
/* ldout(cct, 10) << "sync_write_comit -- no more unsafe writ... - 07:53 PM Revision c36d73ee (ceph): doc: Modified a note section in rbd-snapshot doc.
- Elaborates the note section a bit more as to why stopping I/O before
taking a snapshot is recommended and how to do i... - 07:25 PM rbd Bug #14059: Can't build Ceph with --with-rbd (Cython)
- I'll investigate this further, currently I'm on CentOS 6 and even installing Cython 0.24 doesn't help. I hope I'll be...
- 05:56 PM Bug #13990: Hammer (0.94.3) OSD does not delete old OSD Maps in a timely fashion (maybe at all?)
- I have a log file from an OSD in this state, its 270MB, how would you recommend I get this file into the hands of som...
- 04:02 PM Bug #14066 (Fix Under Review): mount/mtab.cc: memory leaks
- 12:09 PM Bug #14066: mount/mtab.cc: memory leaks
- qiankun zheng wrote:
> the free() should be called to free resources, in order to avoid memory leaks
https://gith... - 11:08 AM Bug #14066 (Resolved): mount/mtab.cc: memory leaks
- the free() should be called to free resources, in order to avoid memory leaks
- 01:56 PM Revision cd0f2876 (ceph): Merge pull request #766 from dachary/wip-buildpackages-log-hammer
- Wip buildpackages log hammer
- 01:53 PM Revision 3c548b0e (ceph): buildpackages: no need to sudo in bootcmd
- Signed-off-by: Loic Dachary <loic@dachary.org>
(cherry picked from commit 64d8a586f72c71834d71a8147ae12bf395d5319a) - 01:53 PM Revision 0e6a9a6f (ceph): buildpackages: log the output of make
- Signed-off-by: Loic Dachary <loic@dachary.org>
(cherry picked from commit 0c1e7f1f6b6137d573984305d4f7519e97a5cdd6) - 01:53 PM Revision a66367b8 (ceph): buildpackages: pkgrepo host also needs user-data setup.
- The packages repo host fails in environments where networking setup is
needed in VMs. Use the same user-data as the b... - 01:53 PM Revision 3e843374 (ceph): buildpackages: copy teuthology user-data.txt changes
- 7b27e1db7: openstack: support /etc/network/intefaces injection
2358562cf: ensure VMs always have /etc/hosts set up
43... - 01:53 PM Revision 5862207b (ceph): buildpackages: catch instance creation failure.
- Split the sleep from the server creation, so we catch 'server create'
failures (eg due to quota):
> Quota exceeded fo... - 01:52 PM Revision dd8388c5 (ceph): buildpackages: Be robust about addresses
- OpenStack could tell us the VM has multiple networks, and offers no
guarantee about the order of addresses either (th... - 01:51 PM CephFS Backport #14067 (Resolved): infernalis : Ceph file system is not freeing space
- https://github.com/ceph/ceph/pull/7431
- 01:51 PM CephFS Bug #13777 (Pending Backport): Ceph file system is not freeing space
- 01:43 PM Revision ac456019 (ceph): Merge pull request #764 from dachary/wip-buildpackages-log
- buildpackages: log the output of make
- 01:42 PM Backport #13929 (In Progress): infernalis: Ceph Pools' MAX AVAIL is 0 if some OSDs' weight is 0
- 01:39 PM Revision b211686d (ceph): mon/PGMonitor: MAX AVAIL is 0 if some OSDs' weight is 0
- In get_rule_avail(), even p->second is 0, it's possible to be used
as divisor and quotient is infinity, then is conve... - 01:27 PM Revision 64d8a586 (ceph): buildpackages: no need to sudo in bootcmd
- Signed-off-by: Loic Dachary <loic@dachary.org>
- 01:03 PM Revision 0c1e7f1f (ceph): buildpackages: log the output of make
- Signed-off-by: Loic Dachary <loic@dachary.org>
12/11/2015
- 10:26 PM Feature #12740 (Pending Backport): ceph-objetstore-tool: add operations to remove clone ids from ...
- 10:24 PM Bug #12738 (Pending Backport): scrub bogus results when missing a clone
- 10:22 PM Feature #13301: Request for method to track client versions which have connected to Ceph cluster
- Well, we could probably log each time a new client authenticates with the mon?
- 10:10 PM Feature #14065 (Rejected): mon commands should allow buckets to be created using straw1 optionally
- Currently, you can only do this by manually editing the crush map.
- 09:53 PM rbd Bug #14059: Can't build Ceph with --with-rbd (Cython)
- Thanks for looking into this Piotr. Sorry for the trouble - looks like we missed adding a lower bound on cython versi...
- 07:17 PM rbd Bug #14059: Can't build Ceph with --with-rbd (Cython)
- Also doesn't work right away on CentOS 7 with Cython 0.19:
@
creating /ceph/src/build
creating /ceph/src/build/t... - 12:50 PM rbd Bug #14059 (Resolved): Can't build Ceph with --with-rbd (Cython)
- When using Cython version 0.14.1 and
@CXXFLAGS="-g -ggdb" ./configure \ ... - 09:50 PM rbd Bug #14064 (Duplicate): TestMockObjectMapSnapshotCreateRequest.ReadMapError librbd/Journal.cc: 71...
- http://jenkins.ceph.dachary.org/job/ceph/LABELS=centos-7&&x86_64/10100/console...
- 09:46 PM Revision 7f627e04 (ceph): Merge pull request #6901 from yehudasa/wip-mstart
- vstart.sh: add mstart, mstop, mrun wrappers for running multiple vstart-style test clusters out of src tree
Reviewed... - 09:37 PM rbd Backport #14063 (Resolved): hammer: rbd merge-diff doesn't properly handle >2GB diffs
- https://github.com/ceph/ceph/pull/6980
- 09:37 PM rbd Backport #14062 (Resolved): infernalis: rbd merge-diff doesn't properly handle >2GB diffs
- https://github.com/ceph/ceph/pull/6981
- 08:46 PM Bug #13993: Use of syslog results in all log messages at priority "emerg"
- *jewel PR*: https://github.com/ceph/ceph/pull/6815
- 08:42 PM Revision b3ad2b93 (ceph): Merge pull request #6698 from mathslinux/auth-crash-fix
- auth: fix crash when bad keyring is passed
Reviewed-by: Sage Weil <sage@redhat.com> - 07:55 PM Bug #13986: TEST_scrub_snaps: timeout > 300
- http://jenkins.ceph.dachary.org/job/ceph/LABELS=centos-7&&x86_64/10094/console
- 05:47 PM Bug #13986: TEST_scrub_snaps: timeout > 300
- http://jenkins.ceph.dachary.org/job/ceph/LABELS=centos-7&&x86_64/10092/
- 05:26 PM Bug #13986: TEST_scrub_snaps: timeout > 300
- http://jenkins.ceph.dachary.org/job/ceph/LABELS=centos-7&&x86_64/10086/console
- 05:26 PM Bug #13986 (Resolved): TEST_scrub_snaps: timeout > 300
- 06:29 PM Revision d4d37876 (ceph): Suites/jewel in the smithi lab will be on centos 7.1
- Signed-off-by: yuriw <yuri.weinstein@gmail.com>
- 05:36 PM Revision f07e0292 (ceph): assert: abort() rather than throw
- Assertion failure should be big. It should be catastrophic. It should
make an impression!
It should also not throw e... - 05:24 PM Revision e5536b27 (ceph): Merge pull request #6890 from dachary/wip-13986-test-pg-scrub
- tests: fix failure for osd-scrub-snap.sh
Reviewed-by: Xinze Chi <xinze@xsky.com>
Reviewed-by: Ning Yao <zay11022@gma... - 05:01 PM Bug #13988: new OSD re-using old OSD id fails to boot
- 05:00 PM Bug #13988: new OSD re-using old OSD id fails to boot
- https://github.com/ceph/ceph/pull/6900
- 03:43 PM Bug #13988: new OSD re-using old OSD id fails to boot
- Before removal:...
- 11:36 AM Bug #13988: new OSD re-using old OSD id fails to boot
- Hi Loic,
I also fail on the master branch with resue-osd-id test cases.
(but not always fail on infernalis)
Ma... - 10:07 AM Bug #13988: new OSD re-using old OSD id fails to boot
- Loic Dachary wrote:
> @Vicente : I do not see an assertion in http://jenkins.ceph.dachary.org/job/ceph/LABELS=centos... - 09:36 AM Bug #13988: new OSD re-using old OSD id fails to boot
- ...
- 08:31 AM Bug #13988: new OSD re-using old OSD id fails to boot
- The bad commit is one of those:...
- 07:57 AM Bug #13988: new OSD re-using old OSD id fails to boot
- @Vicente : I do not see an assertion in http://jenkins.ceph.dachary.org/job/ceph/LABELS=centos-7&&x86_64/10060/consol...
- 07:42 AM Bug #13988: new OSD re-using old OSD id fails to boot
- ...
- 04:50 PM Revision 4e28f9e6 (ceph): osd/OSDMap: clear osd_info, osd_xinfo on osd deletion
- If we destroy an OSD in the map, clear not just the uuid but also
all the metadata about it.
Specifically, we care a... - 04:32 PM Revision 6c4a4219 (ceph): Merge pull request #6709 from xiaoxichen/failure_by_host
- mon: support min_down_reporter by subtree level (default by host)
Reviewed-by: Kefu Chai <kchai@redhat.com>
Reviewed... - 04:30 PM Revision ad15b813 (ceph): vstart.sh: mon osd reporter subtree level = osd
- Since everything is on one host!
Signed-off-by: Sage Weil <sage@redhat.com> - 04:20 PM Revision 8cc6301f (ceph): mstop.sh, mrgw.sh: more scripts to control daemons in dev env
- Signed-off-by: Yehuda Sadeh <yehuda@redhat.com>
- 04:20 PM Revision df92e267 (ceph): mrun: a tool to run commands against specific mstart cluster
- E.g.,
$ ./mstart.sh foo -n
$ ./mrun foo ceph -s
Signed-off-by: Yehuda Sadeh <yehuda@redhat.com> - 04:20 PM Revision dfa403c7 (ceph): mstart.sh: run multiple clusters in a single source tree
- Signed-off-by: Yehuda Sadeh <yehuda@redhat.com>
- 04:14 PM rbd Bug #14058 (Fix Under Review): Can't build Ceph with --without-rbd and --with-debug
- Patch which fixes this: https://github.com/ceph/ceph/pull/6899
- 12:47 PM rbd Bug #14058 (Resolved): Can't build Ceph with --without-rbd and --with-debug
- When using
@CXXFLAGS="-g -ggdb" ./configure \
--without-fuse... - 04:06 PM rbd Bug #14060: TestMockObjectMapResizeRequest.UpdateHeadOnDisk librbd/Journal.cc: 652: FAILED assert...
- I'll append links to failures when I see more.
- 04:05 PM rbd Bug #14060 (Resolved): TestMockObjectMapResizeRequest.UpdateHeadOnDisk librbd/Journal.cc: 652: FA...
- First seen at https://github.com/ceph/ceph/pull/6892 which was merged without a clean run from the bot.
http://jen... - 03:32 PM Revision 61392c1c (ceph): Merge pull request #6823 from oritwas/wip-cmake-fix-authtool
- cmake: add TracepointProvider.cc to libcommon
Reviewed-by: Kefu Chai <kchai@redhat.com> - 03:28 PM Revision dc27e43b (ceph): Merge pull request #6827 from trociny/wip-scrub-debug
- Little improvements for troubleshooting scrub issues
Reviewed-by: Kefu Chai <kchai@redhat.com> - 03:24 PM Revision df56dc9e (ceph): Merge pull request #6732 from aalba6675/patch-custom-cluster-name
- sysvinit: allow custom cluster names
Reviewed-by: Sage Weil <sage@redhat.com> - 03:22 PM Revision 8b3409e5 (ceph): Merge pull request #6730 from kylinstorage/fix-refcounter-in-replica_scrub
- osd: fix MOSDRepScrub reference counter in replica_scrub
Reviewed-by: Kefu Chai <kchai@redhat.com> - 03:21 PM Revision 97dbb7c7 (ceph): Merge pull request #6713 from zhanghuan/fix-wbthrottle
- osd: fix incorrect throttle in WBThrottle
Reviewed-by: Kefu Chai <kchai@redhat.com> - 03:20 PM Revision b6640540 (ceph): Merge pull request #6731 from kylinstorage/clean_up_objecter
- librados: clean up Objecter.h
Reviewed-by: Kefu Chai <kchai@redhat.com> - 02:58 PM Bug #13989: OSD boot fails with os/FileJournal.cc: 1907: FAILED assert(0)
- @vicente, sorry I was confused with 13988. I'm not sure how to reproduce this assert anymore.
- 02:56 PM Revision a75a9a81 (ceph): FileJournal: batch pop and unpop from writeq and completions
- Signed-off-by: Xinze Chi <xinze@xsky.com>
- 02:14 PM Linux kernel client Bug #14022: map_sem for read + request_mutex are held indefinitely
- Yeah, what you attached is pretty useless. Your kernel is compiled w/o lockdep, so I need to see *all* the stacktrac...
- 01:42 PM Linux kernel client Bug #14022: map_sem for read + request_mutex are held indefinitely
- Can't help you with the syslog, because the boot messages are already rotated away.
I could reboot the system, and... - 01:17 PM Linux kernel client Bug #14022: map_sem for read + request_mutex are held indefinitely
- My dmesg buffer seems to be to small, i home this output from journalctl is helpfull, but there are missed messages i...
- 01:35 PM Revision a2c22368 (ceph): Merge pull request #6863 from jdurgin/wip-buffer-header
- buffer: make usable outside of ceph source again
Reviewed-by: Sage Weil <sage@redhat.com> - 01:30 PM rbd Bug #14001: TestMockOperationSnapshotRemoveRequest.Success common/RWLock.h: 58: FAILED assert(!is...
- http://jenkins.ceph.dachary.org/job/ceph/LABELS=centos-7&&x86_64/10082/console
(the pull request from which it origi... - 01:24 PM CephFS Bug #14048: MDS heap stats standby threads
- I think this threads should be cleaned up
- 01:01 PM Revision 61aa107f (ceph): cmake: add run_cmake-check.sh
- currently only check to cmake build process
Signed-off-by: Orit Wasserman <owasserm@redhat.com> - 11:36 AM Revision a13fe7cc (ceph): Merge pull request #762 from dachary/wip-13624-workunit-git-hammer
- workunit: fetch from --ceph-git-url
- 11:34 AM Revision f0e6ff7c (ceph): workunit: allow tag to override sha1
- The sha1 for the workunit task is always set by the suite.py task. The
tag must be checked before the sha1 othewise i... - 11:34 AM Revision 0e084449 (ceph): workunit: keep the path to the list of tests in a variable
- Signed-off-by: Loic Dachary <loic@dachary.org>
(cherry picked from commit 28dc5c0c3347b41bae24a4fa7db491002c4fe2ec) - 11:34 AM Revision c33d2247 (ceph): workunit: fetch from --ceph-git-url
- The commit from which workunits are fetched must be retrieved
from --ceph-git-url via teuth_config.get_ceph_git_url()... - 11:27 AM Revision f19a519b (ceph): Merge pull request #756 from dachary/wip-13624-workunit-git
- workunits: fetch from --ceph-git-url
- 09:47 AM Revision 85f77419 (ceph): Merge pull request #6750 from theanalyst/doc/mailmap/jewel
- mailmap: Jewel updates
Reviewed-by: Loic Dachary <ldachary@redhat.com> - 09:35 AM rbd Bug #14047 (Fix Under Review): rbd: rbd order will be place in 22, when set to 0 in the config_opt
- 02:24 AM rbd Bug #14047: rbd: rbd order will be place in 22, when set to 0 in the config_opt
- https://github.com/ceph/ceph/pull/6886
- 09:06 AM Bug #14056 (Duplicate): include/interval_set.h: 338:FAILED assert(len > 0)
- Environment:the filestore choose Newstore.
In function NewStore::fiemap ,variable offset may be equal to variable st... - 08:47 AM Revision 54a6ba8c (ceph): rados: implement radosstripper force remove when full.
- extend libradosstripper remove interface and rados tool
rm force-full when use_stripper.
Signed-... - 08:46 AM Revision f516f5ae (ceph): Merge pull request #6872 from dachary/wip-osd-typo
- doc: osd: s/schedued/scheduled/
Reviewed-by: Abhishek Lekshmanan <abhishek.lekshmanan@gmail.com> - 08:26 AM Revision db525ef8 (ceph): rados: implement rm --force option to force remove when full.
- librados extend remove interface, add flags parameter, and use
this extended interface to implement force... - 08:05 AM Bug #14055 (Resolved): FAILED assert(len > 0)
- Environment:the filestore choose Newstore.
In function NewStore::fiemap ,variable offset may be equal to variable s... - 07:42 AM Revision b3c28b94 (ceph): mailmap: Nishtha Rai affiliation
- Signed-off-by: Abhishek Lekshmanan <abhishek.lekshmanan@gmail.com>
- 07:42 AM Revision 9f347379 (ceph): mailmap: YankunLi affiliation
- Signed-off-by: Abhishek Lekshmanan <abhishek.lekshmanan@gmail.com>
- 07:41 AM Revision 0828cb43 (ceph): mailmap: John Coyle affiliation
- Signed-off-by: Abhishek Lekshmanan <abhishek.lekshmanan@gmail.com>
Conflicts:
.organizationmap
Trivial merge confli... - 07:40 AM Revision 0d3605f4 (ceph): mailmap: Jashan Kamboj affiliation
- Signed-off-by: Abhishek Lekshmanan <abhishek.lekshmanan@gmail.com>
- 07:40 AM Revision ea8ad14e (ceph): mailmap: Sangdi Xu name normalization
- Signed-off-by: Abhishek Lekshmanan <abhishek.lekshmanan@gmail.com>
- 07:40 AM Revision 5ed7d09f (ceph): mailmap: Wei Luo affiliation
- Signed-off-by: Abhishek Lekshmanan <abhishek.lekshmanan@gmail.com>
- 07:40 AM Revision 59916443 (ceph): mailmap: Wenjun Huang affiliation
- Signed-off-by: Abhishek Lekshmanan <abhishek.lekshmanan@gmail.com>
- 07:40 AM Revision e2cd81d7 (ceph): mailmap: You Ji affiliation
- Signed-off-by: Abhishek Lekshmanan <abhishek.lekshmanan@gmail.com>
- 07:40 AM Revision 458c59a6 (ceph): mailmap: Zeqiang Zhuang affiliation
- Signed-off-by: Abhishek Lekshmanan <abhishek.lekshmanan@gmail.com>
- 07:40 AM Revision 41a18ffd (ceph): mailmap: runsisi affiliation
- Signed-off-by: Abhishek Lekshmanan <abhishek.lekshmanan@gmail.com>
- 07:40 AM Revision c65bab18 (ceph): mailmap: Jacek J. Lakis affiliation
- Signed-off-by: Abhishek Lekshmanan <abhishek.lekshmanan@gmail.com>
- 07:40 AM Revision e108b22d (ceph): mailmap: Adam Kupczyk affiliation
- Signed-off-by: Abhishek Lekshmanan <abhishek.lekshmanan@gmail.com>
- 07:40 AM Revision fee41ad2 (ceph): mailmap: Guang Yang name normalization
- Signed-off-by: Abhishek Lekshmanan <abhishek.lekshmanan@gmail.com>
- 07:40 AM Revision e9aa1328 (ceph): mailmap: Robin H. Johnson name normalization
- Signed-off-by: Abhishek Lekshmanan <abhishek.lekshmanan@gmail.com>
- 07:04 AM Revision a5ce9c29 (ceph): Merge pull request #758 from dreamhost/wip-openstack-dhc
- Fix runs on OpenStack clusters like DreamHost Compute
Reviewed-by: Loic Dachary <ldachary@redhat.com> - 06:58 AM Revision ebf8508c (ceph): buildpackages: pkgrepo host also needs user-data setup.
- The packages repo host fails in environments where networking setup is
needed in VMs. Use the same user-data as the b... - 06:58 AM Revision 89bde0b8 (ceph): buildpackages: copy teuthology user-data.txt changes
- 7b27e1db7: openstack: support /etc/network/intefaces injection
2358562cf: ensure VMs always have /etc/hosts set up
43... - 06:12 AM Bug #14052 (Resolved): Client io blocked when removing snapshot
- I used a rbd command to create a 6TB-size image, And then created a snapshot of this image. After that, I kept writin...
- 02:29 AM Bug #13961 (Resolved): KeyValueStore: ceph_objectstore_bench crashed
- 02:29 AM Revision 1c24b207 (ceph): Merge pull request #6783 from yuyuyu101/wip-13961
- osd: KeyValueStore: don't queue NULL context
Reviewed-by: Sage Weil <sage@redhat.com> - 02:16 AM Revision 58907605 (ceph): test: add test case for c++ osd/pg command
- Signed-off-by: Yunchuan Wen <yunchuan.wen@kylin-cloud.com>
- 02:16 AM CephFS Bug #14029: "[ FAILED ] LibCephFS.OpenLayout" in smoke-master-distro-basic-openstack
- ovh: teuthology-2015-12-07_23:04:02-fs-master---basic-openstack/32792/
Hrm. - 02:15 AM Revision 5f7f8061 (ceph): librados: add osd/pg command interface for c++
- Signed-off-by: Yunchuan Wen <yunchuan.wen@kylin-cloud.com>
- 01:40 AM CephFS Bug #13903: Failure in TestStrays.test_ops_throttle
- http://teuthology.ovh.sepia.ceph.com/teuthology/teuthology-2015-12-08_23:04:01-fs-infernalis---basic-openstack/34531/
- 01:24 AM Revision b2ec53c8 (ceph): Merge pull request #6892 from kylinstorage/wip-fix-rbd-tracepoint
- librbd: fix tracepoint parameter in diff_iterate
Reviewed-by: Josh Durgin <jdurgin@redhat.com>
12/10/2015
- 11:36 PM CephFS Bug #14048: MDS heap stats standby threads
- Huh. The tell command is setting up a new client connection — are we failing to clean these up for some reason?
- 01:00 PM CephFS Bug #14048 (Resolved): MDS heap stats standby threads
- Hello I am have a problem with heap stats (dump, start_profiler, stop_profiler) command:
When I run it - i get som... - 11:32 PM Messengers Bug #12805: msg/simple/SimpleMessenger.cc: 238: FAILED assert(!cleared)
- Do you have some logs or something for this, David?
- 11:28 PM CephFS Bug #14029: "[ FAILED ] LibCephFS.OpenLayout" in smoke-master-distro-basic-openstack
- Hopefully we just got a mismatch between qa suite and ceph with regards to the new pool layout cephx flag? :/
- 10:22 PM devops Bug #11104: dependency problem with Ceph "Firefly" upstream and EPEL
- https://github.com/ceph/ceph/pull/6773 was merged last week. I think that means this is finally resolved.
- 10:22 PM devops Bug #13874: Missing `make` on openstack-provisioned nodes
- > @boris @ken - when can we have the permanent fix?
That would be https://github.com/ceph/ceph/pull/6773 - merged ... - 08:12 PM rgw Bug #11419: DragonDisk fails to create directories via S3: MissingContentLength
- Robin Johnson wrote:
> Max:
> can I get your nginx config and details to check? I tested on lighttpd & apache.
>... - 08:09 PM rgw Bug #11419: DragonDisk fails to create directories via S3: MissingContentLength
- Whenever I try to PUT a new file (tested DragonDisk and the officiall S3 SDK) I'm getting back HTTP/411 with error me...
- 08:06 PM rgw Bug #11419: DragonDisk fails to create directories via S3: MissingContentLength
- Max Grobecker wrote:
> I'm facing the same problem on version 0.94.5 too by using RGW with nginx/fastcgi.
> Should ... - 08:03 PM rgw Bug #11419: DragonDisk fails to create directories via S3: MissingContentLength
- The three commits from https://github.com/ceph/ceph/pull/5118 were released in 0.94.3:...
- 05:36 PM rgw Bug #11419: DragonDisk fails to create directories via S3: MissingContentLength
- I'm facing the same problem on version 0.94.5 too by using RGW with nginx/fastcgi.
Should this fix have been backpor... - 07:58 PM Revision 6955d294 (ceph): Merge pull request #6878 from manens/pybind_rados_ioctx_snap_rollback
- pybind: Implementation of rados_ioctx_snapshot_rollback
Reviewed-by: Josh Durgin <jdurgin@redhat.com> - 07:55 PM Revision 3c695464 (ceph): buildpackages: catch instance creation failure.
- Split the sleep from the server creation, so we catch 'server create'
failures (eg due to quota):
> Quota exceeded fo... - 07:55 PM Revision 0240f576 (ceph): buildpackages: Be robust about addresses
- OpenStack could tell us the VM has multiple networks, and offers no
guarantee about the order of addresses either (th... - 07:52 PM Messengers Bug #13985: Messenger: large message data content length cause int overflow
- Oh. That would happen if you create extremely large directories without enabling the directory fragmentation. It's di...
- 07:46 PM Revision c6cdc334 (ceph): tests: verify it is possible to reuse an OSD id
- When an OSD id is removed via ceph osd rm, it will be reused by the next
ceph osd create command. Verify that and OSD... - 07:20 PM rbd Bug #14030 (Pending Backport): rbd merge-diff doesn't properly handle >2GB diffs
- https://github.com/ceph/ceph/pull/6889
commit:68125dd01349edf93cfa1af5028c2d438b5ae089 - 07:18 PM Revision 134c988e (ceph): Merge pull request #6889 from kylinstorage/wip-fix-merge-diff
- librbd: fix merge-diff for >2GB diff-files
Reviewed-by: Josh Durgin <jdurgin@redhat.com> - 05:50 PM Revision 019a93f6 (ceph): Merge pull request #6888 from prallabh/master
- rgw: fix a glaring syntax error
Reviewed-by: Yehuda Sadeh <yehuda@redhat.com> - 04:34 PM Revision 5876829d (ceph): tests: kill_daemons uses TERM instead of KILL
- When called to teardown a test, kill_daemon should use KILL to ensure
all leftovers are removed as quickly as possibl... - 04:34 PM Revision e2f163a5 (ceph): tests: fix failure for osd-scrub-snap.sh
- Reason to get a false negative only one osd, when calling kill_daemons
$dir KILL osd, which means kill -9 shutdown() ... - 04:05 PM Bug #13350 (New): "Error cancelling global ust listener thread" errors
- Here is one more:
http://teuthology.ovh.sepia.ceph.com/teuthology/teuthology-2015-12-08_17:02:01-upgrade:hammer-x-in... - 03:46 PM Bug #13986: TEST_scrub_snaps: timeout > 300
- https://github.com/ceph/ceph/pull/6890
- 01:04 PM Bug #13986: TEST_scrub_snaps: timeout > 300
- See https://github.com/ceph/ceph/pull/6884 for an explanation of false negatives.
- 12:34 PM Bug #13986: TEST_scrub_snaps: timeout > 300
- @david do you have an opinion on the error log above ? It's the first time it happens.
- 12:32 PM Bug #13986: TEST_scrub_snaps: timeout > 300
- http://jenkins.ceph.dachary.org/job/ceph/LABELS=centos-7&&x86_64/10048/console
it has a new kind of error log. The... - 11:44 AM Bug #13986: TEST_scrub_snaps: timeout > 300
- http://jenkins.ceph.dachary.org/job/ceph/LABELS=centos-7&&x86_64/10041/console
- 03:10 PM Revision 68125dd0 (ceph): librbd: fix merge-diff for >2GB diff-files
- from http://tracker.ceph.com/issues/14030
Fixes: #14030
Signed-off-by: Yunchuan Wen <yunchuan.wen@kylin-cloud.com> - 12:18 PM Revision 213cd780 (ceph): rgw: fix a glaring syntax error
- syntactical error is causing rgw not to start
Signed-off-by: Pavan Rallabhandi <pavan.rallabhandi@sandisk.com> - 12:00 PM Bug #13988: new OSD re-using old OSD id fails to boot
- Hi loic,
I try to trace this problem.
I think the problem is on FileJournal::check()
Due to the same uuid, it... - 11:09 AM Revision 2b50a753 (ceph): Implementation of rados_ioctx_snapshot_rollback
- Signed-off-by: Florent Manens <florent@beezim.fr>
- 10:42 AM rbd Bug #14047: rbd: rbd order will be place in 22, when set to 0 in the config_opt
- Reason:
1.If set in the config_opt rbd_default_order = 0 (not in the range [12, 25]), due to reset the order = 22,... - 08:59 AM rbd Bug #14047 (Resolved): rbd: rbd order will be place in 22, when set to 0 in the config_opt
- rbd order will be place in 22, when set to 0 in the config_opt...
- 10:00 AM Cleanup #13981 (Fix Under Review): configure script doesn't check bzip2-devel presence
- Needs review: https://github.com/ceph/ceph/pull/6885
- 08:51 AM Revision 4671c02a (ceph): Fix end of newlines
- 08:30 AM Revision 02174393 (ceph): Merge pull request #6858 from jcsp/wip-cephfs-auth-doc
- doc: note that cephfs auth stuff is new in jewel
- 07:46 AM rgw Bug #14046 (Can't reproduce): When I get AWS JS API to create buckets, using S3 Browser delete bu...
- When I get AWS JS API to create buckets, using S3 Browser delete buckets, radosgw crashed
2015-12-10 14:36:22.5487... - 01:11 AM rbd Tasks #14045 (Closed): add valgrind to python binding tests
- Especially now that the rbd python bindings are in cython, we should check for memory leaks in the python bindings.
- 01:08 AM rbd Bug #13115 (Resolved): librbd python bindings eat exceptions in callbacks
12/09/2015
- 11:45 PM Revision 701a6626 (ceph): Merge pull request #6768 from marcan/cython
- pybind: port the rbd bindings to Cython
Reviewed-by: Josh Durgin <jdurgin@redhat.com>
Reviewed-by: Nathan Cutler <nc... - 11:11 PM Revision ad47b418 (ceph): CodingStyle: Fix coding-style on PG related Monitor files
- Using uncrustify the coding style of the files has been changed
to match what has been described in the CodingStyle d... - 10:51 PM Backport #14044 (In Progress): infernalis: ceph-disk list fails on /dev/cciss!c0d0
- 10:44 PM Backport #14044 (Resolved): infernalis: ceph-disk list fails on /dev/cciss!c0d0
- https://github.com/ceph/ceph/pull/6880
- 10:48 PM Bug #13970 (Fix Under Review): ceph-disk list fails on /dev/cciss!c0d0
- 10:43 PM Bug #13970 (Pending Backport): ceph-disk list fails on /dev/cciss!c0d0
- 10:40 PM Bug #13970 (Fix Under Review): ceph-disk list fails on /dev/cciss!c0d0
- https://github.com/ceph/ceph/pull/6879
- 10:44 PM Backport #14043 (Resolved): hammer: osd/PG.cc: 288: FAILED assert(info.last_epoch_started >= info...
- https://github.com/ceph/ceph/pull/7177
- 09:49 PM Revision bcd37cf3 (ceph): Merge pull request #6877 from dillaman/wip-14001
- librbd: fix test case race condition for journaling ops
Reviewed-by: Josh Durgin <jdurgin@redhat.com> - 09:44 PM Bug #14028: PG stats are not refresh within osd_pg_stat_report_interval_max
- I just checked the logs again:...
- 04:14 PM Bug #14028 (Won't Fix): PG stats are not refresh within osd_pg_stat_report_interval_max
- OSDs do not seem to refresh their PG stats at least every osd_pg_stat_report_interval_max seconds.
On a test clust... - 09:22 PM rbd Bug #14001 (Fix Under Review): TestMockOperationSnapshotRemoveRequest.Success common/RWLock.h: 58...
- *master PR*: https://github.com/ceph/ceph/pull/6877
- 09:21 PM Revision 7f5ee337 (ceph): librbd: fix test case race condition for journaling ops
- Signed-off-by: Jason Dillaman <dillaman@redhat.com>
- 09:09 PM Messengers Bug #12805: msg/simple/SimpleMessenger.cc: 238: FAILED assert(!cleared)
- 08:34 PM Bug #13988 (In Progress): new OSD re-using old OSD id fails to boot
- 08:34 PM Bug #13988: new OSD re-using old OSD id fails to boot
- To repeat the problem :
* on master https://github.com/ceph/ceph/pull/6876
* on infernalis https://github.com/ce... - 05:34 PM Bug #13988: new OSD re-using old OSD id fails to boot
- Removed all OSDs (0, 1, 2) and then ceph-prepare /dev/vdb which created the osd.0 and shows the same symptoms. The os...
- 04:09 PM Bug #13988: new OSD re-using old OSD id fails to boot
- The attached files were obtained with a slightly different set of commands:...
- 07:34 PM Revision 9162bd29 (ceph): Revert "LifeCycle feature"
- It's not there yet. Created a branch with a few fixes, will need to go
through some more revisions.
This reverts com... - 07:30 PM Feature #14041 (Closed): EC overwrites: deep scrub and checksumming
- EC overwrites preclude our current checksumming strategy. Instead, on each shard, we'll maintain a set of omap entri...
- 07:27 PM Feature #14040 (Resolved): ECBackend support for RMW
- Partial updates will require an RMW. This needs to be implemented. Replace this description with a design strategy ...
- 07:26 PM Feature #14039 (Fix Under Review): ECBackend cache extents with unapplied writes
- While a particular stripe has pending unapplied writes, we need to cache the unstable data to serve reads.
- 07:24 PM rbd Bug #14030: rbd merge-diff doesn't properly handle >2GB diffs
- It only seeks to skip parts of the first diff that were overwritten, so a workaround would be to pass the first diff ...
- 05:41 PM rbd Bug #14030 (Resolved): rbd merge-diff doesn't properly handle >2GB diffs
- It is using lseek instead of lseek64 -- resulting in an overflow....
- 07:15 PM Feature #14038 (Fix Under Review): ECBackend: delay writes which cause a clone to be created on a...
- Otherwise, we have to deal with cloning the temp object as well or something equally wonky. The first step on this i...
- 07:12 PM Feature #14037 (Fix Under Review): PGBackend: add a way for the Backend to notify ReplicatedPG th...
- (Implicit in the above is that ECBackend drives application, and PG is merely notified, this may not be the best desi...
- 07:08 PM Feature #14036 (Fix Under Review): EC overwrites: PGBackend needs to be replumbed to support over...
- It's not quite as simple as just using the write PGTransaction call directly. ECBackend will need to write the overw...
- 07:01 PM Feature #14035 (Fix Under Review): EC overwrites -- cleanup of ReplicatedPG::do_op to actually al...
- Overwrites are currently disallowed, we'll need to change that.
- 06:56 PM Feature #14034 (New): EC Overwrites: Phase 2: PGLog support for recording participants
- EC overwrites require two new concepts:
1) Each log entry needs to include information which records which shards (s... - 06:51 PM Feature #14033 (Fix Under Review): Activation support for EC Overwrites
- During activation, the primary and replicas must all apply entries up to last_update.
- 06:50 PM Feature #14032 (New): EC Overwrites: Phase 2: Peering support for ec overwrites
- The current design for ec overwrites allows for replicas to be missing at the head of their log log entries in which ...
- 06:39 PM Feature #14031 (New): EC overwrites
- This feature is complete when we support overwrites on EC pools. Linked tickets provide details.
https://github.c... - 06:08 PM Revision 82c40b67 (ceph): signal_handler: added support for using reentrant strsignal() implement...
- musl libc does not provide sys_siglist. It does provide a reentrant implementation of strsignal().
Added a cmake bui... - 06:01 PM Revision ae114aea (ceph): cmake: support ccache via a WITH_CCACHE build option
- When -DWITH_CCACHE=ON is specified, try to find and use ccache.
Signed-off-by: John Coyle <dx9err@gmail.com> - 05:48 PM Bug #14015: osd/PG.cc: 288: FAILED assert(info.last_epoch_started >= info.history.last_epoch_star...
- PR to master: https://github.com/ceph/ceph/pull/6793
- 05:41 PM Bug #14015: osd/PG.cc: 288: FAILED assert(info.last_epoch_started >= info.history.last_epoch_star...
- We need to backport to Hammer too, so I added it.
- 05:08 PM CephFS Bug #14029 (Resolved): "[ FAILED ] LibCephFS.OpenLayout" in smoke-master-distro-basic-openstack
- Run: http://pulpito.ovh.sepia.ceph.com:8081/teuthology-2015-12-09_05:00:02-smoke-master-distro-basic-openstack/
Job:... - 02:09 PM Revision b63c3fa2 (ceph): Merge pull request #6331 from chenji-kael/wip-lifecycle
- LifeCycle feature
- 01:02 PM Bug #13986: TEST_scrub_snaps: timeout > 300
- To reproduce the problem:
* in the background run... - 12:56 PM Bug #13986 (In Progress): TEST_scrub_snaps: timeout > 300
- 10:20 AM Bug #13986: TEST_scrub_snaps: timeout > 300
- crash with monitor logs and --osd-scrub-load-threshold=2000 http://dachary.org/loic/crash.gz
- 12:56 PM Bug #14027 (Resolved): tests: set --osd-scrub-load-threshold=2000
- 10:50 AM Bug #14027: tests: set --osd-scrub-load-threshold=2000
- Although it would be worth porting to hammer and infernalis, the race conditions are rare enough that they effectivel...
- 10:49 AM Bug #14027: tests: set --osd-scrub-load-threshold=2000
- https://github.com/ceph/ceph/pull/6871
- 10:47 AM Bug #14027 (Resolved): tests: set --osd-scrub-load-threshold=2000
- In a test environment, consistency is more important than
performances. Effectively disable the test that would post... - 12:55 PM Revision 461a1911 (ceph): Merge pull request #6871 from dachary/wip-14027-scrub-load
- tests: --osd-scrub-load-threshold=2000 for more consistency
Reviewed-by: Xinze Chi <xinze@xsky.com> - 10:48 AM Revision 99bcc7cf (ceph): tests: --osd-scrub-load-threshold=2000 for more consistency
- In a test environment, consistency is more important than
performances. Effectively disable the test that would postp... - 10:36 AM Revision 26008e8a (ceph): ceph-detect-init: Ubuntu >= 15.04 uses systemd
- Ubuntu releases since Ubuntu 15.04 use systemd instead of upstart;
teach ceph-detect-init that this is the case, thus... - 10:21 AM Bug #14026: OSD utilization is abnormal after data disk lost
- in function "OSDService::update_osd_stat", local variable "stbuf"" isn't initialized and it's still used to
update t... - 09:45 AM Bug #14026 (Resolved): OSD utilization is abnormal after data disk lost
- environment: centos7 + ceph-0.94.5
problem: if plug out the data disk, or remove the data directory of a osd, its ub... - 09:47 AM Revision 93cdc989 (ceph): pybind/rbd.pyx: misc typo bugfixes
- 09:46 AM Revision 8fabcbb4 (ceph): ceph-fuse:print usage information when no parameter specified
- Signed-off-by: Bo Cai <cai.bo@h3c.com>
- 09:07 AM Revision 469b55a6 (ceph): pybind/rbd.pyx: PyString -> PyBytes
- 08:12 AM Revision 50022d34 (ceph): os: skip checking pg_meta object existance in FileStore
- pg_meta object in FileStore is actually a logical object without any significant information.
All it data writes to o... - 05:53 AM CephFS Backport #13593 (Resolved): Ceph-fuse won't start correctly when the option log_max_new in ceph.c...
- 05:52 AM Revision f480cea2 (ceph): Merge pull request #6395 from SUSE/wip-13593-infernalis
- Ceph-fuse won't start correctly when the option log_max_new in ceph.conf set to zero
Reviewed-by: Greg Farnum <gfarn... - 12:50 AM Revision 1df97051 (ceph): test/librbd/fsx: musl libc doesn't implement random_r. Use c++11 std::m...
- Signed-off-by: John Coyle <dx9err@gmail.com>
- 12:38 AM Revision 6f1fb161 (ceph): doc: osd: s/schedued/scheduled/
- Signed-off-by: Loic Dachary <loic@dachary.org>
- 12:11 AM rbd Subtask #13293 (Resolved): Journal API + rbd CLI updates
- https://github.com/ceph/ceph/pull/6665
- 12:10 AM rbd Feature #12112 (Resolved): rbd: add --object-size option
12/08/2015
- 11:30 PM Revision f9f4d5c6 (ceph): Merge pull request #6830 from vumrao/wip-12112-vumrao
- rbd: add --object-size option, deprecate --order
Reviewed-by: Josh Durgin <jdurgin@redhat.com> - 11:11 PM rgw Backport #12856 (Resolved): rgw: missing handling of encoding-type=url when listing keys in bucket
- 11:11 PM rgw Backport #13047 (Resolved): Content-Type header should have correct initial capitals
- 11:10 PM rgw Backport #13538 (Resolved): rgw: orphan tool should be careful about removing head objects
- 11:10 PM rgw Backport #13540 (Resolved): rgw: get bucket location returns region name, not region api name
- 11:10 PM rgw Backport #13716 (Resolved): rgw:swift use Civetweb ssl can not get right url
- 11:10 PM Backport #14012 (Resolved): infernalis: ceph-disk workunits must not rely on augtool
- 11:10 PM devops Backport #14019 (Resolved): infernalis: libunwind package missing on CentOS 7
- 07:59 AM devops Backport #14019 (In Progress): infernalis: libunwind package missing on CentOS 7
- 07:55 AM devops Backport #14019 (Resolved): infernalis: libunwind package missing on CentOS 7
- https://github.com/ceph/ceph/pull/6845
- 11:09 PM Revision 7d93cf43 (ceph): buffer: make usable outside of ceph source again
- Add a rados/buffer_fwd.h symlink and remove the "include/" prefix from
buffer.h.
Signed-off-by: Josh Durgin <jdurgin... - 11:06 PM Revision d3aaeae1 (ceph): Merge pull request #6828 from dachary/wip-ceph-disk-augeas
- tests: ceph-disk workunit uses configobj
Reviewed-by: Loic Dachary <ldachary@redhat.com> - 10:47 PM devops Bug #14011: resolve CentOS versions conflict
- Can we just alias "7" to "the latest 7.1.whatever"?
- 10:31 PM devops Bug #14011 (Fix Under Review): resolve CentOS versions conflict
- https://github.com/ceph/teuthology/pull/742
- 10:16 PM Bug #13990: Hammer (0.94.3) OSD does not delete old OSD Maps in a timely fashion (maybe at all?)
- Having never used ceph-post-file, are there any special args to get it to work? I get a "connect to host drop.ceph.c...
- 03:39 AM Bug #13990: Hammer (0.94.3) OSD does not delete old OSD Maps in a timely fashion (maybe at all?)
- Well since restarting got the OSDs to start trimming, we've now restarted all OSDs in our cluster as we were above 80...
- 08:04 PM rgw Bug #14025: Bucket DELETE fails, 404 NoSuchKey, but cannot reproduce on other bucket names or users
- looking at the logs, it seems to me that the bucket entry point still exists but the bucket instance metadata does no...
- 07:56 PM rgw Bug #14025 (Can't reproduce): Bucket DELETE fails, 404 NoSuchKey, but cannot reproduce on other b...
- This is a weird one, I can't reproduce the problem with any other bucket name, or the same bucket name but on a diffe...
- 06:41 PM Revision 0afaeb24 (ceph): tests: ceph-disk workunit uses configobj
- Instead of using augtool to modify the configuration file, use
configobj. It is also used by the install teuthology t... - 06:41 PM Revision ebc5e4ee (ceph): tests: ceph-disk workunit uses the ceph task
- The ceph-disk workunit deploy keys that are not deployed by default by
the ceph teuthology task.
The OSD created by ... - 06:02 PM Revision 93d3dfe0 (ceph): Fixing NULL pointer dereference
- Signed-off-by: Igor Fedotov <ifedotov@mirantis.com>
- 05:24 PM Linux kernel client Bug #14022: map_sem for read + request_mutex are held indefinitely
- Can you also attach the entire syslog for this boot? Compress it, if it ends up too big, just email it to me directly.
- 04:56 PM Linux kernel client Bug #14022: map_sem for read + request_mutex are held indefinitely
- Can you do the following:
# echo "waiting tasks" >/dev/kmsg
# echo w >/proc/sysrq-trigger
# echo "end of waiting... - 09:08 AM Linux kernel client Bug #14022: map_sem for read + request_mutex are held indefinitely
- Output of...
- 09:04 AM Linux kernel client Bug #14022 (Resolved): map_sem for read + request_mutex are held indefinitely
- Hi,
when using rsync, my kernel rbd device hangs, here is the data i collected so far:... - 05:16 PM devops Feature #5014: arm: Build ARM packages
- A quick note just to mention that we are *not currently building ARM binaries* (as of Dec 2015).
- 05:02 PM Bug #13986: TEST_scrub_snaps: timeout > 300
- The relevant lines of the osd logs are:...
- 05:00 PM Bug #13986: TEST_scrub_snaps: timeout > 300
- https://github.com/ceph/ceph/pull/6860
- 04:57 PM Bug #13986 (Fix Under Review): TEST_scrub_snaps: timeout > 300
- 03:35 PM Bug #13986 (Duplicate): TEST_scrub_snaps: timeout > 300
- #13988
- 01:42 PM Bug #13986: TEST_scrub_snaps: timeout > 300
- Running
while : ; do echo -n . ; test/osd/osd-scrub-snaps.sh > test.out 2>&1 || break ; done
on rex003
- 12:58 PM Bug #13986: TEST_scrub_snaps: timeout > 300
- https://github.com/ceph/ceph/pull/6857 to collect more info about the possible race
- 12:29 PM Bug #13986: TEST_scrub_snaps: timeout > 300
- http://jenkins.ceph.dachary.org/job/ceph/LABELS=centos-7&&x86_64/9966/console
- 04:58 PM Revision 72785ee9 (ceph): rgw: enforce SLO part's ETag match during GET on SLO of Swift.
- Signed-off-by: Radoslaw Zarzynski <rzarzynski@mirantis.com>
- 04:58 PM Revision dabf5356 (ceph): rgw: handle errors during GET on Swift's SLO appropriately.
- Signed-off-by: Radoslaw Zarzynski <rzarzynski@mirantis.com>
- 04:58 PM Revision 31f1b89f (ceph): rgw: fix wrong ofs handling during GET on Swift's SLO.
- Signed-off-by: Radoslaw Zarzynski <rzarzynski@mirantis.com>
- 04:58 PM Revision 630eac10 (ceph): rgw: add support for normalisation of ranged SLO requests.
- Signed-off-by: Radoslaw Zarzynski <rzarzynski@mirantis.com>
- 04:58 PM Revision 1cf149d6 (ceph): rgw: append X-Static-Large-Object attribute during PUT on SLO.
- Signed-off-by: Radoslaw Zarzynski <rzarzynski@mirantis.com>
- 04:58 PM Revision 4bab3fca (ceph): rgw: move RGWBulkDelete in rgw/rgw_op.h.
- Signed-off-by: Radoslaw Zarzynski <rzarzynski@mirantis.com>
- 04:58 PM Revision 0a36b957 (ceph): rgw: fix iterate_slo_parts() and improve its declaration's formatting.
- Signed-off-by: Radoslaw Zarzynski <rzarzynski@mirantis.com>
- 04:58 PM Revision 7e06dba6 (ceph): rgw: add support for multipart delete on a SLO of Swift.
- Signed-off-by: Radoslaw Zarzynski <rzarzynski@mirantis.com>
- 04:58 PM Revision fa71082e (ceph): rgw: make response for multipart delete on a SLO compliant with Swift.
- Signed-off-by: Radoslaw Zarzynski <rzarzynski@mirantis.com>
- 04:58 PM Revision d8593b16 (ceph): rgw: read and parse put slo obj request
- Signed-off-by: Yehuda Sadeh <yehuda@redhat.com>
Signed-off-by: Radoslaw Zarzynski <rzarzynski@mirantis.com>
Conflict... - 04:58 PM Revision 7f5fdced (ceph): rgw: store slo info when putting object
- Signed-off-by: Yehuda Sadeh <yehuda@redhat.com>
Signed-off-by: Radoslaw Zarzynski <rzarzynski@mirantis.com>
Conflict... - 04:58 PM Revision 1df28c8e (ceph): rgw: iterate over slo parts
- Read slo parts when a read request was sent.
Signed-off-by: Yehuda Sadeh <yehuda@redhat.com>
Signed-off-by: Radoslaw... - 04:57 PM Revision ffb4691c (ceph): rgw: s/obj_manifest/dlo_manifest
- Be more clear, as slo_manifest will come later.
Signed-off-by: Yehuda Sadeh <yehuda@redhat.com>
Signed-off-by: Rados... - 04:57 PM Revision b665b046 (ceph): rgw: add support for Bulk Delete API of Swift.
- Signed-off-by: Radoslaw Zarzynski <rzarzynski@mirantis.com>
- 04:57 PM Revision c421037f (ceph): rgw: enable chunking in Bulk Delete API implementation.
- Signed-off-by: Radoslaw Zarzynski <rzarzynski@mirantis.com>
- 04:57 PM Revision 60018d92 (ceph): rgw: improve response code handling for Bulk Delete API.
- Signed-off-by: Radoslaw Zarzynski <rzarzynski@mirantis.com>
- 04:57 PM Revision 8eb80485 (ceph): rgw: support bucket removal in Bulk Delete API implementation.
- Signed-off-by: Radoslaw Zarzynski <rzarzynski@mirantis.com>
- 04:57 PM Revision 4ea0c7c6 (ceph): rgw: XMLFormatter may print XML tags lowercased and underscored now.
- Signed-off-by: Radoslaw Zarzynski <rzarzynski@mirantis.com>
- 04:57 PM Revision 2ff837f4 (ceph): rgw: rework authorization parts to accomodate Bulk Delete API.
- Signed-off-by: Radoslaw Zarzynski <rzarzynski@mirantis.com>
- 04:57 PM Revision 619e9452 (ceph): rgw: improve code formatting in Swift's DLO implementation.
- Signed-off-by: Radoslaw Zarzynski <rzarzynski@mirantis.com>
- 04:57 PM Revision 2b45910f (ceph): rgw: implement std::streambuf over RGWClientIO.
- Signed-off-by: Radoslaw Zarzynski <rzarzynski@mirantis.com>
- 04:57 PM Revision 3ecc2c3e (ceph): rgw: RGWFormatter_Plain does support key-value style for Bulk Delete API.
- Signed-off-by: Radoslaw Zarzynski <rzarzynski@mirantis.com>
- 04:57 PM Revision 2945bef8 (ceph): rgw: add support for putting Swift's X-Object-Manifest through POST.
- Fixes: #12886
Signed-off-by: Radoslaw Zarzynski <rzarzynski@mirantis.com> - 04:57 PM Revision 66f47f35 (ceph): rgw: fix wrong first byte pos in Content-Range HTTP header of Swift's DLO.
- Fixes: #13452
Signed-off-by: Radoslaw Zarzynski <rzarzynski@mirantis.com> - 04:57 PM Revision b48f3d77 (ceph): rgw: fix wrong length in Content-Range HTTP header of Swift's DLO.
- Fixes: #13452
Signed-off-by: Radoslaw Zarzynski <rzarzynski@mirantis.com> - 04:53 PM rbd Subtask #13296 (In Progress): Mirroring API + rbd CLI updates for images
- 04:52 PM rbd Subtask #13294 (Resolved): Journal repair tools for rbd CLI
- https://github.com/ceph/ceph/pull/6665
- 04:52 PM rbd Feature #11287 (Resolved): librbd: write to journal
- 04:52 PM rbd Feature #12698 (Resolved): librbd: write maintenance operations to journal
- 04:51 PM rbd Subtask #13292 (Resolved): Mirroring API + rbd CLI updates for pool
- 04:19 PM Revision 94be4526 (ceph): rgw: refactor arrangement of parameters for rgw_read_user_buckets().
- Signed-off-by: Radoslaw Zarzynski <rzarzynski@mirantis.com>
- 04:15 PM Revision 58adddaa (ceph): rgw: add support for "end_marker" on listing Swift account.
- Fixes: #10682
Signed-off-by: Radoslaw Zarzynski <rzarzynski@mirantis.com> - 04:03 PM Revision 2d8d7061 (ceph): Merge pull request #6857 from dachary/wip-13986-test-pg-scrub
- tests: osd-scrub-snaps.sh to display full osd logs on error
Reviewed-by: Xinze Chi <xinze@xsky.com> - 03:09 PM Feature #13867 (In Progress): Make non-client op users of OpContext use something else
- 03:09 PM Feature #11676: suite: extend the Product part of build_matrix to pass mincyclity properly
- 02:46 PM rbd Bug #14016 (Duplicate): ./common/RWLock.h: 58: FAILED assert(!is_locked())
- Duplicate of #14001
- 01:48 AM rbd Bug #14016 (Duplicate): ./common/RWLock.h: 58: FAILED assert(!is_locked())
Seen in make -j4 check with parallel make, when it runs test/run-rbd-unit-tests.sh
on master (b6672594702eb1b6b3dd...- 02:20 PM Backport #13929 (New): infernalis: Ceph Pools' MAX AVAIL is 0 if some OSDs' weight is 0
- 02:10 PM Revision 1b6a7e36 (ceph): doc: note that cephfs auth stuff is new in jewel
- + bonus whitespace fixes to the code blocks to
get them to render properly
Signed-off-by: John Spray <john.spray@red... - 12:54 PM Revision 2c7c7129 (ceph): tests: osd-scrub-snaps.sh to display full osd logs on error
- This generates a large output but is the only way to get more
information about race conditions that turn out to show... - 12:31 PM Revision 76467ee1 (ceph): Merge pull request #6850 from H3C/wip-zcl-ceph-mds
- ceph-mds:add --help/-h
Reviewed-by: John Spray <john.spray@redhat.com> - 11:47 AM Revision 534ff381 (ceph): Merge pull request #6848 from H3C/wip-zcl-layout-rst
- doc:file must be empty when writing layout fields of file use "setfattr"
Reviewed-by: John Spray <john.spray@redhat.... - 10:44 AM Revision 71690e70 (ceph): rbd: add --object-size option
- Object size can be specified when creating an image with the --order option,
as a number of bits in the size.
This p... - 10:41 AM Bug #13952 (Duplicate): /lib/systemd/system/ceph-disk@.service il using /bin/flock instead of /us...
- 09:50 AM Bug #13952: /lib/systemd/system/ceph-disk@.service il using /bin/flock instead of /usr/bin/flock ...
- Dup with #14013. To be closed.
- 09:27 AM CephFS Backport #13889 (In Progress): infernalis: Segmentation fault accessing file using fuse mount
- 09:20 AM Backport #14013 (In Progress): infernalis: systemd/ceph-disk@.service assumes /bin/flock
- 09:17 AM Backport #14018 (In Progress): infernalis: osd/PG.cc: 288: FAILED assert(info.last_epoch_started ...
- 07:55 AM Backport #14018 (Resolved): infernalis: osd/PG.cc: 288: FAILED assert(info.last_epoch_started >= ...
- https://github.com/ceph/ceph/pull/6851
- 09:12 AM Backport #13979 (In Progress): infernalis : "shard missing" errors in logs during Teuthology run
- 09:01 AM Revision a69367d9 (ceph): ceph-mds:add --help/-h
- When use ceph-mds --hlep/-h, the usage will be showed
Signed-off-by: Cilang Zhao <zhao.cilang@h3c.com> - 08:57 AM Backport #13638 (In Progress): FileStore: potential memory leak if getattrs fails.
- 08:45 AM Revision 07c334a2 (ceph): doc:file must be empty when writing layout fields of file use "setfattr"
- When the layout fields of a file are modified, this file must be empty, otherwise an error would occur.
Signed-off-b... - 08:44 AM Bug #9978: keyvaluestore: void ECBackend::handle_sub_read
- Unfortunately, we encountered this bug recently. I found a workaround and hope anyone who came to this page desperate...
- 08:42 AM devops Bug #13997: libunwind package missing on CentOS 7
- Not backporting to hammer which has an ancient version anyway
- 08:42 AM devops Bug #13997 (Resolved): libunwind package missing on CentOS 7
- 07:50 AM devops Bug #13997 (Pending Backport): libunwind package missing on CentOS 7
- 07:23 AM devops Bug #13997: libunwind package missing on CentOS 7
- https://github.com/ceph/ceph/pull/6844
- 06:48 AM devops Bug #13997 (In Progress): libunwind package missing on CentOS 7
- 06:27 AM devops Bug #13997: libunwind package missing on CentOS 7
- Created https://github.com/ceph/ceph/pull/6842
- 02:05 AM devops Bug #13997: libunwind package missing on CentOS 7
- Fresh Centos 7 install....
- 08:35 AM Revision a2644ed5 (ceph): client: use null snapc to check pool permission
- snap inodes' ->snaprealm can be NULL, so dereferencing it in
check_pool_perm() can cause segment fault. The pool perm... - 08:34 AM Revision 476af491 (ceph): Merge pull request #6845 from dachary/wip-14019-infernalis
- infernalis: libunwind package missing on CentOS 7
Reviewed-by: Brad Hubbard <bhubbard@redhat.com> - 08:29 AM Revision 7d90c018 (ceph): build/ops: systemd ceph-disk unit must not assume /bin/flock
- The flock command may be installed elsewhere, depending on the
system. Let the PATH search figure that out.
http://t... - 08:25 AM Revision 5ef86d78 (ceph): osd: Test osd_find_best_info_ignore_history_les config in another assert
- Signed-off-by: David Zafman <dzafman@redhat.com>
(cherry picked from commit 02a9a41f151a3d968bf8066749658659dc6e3ac4) - 08:17 AM Revision 7ac5b151 (ceph): osd: call on_new_interval on newly split child PG
- We must call on_new_interval() on any interval change *and* on the
creation of the PG. Currently we call it from PG:... - 08:05 AM Revision 5d1e6ae4 (ceph): FileStore: potential memory leak if _fgetattrs fails
- Memory leak happens if _fgetattrs encounters some error and simply returns.
Fixes: #13597
Signed-off-by: xie xingguo ... - 08:03 AM Revision 84ee7c33 (ceph): Merge remote-tracking branch 'ceph/jewel'
- 08:02 AM Bug #14021 (Closed): osd coredump on rhel7.1
- we find 2 coredumps in log file after some time of running, can anyone help to check this?
coredump1:
ceph version ... - 07:58 AM Revision 54c8af80 (ceph): build/ops: enable CR in CentOS 7
- To get libunwind from the CR repositories until CentOS 7.2.1511 is released.
http://tracker.ceph.com/issues/13997 Fi... - 07:55 AM Bug #14015: osd/PG.cc: 288: FAILED assert(info.last_epoch_started >= info.history.last_epoch_star...
- @David, this is to be backported to infernalis ? Or also hammer ?
- 07:52 AM Revision 2c1af382 (ceph): Merge pull request #6844 from dachary/wip-13997-centos-libunwind-jewel
- build/ops: enable CR in CentOS 7
Reviewed-by: Brad Hubbard <bhubbard@redhat.com> - 07:50 AM Bug #8379: cephx: clock skew causes verify_reply decode error
- karan singh wrote:
> Looks like this this error is not ONLY because of date / time sync.
>
> I have date / time s... - 07:25 AM Revision 247ee608 (ceph): build/ops: enable CR in CentOS 7
- To get libunwind from the CR repositories until CentOS 7.2.1511 is released.
http://tracker.ceph.com/issues/13997 Fi... - 07:20 AM Revision f9a230aa (ceph): test/osd: add test case for shutdown_when_flapping
- TEST_markdown_exceed_maxdown_count :
down N+1 times within period, should be UP in first N times but DOWN in last t... - 06:47 AM Revision 91264778 (ceph): Merge pull request #6842 from badone/wip-13997
- tests: centos7 needs the Continuous Release (CR) Repository enabled for
Reviewed-by: Loic Dachary <ldachary@redhat.com> - 06:07 AM Revision b47eeae1 (ceph): tests: centos7 needs the Continuous Release (CR) Repository enabled for
- libunwind
Fixes: #13997
Signed-off-by: Brad Hubbard <bhubbard@redhat.com> - 03:22 AM Revision 7d48f62f (ceph): LifeCycle feature
- As same as amazon S3 interface,"PUT Bucket lifecycle" and
"DELETE Bucket lifecycle" have been implemented,
"GET Bucke... - 03:15 AM CephFS Bug #12715: "[ERR] bad backtrace on dir ino 600" in cluster log"
- it's safe to ignore them
- 02:33 AM Revision 0565055b (ceph): rgw: fix a typo in init-radosgw
- Signed-off-by: Zhi Zhang zhangz.david@outlook.com
- 01:03 AM Messengers Bug #13985: Messenger: large message data content length cause int overflow
- we setup a cluster, and make 4 million dirs in fs, which result mds hang, then we restart mds, it sends getomapvals ...
- 12:48 AM Revision 78bb6034 (ceph): Merge pull request #757 from dachary/wip-dummy
- Revert "dummy: only use the user playbook"
- 12:47 AM Revision d0280d79 (ceph): Revert "dummy: only use the user playbook"
- This reverts commit a016f29e211e378a50fd833d940abbb49449c37b.
- 12:06 AM Revision 8676120b (ceph): Merge pull request #6818 from trociny/fix-cmake
- cmake: update for recent rbd changes
Reviewed-by: Josh Durgin <jdurgin@redhat.com>
12/07/2015
- 11:52 PM Revision caedebde (ceph): Merge pull request #6837 from dreamhost/wip-test-cephtool-filenames
- qa/workunits/cephtool/test.sh: false positive fail on /tmp/obj1.
Reviewed-by: Josh Durgin <jdurgin@redhat.com> - 11:27 PM Revision cd0c13bb (ceph): qa/workunits/cephtool/test.sh: false positive fail on /tmp/obj1.
- If /tmp/obj1 happened to exist already, and was not writable by the
testing user, then this test failed!
Signed-off-... - 11:21 PM Revision f112e25f (ceph): Merge pull request #6838 from stiopaa1/Wreorder_ceph_context
- common/ceph_context.cc:fix order of initialisers
Reviewed-by: Josh Durgin <jdurgin@redhat.com> - 11:11 PM Bug #14015 (Resolved): osd/PG.cc: 288: FAILED assert(info.last_epoch_started >= info.history.last...
Hit this assert when setting osd_find_best_info_ignore_history_les to get incomplete PGs to active+clean with known...- 11:06 PM Bug #13805: segfault in Objecter::handle_watch_notify
- Sage, is this still in testing? Do we have a commit hash?
- 11:02 PM Backport #13791 (In Progress): Objecter: potential null pointer access when do pool_snap_list.
- 11:02 PM Linux kernel client Bug #14010 (Resolved): xfstests generic/247 failure
- 05:29 PM Linux kernel client Bug #14010: xfstests generic/247 failure
- https://github.com/ceph/ceph/pull/6831
- 05:19 PM Linux kernel client Bug #14010 (Resolved): xfstests generic/247 failure
- xfstests generic/247 often fails with a warning:
[ 1135.748673] WARNING: CPU: 0 PID: 16554 at fs/xfs/xfs_file.c:75... - 11:02 PM Revision af2ced59 (ceph): Objecter: remove redundant result-check of _calc_target in _map_session.
- Result-code check is currently redundant since _calc_target never returns a negative value.
Signed-off-by: xie xinggu... - 11:01 PM Revision 43dede31 (ceph): Objecter: potential null pointer access when do pool_snap_list.
- Objecter: potential null pointer access when do pool_snap_list. Shall check pool existence first.
Fixes: #13639
Signe... - 11:01 PM Backport #13789 (In Progress): Objecter: potential null pointer access when do pool_snap_list.
- 11:01 PM Revision b6672594 (ceph): Merge pull request #6831 from fullerdj/wip-14010
- qa/krbd: Expunge generic/247
Reviewed-by: Josh Durgin <jdurgin@redhat.com> - 11:00 PM Revision 6cf261cd (ceph): Objecter: remove redundant result-check of _calc_target in _map_session.
- Result-code check is currently redundant since _calc_target never returns a negative value.
Signed-off-by: xie xinggu... - 11:00 PM Revision 5d0b5c13 (ceph): Objecter: potential null pointer access when do pool_snap_list.
- Objecter: potential null pointer access when do pool_snap_list. Shall check pool existence first.
Fixes: #13639
Signe... - 10:56 PM Revision 74ee7bd2 (ceph): Merge pull request #6793 from dzafman/wip-hist-les-fix
- osd: Allow repair of history.last_epoch_started using config
Reviewed-by: Samuel Just <sjust@redhat.com> - 10:43 PM Backport #13769: EC async "fast read" patchset for hammer (EC read latency improvements)
- The three commits to be cherry-picked are listed in http://tracker.ceph.com/issues/13769#note-1
Sam cannot get at ... - 10:32 PM Backport #13769: EC async "fast read" patchset for hammer (EC read latency improvements)
- Sam, can this be backported to hammer now?
- 10:39 PM Revision a80ff1fe (ceph): common/ceph_context.cc:fix order of initialisers
- Fix order of member initialisers causing [-Wreorder] warning.
Signed-off-by: Michal Jarzabek <stiopa@gmail.com> - 10:31 PM Backport #13891 (In Progress): infernalis: auth/cephx: large amounts of log are produced by osd
- 10:30 PM Revision ac31f7f8 (ceph): auth/cephx: large amounts of log are produced by osd
- if the auth of osd is deleted when the osd is running, the osd will produce large amounts of log.
Fixes:#13610
Signe... - 10:29 PM Backport #13892 (In Progress): hammer: auth/cephx: large amounts of log are produced by osd
- 10:29 PM Revision 66ff0aa0 (ceph): auth/cephx: large amounts of log are produced by osd
- if the auth of osd is deleted when the osd is running, the osd will produce large amounts of log.
Fixes:#13610
Signe... - 10:26 PM Backport #13930 (In Progress): hammer: Ceph Pools' MAX AVAIL is 0 if some OSDs' weight is 0
- 10:26 PM Revision 61da13bb (ceph): mon/PGMonitor: MAX AVAIL is 0 if some OSDs' weight is 0
- In get_rule_avail(), even p->second is 0, it's possible to be used
as divisor and quotient is infinity, then is conve... - 10:22 PM Backport #13935 (In Progress): infernalis: Ceph daemon failed to start, because the service name ...
- 10:22 PM Revision c968a40d (ceph): init-ceph: fix systemd-run cant't start ceph daemon sometimes
- Fixes: #13474
Signed-off-by: Chuanhong Wang <wang.chuanhong@zte.com.cn>
(cherry picked from commit 2f36909e1e08bac993... - 10:15 PM Backport #13936 (In Progress): hammer: Ceph daemon failed to start, because the service name was ...
- 10:14 PM Revision 53e81aab (ceph): init-ceph: fix systemd-run cant't start ceph daemon sometimes
- Fixes: #13474
Signed-off-by: Chuanhong Wang <wang.chuanhong@zte.com.cn>
(cherry picked from commit 2f36909e1e08bac993... - 10:14 PM CephFS Bug #12715: "[ERR] bad backtrace on dir ino 600" in cluster log"
- This kind of warning is still happen in the following upgrading path.
Dumpling -> FireFly -> Hammer
Each vers... - 09:55 PM CephFS Bug #13907 (Duplicate): libcephfs-java failure on hammer
- 09:52 PM CephFS Backport #13928 (Duplicate): hammer: libcephfs-java failure on hammer
- 09:45 PM Bug #13990: Hammer (0.94.3) OSD does not delete old OSD Maps in a timely fashion (maybe at all?)
- If the log is too large, you can use ceph-post-file.
- 09:41 PM Bug #13990: Hammer (0.94.3) OSD does not delete old OSD Maps in a timely fashion (maybe at all?)
- The next step would be to post an osd log from an osd as it fails to trim maps. I suggest 1 hour of logs at
debug... - 08:40 PM Bug #13990: Hammer (0.94.3) OSD does not delete old OSD Maps in a timely fashion (maybe at all?)
- It was HEALTH_OK prior to upgrade for months. We've grown the cluster various times, and it has backfilled successfu...
- 07:04 PM Bug #13990: Hammer (0.94.3) OSD does not delete old OSD Maps in a timely fashion (maybe at all?)
- What's the history of this cluster during the upgrade? It sounds like you had dirty state for a very long time (preve...
- 09:43 PM Backport #14012 (In Progress): infernalis: ceph-disk workunits must not rely on augtool
- 09:42 PM Backport #14012 (Resolved): infernalis: ceph-disk workunits must not rely on augtool
- https://github.com/ceph/ceph/pull/6828
- 09:42 PM Backport #14013 (Resolved): infernalis: systemd/ceph-disk@.service assumes /bin/flock
- https://github.com/ceph/ceph/pull/6852
- 09:09 PM Bug #13986: TEST_scrub_snaps: timeout > 300
- http://jenkins.ceph.dachary.org/job/ceph/LABELS=ubuntu-14.04&&x86_64/9932/consoleFull
- 08:36 PM devops Bug #14011: resolve CentOS versions conflict
- @zack agree it's a good idea.
can ovh run centos 7.1 ? - 07:40 PM devops Bug #14011: resolve CentOS versions conflict
- Yes, I think that's a good solution Zack. We don't differentiate beyond that afaik.
- 07:05 PM devops Bug #14011: resolve CentOS versions conflict
- I could tell teuthology to only submit X.Y:
14.04 -> 14.04
7.1.1503 -> 7.1
9 -> 9
Do we want this? - 06:44 PM devops Bug #14011 (Resolved): resolve CentOS versions conflict
- on smithi's centos version is 7.1.1503 and it fails current tests as they ask for 7.0
and we need to reconcile this ... - 07:54 PM Revision 0a883a48 (ceph): Merge branch 'unlinkcallbyrmbucket'
- Conflicts:
src/rgw/rgw_bucket.cc
Reviewed-by: Yehuda Sadeh <yehuda@redhat.com> - 07:51 PM Revision 0186cdc6 (ceph): rgw:dont update entrypoint when removing bucket
- when removing bucket,we unlink the bucket bucket but we don't want
to update the entrypoint here.Because the update_e... - 07:10 PM Revision 23554c46 (ceph): Merge pull request #6735 from rzarzynski/wip-rgw-clean-putmetabucket
- rgw: remove unused variable in RGWPutMetadataBucket::execute
Reviewed-by: Yehuda Sadeh <yehuda@redhat.com> - 07:09 PM Revision 54a62225 (ceph): Merge pull request #6829 from LenzGr/master
- doc: Fixed incorrect name of a "List Multipart Upload Parts" Response Entity
Reviewed-by: Yehuda Sadeh <yehuda@redha... - 06:55 PM Messengers Bug #13985: Messenger: large message data content length cause int overflow
- Fixing this overflow is fine, but...what are you doing that you have a 2GB message? There's no way that's a safe thin...
- 06:42 PM Revision eb81cc44 (ceph): Merge pull request #6774 from ceph/wip-rgw-cap-fix
- rgw: add a missing cap type
- 06:41 PM Revision 15b24058 (ceph): osd: use hexdump when logging CRC values
- Signed-off-by: Mykola Golub <mgolub@mirantis.com>
- 06:41 PM Revision 9d88f1ac (ceph): osd: more scrubber fields in pg query output
- Signed-off-by: Mykola Golub <mgolub@mirantis.com>
- 06:41 PM Revision a828a4df (ceph): osd: fix arithmetic exception when scrub_min_interval is small
- Signed-off-by: Mykola Golub <mgolub@mirantis.com>
- 06:40 PM Bug #14009: FAILED assert(old_size == total_chunk_size) in 0.94.5 / strange file size
- I have workaround our issue by stopping the three OSDs that crash, and truncate this object to the right size:
cd ... - 05:24 PM Bug #14009: FAILED assert(old_size == total_chunk_size) in 0.94.5 / strange file size
- I think it is safe for you to remove the corrupted shard (thanks to Kefu for clarifying this ;-). It will be rebuilt ...
- 05:24 PM Bug #14009: FAILED assert(old_size == total_chunk_size) in 0.94.5 / strange file size
- xattr just in case....
- 05:24 PM Bug #14009: FAILED assert(old_size == total_chunk_size) in 0.94.5 / strange file size
- As requested by Loic:
# ceph osd pool get ec8p2 erasure_code_profile
erasure_code_profile: ec8p2profile
# ce... - 05:23 PM Bug #14009: FAILED assert(old_size == total_chunk_size) in 0.94.5 / strange file size
- It looks like the abnormal sized file is containing data for the first 524288 bytes then only zeroes....
- 05:23 PM Bug #14009: FAILED assert(old_size == total_chunk_size) in 0.94.5 / strange file size
- ...
- 05:22 PM Bug #14009: FAILED assert(old_size == total_chunk_size) in 0.94.5 / strange file size
- Per Loic request:...
- 05:22 PM Bug #14009: FAILED assert(old_size == total_chunk_size) in 0.94.5 / strange file size
- [0(9),524288, 8(7),524288, 9(2),1048576, 13(8),524288])
[2,1048576, 7,524288, 8,524288, 9,524288]
=> one piece i... - 05:22 PM Bug #14009: FAILED assert(old_size == total_chunk_size) in 0.94.5 / strange file size
- Missing part after len=276, (it's on two lines) :...
- 05:21 PM Bug #14009: FAILED assert(old_size == total_chunk_size) in 0.94.5 / strange file size
- For the record osd 30 is the one crashing the most frequently, OSD 33 and 9 are also crashing quickly, other OSD are ...
- 05:20 PM Bug #14009: FAILED assert(old_size == total_chunk_size) in 0.94.5 / strange file size
- debug osd = 20 log on osd.30, 300 MByte :
http://guerby.org/ftp/ceph-osd.30.log.xz
I did update ceph.conf, rest... - 05:19 PM Bug #14009: FAILED assert(old_size == total_chunk_size) in 0.94.5 / strange file size
- With:
ceph osd set norecover
The 3 OSDs do not crash.
But of course no recovery is happening so we're k... - 05:18 PM Bug #14009: FAILED assert(old_size == total_chunk_size) in 0.94.5 / strange file size
- The three OSD keep having the same error a short time after restart, log with ms = 1 objecter = 20 on osd.30 is attac...
- 05:16 PM Bug #14009: FAILED assert(old_size == total_chunk_size) in 0.94.5 / strange file size
We started seeing this tonight on our ceph 0.94.5 cluster tonight, RBD over EC pools (one 4+1, one 8+2), ubuntu tru...- 05:16 PM Bug #14009: FAILED assert(old_size == total_chunk_size) in 0.94.5 / strange file size
- h3. COPYING THE INITIAL REPORT
- 05:11 PM Bug #14009: FAILED assert(old_size == total_chunk_size) in 0.94.5 / strange file size
- There is a strange file on disk described here
http://tracker.ceph.com/issues/13972#note-32
http://tracker.ceph.c... - 05:09 PM Bug #14009: FAILED assert(old_size == total_chunk_size) in 0.94.5 / strange file size
- OSD crashing the most often is 30. The other two are 9 and 33.
Various log files:
http://guerby.org/ftp/ooops-30.... - 05:06 PM Bug #14009: FAILED assert(old_size == total_chunk_size) in 0.94.5 / strange file size
- ...
- 05:05 PM Bug #14009 (Can't reproduce): FAILED assert(old_size == total_chunk_size) in 0.94.5 / strange fil...
- Three of our OSD are failing with
2015-12-06 09:44:22.271873 7f47091fc700 -1 osd/ECUtil.cc: In function 'int ECUt... - 06:27 PM Revision 794e5182 (ceph): msg/Dispatcher.h:removed unneeded includes
- Removed includes and added forward declarations instead.
Signed-off-by: Michal Jarzabek <stiopa@gmail.com> - 05:24 PM Revision 9af8d6f0 (ceph): qa/krbd: Expunge generic/247
- xfstests generic/247 exercises XFS DIO and AIO to detect races. Some
races apparently exist in the XFS code as this i... - 04:52 PM Revision b04fc891 (ceph): rgw: fix wrong checking for JSONParser::parse() return
- The type of return value from JSONParser::parse() is bool, it never returns
a negative value.
Signed-off-by: Dunrong... - 04:27 PM Revision 2eb0bf32 (ceph): tests: make sure no segfault occurs when using some bad keyring
- Signed-off-by: Dunrong Huang <riegamaths@gmail.com>
- 04:22 PM rgw Documentation #14003 (In Progress): Incorrect name of the "List Multipart Upload Parts" Response ...
- 02:59 PM rgw Documentation #14003: Incorrect name of the "List Multipart Upload Parts" Response Entities
- Pull request: https://github.com/ceph/ceph/pull/6829
- 02:51 PM rgw Documentation #14003: Incorrect name of the "List Multipart Upload Parts" Response Entities
- The subject line of this bug is incorrect - it's the *name*, not *description* of the entity.
I'm about to prepare... - 02:01 PM rgw Documentation #14003 (Resolved): Incorrect name of the "List Multipart Upload Parts" Response Ent...
- The RADOS Gateway documentation at http://docs.ceph.com/docs/master/radosgw/s3/objectops/#list-multipart-upload-parts...
- 04:14 PM Bug #14008 (Can't reproduce): "FAILED assert(0 == "out of order op")" in rbd-hammer-distro-basic-...
- Run: http://pulpito.ovh.sepia.ceph.com:8081/teuthology-2015-12-06_18:00:02-rbd-hammer-distro-basic-openstack/
Job: [... - 04:07 PM rbd Bug #14007 (Resolved): TestLibRBD.* tests failed in upgrade:client-upgrade-jewel-distro-basic-ope...
- Run: http://teuthology.ovh.sepia.ceph.com/teuthology/teuthology-2015-12-04_23:20:01-upgrade:client-upgrade-jewel-dist...
- 03:22 PM Revision f2f92c84 (ceph): cmake: add TracepointProvider.cc to libcommon
- Signed-off-by: Orit Wasserman <owasserm@redhat.com>
- 03:16 PM Revision 12d5b65f (ceph): Merge pull request #6805 from zaitcev/cmake-pluginregistry
- cmake: Add common/PluginRegistry.cc to CMakeLists.txt
- 02:58 PM Backport #13929 (In Progress): infernalis: Ceph Pools' MAX AVAIL is 0 if some OSDs' weight is 0
- 02:57 PM rbd Feature #14006 (Resolved): Track IO dependencies in journal
- We already track when maintenance operations (resize, snapshot add/remove/etc, rename, etc) complete in the journal, ...
- 02:55 PM Bug #13993 (Fix Under Review): Use of syslog results in all log messages at priority "emerg"
- 01:25 AM Bug #13993 (Resolved): Use of syslog results in all log messages at priority "emerg"
- In src/log/Log.cc we call syslog like so;
syslog(LOG_USER, "%s%s", buf, s.c_str());
syslog(LOG_USER, "%s", s);
... - 02:54 PM rgw Bug #14005 (Resolved): RGW shouldn't send Content-Type nor Content-Length for 304 responses
- Currently if you get a 304 response (using If-None-Match in the request for instance), then you get in the response ...
- 02:54 PM Revision 3e23e45d (ceph): doc: Fixed incorrect name of a "List Multipart Upload Parts" Response E...
- Fixed incorrect name of a "List Multipart Upload Parts" Response Entity:
"InitiatedMultipartUploadsResult" -> "ListPa... - 02:40 PM Bug #14004 (Resolved): ceph-disk workunits must not rely on augtool
- Instead of using augtool to modify the configuration file, use
configobj. It is also used by the install teuthology ... - 02:02 PM rgw Bug #11494: rgw: multipart list part response returns incorrect field
- FYI, I filed bug #14003 for the documentation update.
- 01:26 PM Revision 1d84219b (ceph): mon/Monitor.cc: improve error handling
- sync_obtain_latest_monmap will fail with assertion failure if store->get()
returns error, but there's an extra assert... - 01:26 PM Revision bb4936aa (ceph): mon/Monitor.cc: attempt to decode and extract keyring only if it is found.
- Monitor::preinit wrongly assumes that keyring always exists. This worked
so far because the logic that followed made ... - 01:26 PM Revision 08ae31c0 (ceph): mon/ceph_mon.cc: fail on any error, not just empty magicbl
- Assume failure when either MonDBStore returns error or returns empty
bufferlist.
Signed-off-by: Piotr Dałek <piotr.d... - 01:26 PM Revision 3da96b96 (ceph): mon/Monitor.cc: Reuse empty bufferlist
- When Monitor::read_features_off_disk doesn't find features on disk, it
tries to write them back to disk. featuresbl i... - 01:26 PM Revision f5a7a3b1 (ceph): mon/MonitorDBStore: use single-key LevelDB::get() method
- Make MonitorDBStore use single-key ::get() method for some performance
increase in heavy K/V workloads.
Signed-off-b... - 01:26 PM Revision 1e6cda08 (ceph): mon/AuthMonitor: don't crash on ENOENT
- When MonDBStore returns ENOENT when trying to get mkfs keyring, monitor
crashes with assert. On the other hand, if it... - 01:26 PM Revision eec46112 (ceph): mon: remove unnecessary key existence check
- Since store->get() returns -ENOENT when key doesn't exists, there's no longer
needed to explicitly check for its exis... - 12:46 PM rbd Bug #14001: TestMockOperationSnapshotRemoveRequest.Success common/RWLock.h: 58: FAILED assert(!is...
- This is the first time I've seen this one, I'll add more if it happens again
http://jenkins.ceph.dachary.org/job/c... - 12:45 PM rbd Bug #14001 (Resolved): TestMockOperationSnapshotRemoveRequest.Success common/RWLock.h: 58: FAILED...
- ...
- 12:02 PM CephFS Bug #12175: Fix ceph-fuse --help
- Related: https://github.com/ceph/ceph/pull/6821
- 11:59 AM CephFS Feature #12273 (Resolved): Apply scrub tag to objects during forward scrub
- ...
- 11:58 AM CephFS Bug #13898 (Resolved): Bad layouts injected by cephfs-data-scan scan_inodes
- https://github.com/ceph/ceph/pull/6719
- 10:25 AM Linux kernel client Bug #13712: Page allocation failure in Kernel 4.2.3.
- The structure in question manages snapshots, but is allocated even if you have no snapshots. It just gets bigger if ...
- 09:45 AM Bug #13972: osd/ECUtil.h: 117: FAILED assert(old_size == total_chunk_size) in 0.80.10
- pg dump
- 09:44 AM Bug #13972: osd/ECUtil.h: 117: FAILED assert(old_size == total_chunk_size) in 0.80.10
- osd and pg dump
- 09:33 AM Bug #13972: osd/ECUtil.h: 117: FAILED assert(old_size == total_chunk_size) in 0.80.10
- crushmap per Loic request.
- 09:13 AM Revision c83d6dbf (ceph): cmake: update for recent rbd changes
- Signed-off-by: Mykola Golub <mgolub@mirantis.com>
- 09:02 AM Revision 8eb0ed1d (ceph): Merge pull request #6816 from drunkard/doc
- doc: rst style fix for pools document
Reviewed-by: Loic Dachary <ldachary@redhat.com> - 09:00 AM CephFS Feature #13999 (New): client: richacl support
- http://www.bestbits.at/richacl/
So far no other project has merged richacl patch. Should we be the first one.
- 08:57 AM CephFS Feature #13998 (Resolved): posix acl support
- https://github.com/ceph/ceph/pull/5658
- 08:52 AM Revision 273c52a5 (ceph): Merge pull request #6820 from mikulely/add-umcloud-affiliation
- Add UMCloud affiliation
Reviewed-by: Loic Dachary <ldachary@redhat.com> - 08:38 AM devops Bug #13997 (Resolved): libunwind package missing on CentOS 7
- As of December 6th, 2015...
- 08:25 AM Revision c3619f4c (ceph): tests: add destroy_ec_profile{,_pp} helpers
- Signed-off-by: Loic Dachary <ldachary@redhat.com>
(cherry picked from commit ab46d79bc09fc711fa35302f49eecac81a98519b) - 08:25 AM Revision 26adf7a3 (ceph): tests: destroy testprofile before creating one
- The testprofile erasure code profile is destroyed before creating a new
one so that it does not fail when another tes... - 08:25 AM Revision 0eb2ddb0 (ceph): tests: test/librados/test.cc must create profile
- Now that the create_one_ec_pool function removes the testprofile each
time it is called, it must create the testprofi... - 08:17 AM Revision 25c70cfc (ceph): Update Jiaying Ren affinity
- Signed-off-by: Jiaying Ren <jiaying.ren@umcloud.com>
- 08:17 AM Revision c11ca426 (ceph): Update Rongze Zhu affinity
- Signed-off-by: Jiaying Ren <jiaying.ren@umcloud.com>
- 08:14 AM Bug #13996 (Rejected): function:Monitor::preinit() dead lock if option sanitization failed
- version:0.94.5
function:monitor::preinit()
dead lock if option sanitization failed - 08:12 AM Bug #13995 (Can't reproduce): ceph-osd "nonce collision" due to unshared pid namespaces
- I found Issue #13032 yesterday after I've been hitting a brick wall while trying to deploy ceph on and off during the...
- 05:10 AM Revision da8a891b (ceph): Merge pull request #744 from ceph/wip-inode-claim
- tasks/cephfs: add TestJournalRepair.test_table_tool_take_inos
Reviewed-by: Greg Farnum <gfarnum@redhat.com> - 05:09 AM Revision 47c78934 (ceph): Merge pull request #6655 from jcsp/wip-inode-claim
- tools: add cephfs-table-tool 'take_inos'
Reviewed-by: Yan, Zheng <zyan@redhat.com> - 05:05 AM Revision 5e714c69 (ceph): Merge pull request #703 from ceph/wip-layout-restrict
- tasks/cephfs: test for auth caps pool-setting restriction
Reviewed-by: Greg Farnum <gfarnum@redhat.com> - 05:05 AM Revision 66de0e32 (ceph): Merge pull request #6567 from jcsp/wip-layout-restrict
- mds: add 'p' flag in auth caps to control setting pool in layout
Reviewed-by: Sage Weil <sage@redhat.com> - 05:04 AM Revision c619c161 (ceph): Merge pull request #728 from ceph/wip-snap-strays
- tasks/cephfs: cover snaps in test_strays
- 05:03 AM Revision b1d5c481 (ceph): Merge pull request #6601 from jcsp/wip-13777-caps
- mon: don't require OSD W for MRemoveSnaps
Reviewed-by: Kefu Chai <kchai@redhat.com>
Reviewed-by: Yan, Zheng <zyan@re... - 05:01 AM Revision 543d48db (ceph): Merge pull request #709 from ceph/wip-scrub-basic
- tasks/cephfs: add TestForwardScrub
- 04:33 AM Feature #13994 (Closed): librbd: Add iovec support for RBD C-API
- Hmm, in current librbd impl, it would be useless for reducing copying
- 03:56 AM Feature #13994 (Closed): librbd: Add iovec support for RBD C-API
- If caller has a iovec structure needed to passed in rbd, for c++ program, it could use aio_write(uint64_t off, size_t...
- 04:09 AM CephFS Bug #6613: samba is crashing in teuthology
- http://pulpito.ovh.sepia.ceph.com:8081/teuthology-2015-12-06_18:14:01-samba-hammer---basic-openstack/30493/
http:/... - 02:45 AM Revision 593c1249 (ceph): doc: rst style fix for pools document
- Signed-off-by: Drunkard Zhang <gongfan193@gmail.com>
12/06/2015
- 11:44 PM Linux kernel client Bug #13712: Page allocation failure in Kernel 4.2.3.
- We are also not using snapshots. Just a simple RBD, formatted as ext4. We had another panic just under two days ago, ...
- 02:24 PM Revision e62954e0 (ceph): deb,rpm: package buffer_fwd.h
- Introduced by 206b8e2cc6157451d5c7548e294461408e0b0c8f.
Signed-off-by: Sage Weil <sage@redhat.com> - 02:16 PM Bug #13992 (Resolved): LibRadosMiscConnectFailure.ConnectFailure (not so intermittent) failure in...
- ...
- 12:23 PM Revision b0a770a3 (ceph): workunit: fetch from --ceph-git-url
- The commit from which workunits are fetched must be retrieved
from --ceph-git-url via teuth_config.get_ceph_git_url()... - 12:07 PM Revision 28dc5c0c (ceph): workunit: keep the path to the list of tests in a variable
- Signed-off-by: Loic Dachary <loic@dachary.org>
- 11:34 AM Revision 7de33f08 (ceph): workunit: allow tag to override sha1
- The sha1 for the workunit task is always set by the suite.py task. The
tag must be checked before the sha1 othewise i... - 08:57 AM Bug #13991: osd: fix ScrubJob compare function
- If we not consider the deadline in compare function. Some pgs would not
be scheduled when the deadline arriv... - 08:38 AM Bug #13991: osd: fix ScrubJob compare function
- make a mistake, close it.
- 08:26 AM Bug #13991 (Won't Fix): osd: fix ScrubJob compare function
- If we not consider the deadline in compare function. Some pgs would not
be scheduled when the deadline arrives.
S...
12/05/2015
- 09:01 PM Bug #13970: ceph-disk list fails on /dev/cciss!c0d0
- nevermind, now I see the email - it comes from /dev/cciss/c0d0 being mapped to /sys/block/cciss!c0d0
- 08:35 PM Bug #13970: ceph-disk list fails on /dev/cciss!c0d0
- Is that exclamation point supposed to be there??
Look at http://man7.org/linux/man-pages/man4/cciss.4.html
<pre... - 06:27 PM Bug #13970: ceph-disk list fails on /dev/cciss!c0d0
A tentative workaround is to replace in /usr/sbin/ceph-disk...- 05:48 PM Bug #13970 (In Progress): ceph-disk list fails on /dev/cciss!c0d0
- 07:44 PM Bug #13990: Hammer (0.94.3) OSD does not delete old OSD Maps in a timely fashion (maybe at all?)
- One last comment for now. I have found this morning that restarting an OSD does trigger the cleanup. It just takes a...
- 05:25 PM Bug #13990: Hammer (0.94.3) OSD does not delete old OSD Maps in a timely fashion (maybe at all?)
- Ok.. Sorry for the spam, the last log snippet didn't quite show what I indicated, as it appears 1199 did finally star...
- 05:04 PM Bug #13990: Hammer (0.94.3) OSD does not delete old OSD Maps in a timely fashion (maybe at all?)
- Request entity too large... here's a snippet that I believe shows what I said in the bug:
Dec 5 09:23:32 str-slc-06... - 04:58 PM Bug #13990 (Resolved): Hammer (0.94.3) OSD does not delete old OSD Maps in a timely fashion (mayb...
- We recently upgraded from firefly 0.80.9 to Hammer 0.94.3. Since our upgrade 3 weeks ago we have accumulated over 25...
- 06:29 PM Messengers Bug #13985: Messenger: large message data content length cause int overflow
- https://github.com/ceph/ceph/pull/6809
- 10:44 AM Messengers Bug #13985 (In Progress): Messenger: large message data content length cause int overflow
- 07:02 AM Messengers Bug #13985 (Resolved): Messenger: large message data content length cause int overflow
- OSD aborted when sending message:
-1103> 2015-12-04 16:12:23.691914 7f9282bfe700 1 -- 172.19.10.1:6801/186389 --> 1... - 05:42 PM Bug #13975 (Pending Backport): systemd/ceph-disk@.service assumes /bin/flock
- 05:42 PM Revision a60a9c97 (ceph): Merge pull request #6803 from dachary/wip-13975-systemd-flock
- build/ops: systemd ceph-disk unit must not assume /bin/flock
Reviewed-by: Nathan Cutler <ncutler@suse.cz> - 05:11 PM Revision 59c7fa09 (ceph): Merge pull request #6811 from smithfarm/wip-submitting-jewel
- SubmittingPatches: there is no next; only jewel
Reviewed-by: Loic Dachary <ldachary@redhat.com> - 04:46 PM Bug #13989 (Duplicate): OSD boot fails with os/FileJournal.cc: 1907: FAILED assert(0)
- Looks a little like http://tracker.ceph.com/issues/13511...
- 04:40 PM Bug #13988 (Resolved): new OSD re-using old OSD id fails to boot
- Steps to reproduce...
- 04:34 PM Revision b23b92df (ceph): AsyncConnection: Fix potential return code overflow
- Signed-off-by: Haomai Wang <haomai@xsky.com>
- 04:34 PM Revision 3443b6d1 (ceph): Pipe: Fix potential return code overflow
- Signed-off-by: Haomai Wang <haomai@xsky.com>
- 04:20 PM Revision 1adf3063 (ceph): SubmittingPatches: there is no next; only jewel
- Signed-off-by: Nathan Cutler <ncutler@suse.com>
- 11:19 AM Revision c5ff6f60 (ceph): Pipe: Fix large message data content length causing overflow
- Fix #13985
Signed-off-by: Haomai Wang <haomai@xsky.com>
Signed-off-by: Jun Huang <hjwsm1989@gmail.com> - 09:37 AM rbd Bug #13969: TestLibRBD.ExclusiveLockTransition Journal.cc: 639: FAILED assert(m_events.empty())
- @Jason thanks for being so responsive :-)
- 09:36 AM rbd Bug #13969 (Resolved): TestLibRBD.ExclusiveLockTransition Journal.cc: 639: FAILED assert(m_events...
- 09:34 AM Bug #13986 (Resolved): TEST_scrub_snaps: timeout > 300
- http://jenkins.ceph.dachary.org/job/ceph/LABELS=centos-7&&x86_64/9864/consoleFull...
- 06:16 AM Feature #13441: change PG::do_command list_missing to iterate over missing_loc structure instead ...
- PR: https://github.com/ceph/ceph/pull/6298
Test: https://github.com/ceph/ceph-qa-suite/pull/755 - 06:15 AM Feature #13441: change PG::do_command list_missing to iterate over missing_loc structure instead ...
- 06:14 AM Feature #12754 (Resolved): pg auto repair for EC pool
- 01:36 AM Revision b0997bc5 (ceph): Merge pull request #6792 from smithfarm/wip-index-20151202
- Wip index 20151202
Reviewed-by: Loic Dachary <ldachary@redhat.com> - 01:31 AM Revision fb96b514 (ceph): Merge pull request #6795 from ceph/pybind-rados-exec
- pybind: support ioctx:exec
Reviewed-by: Josh Durgin <jdurgin@redhat.com>
12/04/2015
- 11:24 PM Revision 8160f9eb (ceph): Add common/PluginRegistry.cc to CMakeLists.txt
- Else this happens:
[ 15%] Linking CXX executable ceph-authtool
libcommon.a(ceph_context.cc.o): In function `Ceph... - 11:03 PM Revision 6fb44db9 (ceph): pg: add test cases for pg list_missing on EC pool
- Reviewed-by: Guang Yang <yguang@yahoo-inc.com>
- 10:58 PM Revision c1daf4e6 (ceph): doc/dev/index.rst: wholesale refactor
- Signed-off-by: Nathan Cutler <ncutler@suse.com>
- 10:56 PM Revision a62b5ac9 (ceph): pybind: support ioctx:exec
- Signed-off-by: Noah Watkins <noahwatkins@gmail.com>
- 09:40 PM Revision 371e897b (ceph): Removed CentOS 7.0 and added CentOS 7.1.1503 instead
- Signed-off-by: Yuri Weinstein <yweinste@redhat.com>
- 09:09 PM Revision 7f817288 (ceph): common/Makefile: ship common/event_socket.h
- Broken in commit 5eb64dd324a0e0ec609acf2b178c02dee901aa5d.
Signed-off-by: Sage Weil <sage@redhat.com> - 08:13 PM Bug #13975 (Fix Under Review): systemd/ceph-disk@.service assumes /bin/flock
- https://github.com/ceph/ceph/pull/6803
- 01:07 PM Bug #13975 (In Progress): systemd/ceph-disk@.service assumes /bin/flock
- 08:11 PM Revision c8f7d44c (ceph): build/ops: systemd ceph-disk unit must not assume /bin/flock
- The flock command may be installed elsewhere, depending on the
system. Let the PATH search figure that out.
http://t... - 07:15 PM Bug #13964 (Resolved): jewel: ceph_test_objectcacher_stress segfault
- Same as #13877. cherry-picked to jewel as commit:73aab5e93eb609f42fcc22b6a7f18d5415d02ef7
- 07:05 PM Bug #13964 (In Progress): jewel: ceph_test_objectcacher_stress segfault
- 07:11 PM Revision 73aab5e9 (ceph): test: use sequential journal_tid for object cacher test
- This matches the real usage by librbd.
Fixes: #13877
Signed-off-by: Josh Durgin <jdurgin@redhat.com>
(cherry picked ... - 06:30 PM Bug #13984 (Duplicate): ceph-disk prepare activates the osd on 7.1
- There are two ways to activate osd block device,
1) ceph-deploy prepare + ceph-deploy activate
or
2) ceph-... - 06:20 PM Bug #13878 (Resolved): test/librados/tier.cc doesn't completely clean up EC pools
- 06:19 PM Revision 014b8e25 (ceph): Merge pull request #6788 from dachary/wip-13878-ec-pool
- test/librados/test.cc: clean up EC pools' crush rules too
Reviewed-by: Dan Mick <dmick@redhat.com> - 05:41 PM Bug #13921 (Can't reproduce): Error connecting to cluster: TypeError
- 12:03 AM Bug #13921: Error connecting to cluster: TypeError
- after rebasing wip-sage-testing on master...
- 05:33 PM Revision 534bd2b5 (ceph): test/librados: increase timeout in wait_for_healthy
- I set this with unperturbed vstart clusters in mind.
The tests are sometimes run with thrashers enabled,
so things mi... - 05:20 PM Revision 9e07b8af (ceph): Merge pull request #6665 from trociny/wip-librbd-journal
- rbd: journal: configuration via conf, cli, api and some fixes
Reviewed-by: Josh Durgin <jdurgin@redhat.com> - 05:13 PM Revision a5d3370b (ceph): Merge pull request #6800 from dillaman/wip-13969
- librbd: fix known test case race condition failures
Reviewed-by: Josh Durgin <jdurgin@redhat.com>
Reviewed-by: Loic ... - 03:42 PM rbd Bug #13969 (Fix Under Review): TestLibRBD.ExclusiveLockTransition Journal.cc: 639: FAILED assert(...
- *master PR*: https://github.com/ceph/ceph/pull/6800
- 03:40 PM rbd Bug #13969 (In Progress): TestLibRBD.ExclusiveLockTransition Journal.cc: 639: FAILED assert(m_eve...
- 12:28 PM rbd Bug #13969: TestLibRBD.ExclusiveLockTransition Journal.cc: 639: FAILED assert(m_events.empty())
- http://jenkins.ceph.dachary.org/job/ceph/LABELS=centos-7&&x86_64/9832/console...
- 11:53 AM rbd Bug #13969: TestLibRBD.ExclusiveLockTransition Journal.cc: 639: FAILED assert(m_events.empty())
- Re-opening as it seems to re-surface since the fix with variants.
http://jenkins.ceph.dachary.org/job/ceph/LABELS=... - 03:39 PM Revision 6cbf128b (ceph): tests: wait for mocked requests to complete
- Fixes: #13969
Signed-off-by: Jason Dillaman <dillaman@redhat.com> - 03:37 PM Revision b91cd9d4 (ceph): Merge pull request #6641 from aclamk/wip-log-alloc-predictor
- common: log: predict log message buffer allocation size
Reviewed-by: Sage Weil <sage@redhat.com> - 03:35 PM Revision c5e53526 (ceph): mon: fix OSDMonitor::_check_become_tier cannot prevent client
- from setting multiple tiers in a particular way
Fixes:#13950
Signed-off-by: Mingxin Liu <mingxin.liu@kylin-cloud.com> - 03:34 PM Revision 89435f21 (ceph): Merge pull request #6642 from 0003ydlom/decode_reorder
- osd: move newest decode version of MOSDOp and MOSDOpReply to the front
Reviewed-by: Sage Weil <sage@redhat.com> - 03:25 PM Revision fde9f785 (ceph): librbd: do not complete AIO callbacks within caller's thread context
- Avoid rare, racy issues when individual requests associated with an AIO
completion can complete prior to marking the ... - 03:19 PM Revision 375137bf (ceph): Merge pull request #6696 from Ved-vampir/wip_common_plugin
- common: add generic plugin infrastructure
Reviewed-by: Varada Kari <varada.kari@sandisk.com> - 03:16 PM Bug #13944 (Resolved): librados assumes malloc(0) returns valid pointer
- not sure this is worth backporting.
- 03:15 PM Revision 8d5e05a5 (ceph): Merge pull request #6779 from ceph/wip-librados-xattr
- librados: avoid malloc(0) (which can return NULL on some platforms)
Reviewed-by: Sage Weil <sage@redhat.com> - 03:14 PM Revision 43268f20 (ceph): Merge pull request #6759 from liewegas/wip-bdev-probe
- osd: make backend and block device code a bit more generic
Reviewed-by: Loic Dachary <ldachary@redhat.com>
Reviewed-... - 03:13 PM Revision ff699d7d (ceph): Merge pull request #6686 from stiopaa1/namespace_buffer
- common: use namespace instead of subclasses for buffer
Reviewed-by: Sage Weil <sage@redhat.com> - 03:13 PM Revision 7643ce9c (ceph): Merge pull request #6535 from mslovy/wip-crc-cache-tracker
- common: buffer: add cached_crc and cached_crc_adjust counts to perf dump
Reviewed-by: Sage Weil <sage@redhat.com> - 03:10 PM Revision 97e74a1d (ceph): Merge pull request #6058 from guangyy/wip-pg-query
- osd: dump number of missing objects for each peer with pg query
Reviewed-by: Sage Weil <sage@redhat.com> - 03:09 PM Feature #13121 (Resolved): osd: add pool level setting for recovery priority?
- 03:09 PM Revision c3e3ee05 (ceph): Merge pull request #5953 from guangyy/wip-13121
- osd: support pool level recovery_priority and recovery_op_priority
Reviewed-by: David Zafman <dzafman@redhat.com> - 03:08 PM rbd Feature #13983 (New): Optionally no-op flush requests when journal + cache are enabled
- If both the journal and cache are enabled, it is safe to no-op flush requests since read-after-write consistency is p...
- 03:01 PM Bug #12789 (Resolved): Long tail latency due to journal aio io_submit takes long time to return?
- 03:00 PM Revision 673d8521 (ceph): Merge pull request #5670 from david-z/master
- osd: FileJournal: reduce locking scope in write_aio_bl
Reviewed-by: Haomai Wang <haomai@xsky.com> - 01:40 PM Revision 1fffb25e (ceph): Merge pull request #6798 from trociny/fix-rbd_mirror-bool
- librbd: include missing header for bool type
Reviewed-by: Jason Dillaman <dillaman@redhat.com> - 01:28 PM Bug #13982 (Won't Fix): tiering: target_max_bytes won't work as expected once you use 'rbd rm'
- When 'rbd rm' has been used, "rados df" reports the same objects as before the 'rm' while the used size is decreased ...
- 01:22 PM CephFS Bug #13977: ovh: sambatorture.scan-pipe_number fails on ENOMEM
- http://pulpito.ovh.sepia.ceph.com:8081/teuthology-2015-12-02_23:14:01-samba-master---basic-openstack/
- 03:46 AM CephFS Bug #13977 (New): ovh: sambatorture.scan-pipe_number fails on ENOMEM
- Open of pipe \WKSSVC failed with error (socket_error)
2015-12-03T18:22:43.418 INFO:teuthology.orchestra.run.target09... - 01:10 PM Revision 32597cab (ceph): librbd: include missing header for bool type
- Signed-off-by: Mykola Golub <mgolub@mirantis.com>
- 11:58 AM Cleanup #13981 (Resolved): configure script doesn't check bzip2-devel presence
- After cloning git master (state as for 4 Dec 2015), it is not possible to build Ceph from sources when bzip2-devel (o...
- 11:53 AM rbd Bug #13976 (Duplicate): TestMockOperationSnapshotProtectRequest.Success include/xlist.h: 77: FAIL...
- clustering with #13969
- 11:18 AM Revision e8a584ff (ceph): rbd: add new journal rbd commands
- Also, extend 'rbd info' to output journal spec if journaling is enabled.
Signed-off-by: Mykola Golub <mgolub@miranti... - 11:18 AM Revision e72fc02c (ceph): librbd: journal: mark entry committed after replay
- After replying a journal entry we have to call committed(). Otherwise,
the entries remain with flag committed=false i... - 11:18 AM Revision 89c716ff (ceph): librbd: output operator for image options
- It is going to be used for diagnostic.
Signed-off-by: Mykola Golub <mgolub@mirantis.com> - 11:18 AM Revision 47f56108 (ceph): librbd: journal options on image create/clone/copy
- Signed-off-by: Mykola Golub <mgolub@mirantis.com>
- 11:18 AM Revision 5622d3ff (ceph): rbd: alow to specify options for created journal
- Signed-off-by: Mykola Golub <mgolub@mirantis.com>
- 11:15 AM Revision 7248fb5a (ceph): librbd: new config options to tweak journal settings
- Signed-off-by: Mykola Golub <mgolub@mirantis.com>
- 11:15 AM Revision ebadfefa (ceph): librbd: debug: log journal metadata
- Signed-off-by: Mykola Golub <mgolub@mirantis.com>
- 11:15 AM Revision 61ad15c7 (ceph): librbd: allow alternate pool for journal objects
- Signed-off-by: Mykola Golub <mgolub@mirantis.com>
- 11:15 AM Revision bb10815f (ceph): librbd: add image pool ID to journal object names.
- The data objects are named "journal_data.<pool id>.<image id>.<num>".
This is to avoid conflict when a journal object... - 11:15 AM Revision 7311185a (ceph): librbd: humanize ImageWatcher and Journal states in debug log
- Signed-off-by: Mykola Golub <mgolub@mirantis.com>
- 08:16 AM Revision 77131e0b (ceph): Merge pull request #6129 from dillaman/wip-13292
- RBD pool mirroring configuration API + CLI
Reviewed-by: Josh Durgin <jdurgin@redhat.com> - 08:14 AM Revision e9317f00 (ceph): SubProcess: fix multiple definition bug
- In current implementation, the declare of the
SubProcess class was seperate from the
implementation, like this:
clas... - 07:45 AM CephFS Bug #13948 (Rejected): fs has no limit to PATH_MAX
- 05:53 AM CephFS Bug #13948: fs has no limit to PATH_MAX
- Zheng is right. I finally had a chance to re-test this on an idle filesystem and it actually completed. I think it wa...
- 02:09 AM CephFS Bug #13948: fs has no limit to PATH_MAX
- getcwd() is handle by VFS, nothing do with cephfs kernel driver. besides,it's not feasible to check how deep a direct...
- 07:42 AM CephFS Bug #13980 (Resolved): all nfs v3 mounts fail in ovh lab
- http://pulpito.ovh.sepia.ceph.com:8081/teuthology-2015-12-02_23:10:02-knfs-jewel-testing-basic-openstack/
http://pul... - 06:54 AM Backport #13979 (Resolved): infernalis : "shard missing" errors in logs during Teuthology run
- https://github.com/ceph/ceph/pull/6849
- 06:47 AM rbd Feature #13973: ls hangs in mounted folder (rbd)
- The reason is that all osds were full and cluster was rejecting updates, it would help if rbd command line tools retu...
- 05:44 AM Bug #13950 (Fix Under Review): OSDMonitor::_check_become_tier cannot prevent client from setting ...
- 01:48 AM Bug #13897 (Need More Info): (jewel) rados/objectstore/objectstore.yaml common/Mutex.cc: 105: FAI...
- Loic, do you have the path for the original run? The logs should help track this down.
- 01:33 AM Revision 02a9a41f (ceph): osd: Test osd_find_best_info_ignore_history_les config in another assert
- Signed-off-by: David Zafman <dzafman@redhat.com>
- 12:51 AM Bug #13939: PGs stuck in "creating" when created just after deleting another pool
- Nope, issue is still there, showed up again in next run.
http://pulpito.ceph.com/jspray-2015-12-03_04:11:11-fs:recov... - 12:38 AM Bug #13962 (Pending Backport): "shard missing" errors in logs during Teuthology run
- 12:38 AM Revision 9f18962a (ceph): Merge pull request #6778 from liewegas/wip-13962
- osd: call on_new_interval on newly split child PG
Reviewed-by: Samuel Just <sjust@redhat.com> - 12:08 AM Revision 04b4795f (ceph): test/librados/test.cc: clean up EC pools' crush rules too
- SetUp was adding an erasure-coded pool, which automatically adds
a new crush rule named after the pool, but only remo...
12/03/2015
- 11:57 PM Bug #13921: Error connecting to cluster: TypeError
- With wip-sage-testing...
- 12:07 PM Bug #13921: Error connecting to cluster: TypeError
- I don't see anything even remotely capable of causing the error in the commits from wip-sage-testing that were not al...
- 11:37 AM Bug #13921: Error connecting to cluster: TypeError
- ...
- 11:14 AM Bug #13921 (In Progress): Error connecting to cluster: TypeError
- 10:53 PM Linux kernel client Bug #13712: Page allocation failure in Kernel 4.2.3.
- We were not using snapshots before, then after the first corruption, I created a snapshot and a clone and if anything...
- 04:09 PM Linux kernel client Bug #13712: Page allocation failure in Kernel 4.2.3.
- > The data structure I mentioned above manages snapshots. Jérôme, Daniel, do
> you use rbd snapshots? More specifi... - 10:52 PM Revision 2fd95185 (ceph): Merge pull request #753 from ceph/wip_fix_fuse_hammer
- Fixed missing ceph-fuse
Reviewed-by: Josh Durgin <jdurgin@redhat.com> - 10:26 PM Revision ef9cdc56 (ceph): Fixed missing ceph-fuse
- Signed-off-by: Yuri Weinstein <yweinste@redhat.com>
- 09:56 PM rbd Bug #13976: TestMockOperationSnapshotProtectRequest.Success include/xlist.h: 77: FAILED assert(_s...
- @Jason I think it may be a false negative : https://github.com/ceph/ceph/pull/6440/files does not look like it could ...
- 09:56 PM rbd Bug #13976 (Duplicate): TestMockOperationSnapshotProtectRequest.Success include/xlist.h: 77: FAIL...
- First seen at http://jenkins.ceph.dachary.org/job/ceph/LABELS=centos-7&&x86_64/9820/console...
- 09:40 PM Revision 15a81bb7 (ceph): Merge pull request #6780 from smithfarm/wip-index-20151201
- doc/dev/index.rst: fix headings
Reviewed-by: Loic Dachary <ldachary@redhat.com> - 09:39 PM rbd Bug #13969 (Resolved): TestLibRBD.ExclusiveLockTransition Journal.cc: 639: FAILED assert(m_events...
- 09:38 PM rbd Bug #13969: TestLibRBD.ExclusiveLockTransition Journal.cc: 639: FAILED assert(m_events.empty())
- http://jenkins.ceph.dachary.org/job/ceph/LABELS=centos-7&&x86_64/9801/console
- 12:56 PM rbd Bug #13969 (Fix Under Review): TestLibRBD.ExclusiveLockTransition Journal.cc: 639: FAILED assert(...
- *master PR*: https://github.com/ceph/ceph/pull/6789
- 09:56 AM rbd Bug #13969: TestLibRBD.ExclusiveLockTransition Journal.cc: 639: FAILED assert(m_events.empty())
- https://github.com/ceph/ceph/pull/6665#issuecomment-161560916
- 09:53 AM rbd Bug #13969: TestLibRBD.ExclusiveLockTransition Journal.cc: 639: FAILED assert(m_events.empty())
- http://jenkins.ceph.dachary.org/job/ceph/LABELS=centos-7&&x86_64/9774/console
- 09:31 AM rbd Bug #13969: TestLibRBD.ExclusiveLockTransition Journal.cc: 639: FAILED assert(m_events.empty())
- @Jason, this happened today during make check on an unrelate pull request. If it does not ring a bell, feel free to u...
- 09:30 AM rbd Bug #13969 (Resolved): TestLibRBD.ExclusiveLockTransition Journal.cc: 639: FAILED assert(m_events...
- ...
- 09:33 PM CephFS Bug #13948 (New): fs has no limit to PATH_MAX
- I don't think that's an accurate assessment of the problem. Presumably this test passes on local FSes and fails for u...
- 09:36 AM CephFS Bug #13948 (Rejected): fs has no limit to PATH_MAX
- no other FS impose this limitation. PATH_MAX is max length of path buffer. For very deep directory trees, you still c...
- 09:16 PM Bug #13975 (Resolved): systemd/ceph-disk@.service assumes /bin/flock
- It may be elsewhere (/usr/bin/flock etc.)
- 08:34 PM Bug #13972 (Need More Info): osd/ECUtil.h: 117: FAILED assert(old_size == total_chunk_size) in 0....
- 06:49 PM Bug #13972: osd/ECUtil.h: 117: FAILED assert(old_size == total_chunk_size) in 0.80.10
- The log above only contains two writes on the object at offset 2096640~2096640 and 4193280~1024. If it recurs after ...
- 06:35 PM Bug #13972: osd/ECUtil.h: 117: FAILED assert(old_size == total_chunk_size) in 0.80.10
- 2015-12-03 09:53:41.845187 7f68aecb2700 10 osd.80 8518 dequeue_op 0x7f690866f3b0 prio 63 cost 2096640 latency 24.5771...
- 06:34 PM Bug #13972: osd/ECUtil.h: 117: FAILED assert(old_size == total_chunk_size) in 0.80.10
- After rolling the rgw nodes the cluster went back to active+clean and the osds aren't crashing anymore.
- 06:32 PM Bug #13972: osd/ECUtil.h: 117: FAILED assert(old_size == total_chunk_size) in 0.80.10
- The workload that was happening before the crash was a boto multipart upload. There was a cron job of about 24 batch...
- 06:14 PM Bug #13972: osd/ECUtil.h: 117: FAILED assert(old_size == total_chunk_size) in 0.80.10
- The ceph osd log is located here: ceph-post-file: 5023e3eb-eeba-472d-aeea-86022efdc83d
- 06:05 PM Bug #13972: osd/ECUtil.h: 117: FAILED assert(old_size == total_chunk_size) in 0.80.10
- Ceph.conf: https://gist.github.com/cholcombe973/5293a14d63d4d573d4c6
ceph df: https://gist.github.com/cholcombe973/c... - 05:34 PM Bug #13972 (Can't reproduce): osd/ECUtil.h: 117: FAILED assert(old_size == total_chunk_size) in 0...
- I have an erasure code pool setup with k=10,m=3 jerasure and ruleset-failure-domain=osd and I seem to be hitting a cr...
- 08:28 PM Bug #13894: RBD over cache tier over EC pool: rbd rm doesn't remove objects
- Samuel,
I have following setting for hitset only:
./ceph -c ceph.conf osd pool set cachepool hit_set_type bloom
... - 07:02 PM Bug #13894 (Need More Info): RBD over cache tier over EC pool: rbd rm doesn't remove objects
- Laurent, I agree that the size bug is a problem, can you open a new ticket specifically for that? This ticket is abo...
- 07:01 PM Bug #13894: RBD over cache tier over EC pool: rbd rm doesn't remove objects
- Igor: do you have hitsets configured? It looks like it always promotes if hitsets are disabled.
- 06:37 PM Bug #13894: RBD over cache tier over EC pool: rbd rm doesn't remove objects
- The issue we're facing is that the size reported by all ceph tools is completely wrong, from my ticket:...
- 06:25 PM Bug #13894: RBD over cache tier over EC pool: rbd rm doesn't remove objects
- More detailed logs with following debug settings enabled:
debug osd = 25
debug filestore = 20
debug ms = 1 - 06:24 PM Bug #13894: RBD over cache tier over EC pool: rbd rm doesn't remove objects
- Please find appropriate logs attached too.
IMO this shows object promotion is still present in master branch.
- 06:23 PM Bug #13894: RBD over cache tier over EC pool: rbd rm doesn't remove objects
- I made some investigations on the topic in current master. Here are my observations.
I added some additional loggi... - 08:16 PM CephFS Feature #13974 (New): MDS: mon: expose state of log flushing
- We warn users when their MDS isn't logging, but we don't seem to expose much state about why that is. Do so.
- 06:43 PM rbd Feature #13973: ls hangs in mounted folder (rbd)
- after that ...
- 06:37 PM rbd Feature #13973 (Rejected): ls hangs in mounted folder (rbd)
- ls (and all other file system operations) hang forever in the folder mounted with rbd.
Context:
I've mounted a fo... - 06:22 PM Revision 6c743813 (ceph): test: fix linkage on rbd/journal tests
- These weren't explicitly linking librados,
so were prone to missing symbols on changes
in librados_test
Signed-off-b... - 05:41 PM Revision 46493d4e (ceph): Merge pull request #6786 from oritwas/wip-cmake-fix-radosgw
- cmake: add rgw_basic_types.cc to librgw.a
Reviewed-by: Yehuda Sadeh <yehuda@redhat.com> - 05:31 PM Revision 0857da4c (ceph): Merge pull request #6767 from oritwas/wip-13529-jewel
- rgw: use smart pointer for C_Reinitwatch
Reviewed-by: Yehuda Sadeh <yehuda@redhat.com> - 05:11 PM Revision 6807b35a (ceph): doc/dev/index.rst: fix links
- Signed-off-by: Nathan Cutler <ncutler@suse.com>
- 05:11 PM Revision f3366401 (ceph): doc/dev/index.rst: Flesh out the IRC section
- Signed-off-by: Nathan Cutler <ncutler@suse.com>
- 05:11 PM Revision 8d281bc5 (ceph): doc/dev/index.rst: Make Issue tracker a separate chapter
- Signed-off-by: Nathan Cutler <ncutler@suse.com>
- 03:59 PM Bug #13511: core dump during osd start
- See similar in run:
http://pulpito.ovh.sepia.ceph.com:8081/teuthology-2015-12-02_20:55:02-rados-hammer-distro-basic-... - 03:54 PM Bug #13971 (Can't reproduce): "test/multi_stress_watch.cc: 61: FAILED assert(!ret)" in rados-hamm...
- Run: http://pulpito.ovh.sepia.ceph.com:8081/teuthology-2015-12-02_20:55:02-rados-hammer-distro-basic-openstack/
Job:... - 02:57 PM Revision 138f5849 (ceph): osdc/Objecter: new-style pgls
- Signed-off-by: John Spray <john.spray@redhat.com>
Signed-off-by: Sage Weil <sage@redhat.com> - 02:57 PM Revision c3524d7b (ceph): librados: new style pgls
- Signed-off-by: John Spray <john.spray@redhat.com>
Signed-off-by: Sage Weil <sage@redhat.com> - 02:57 PM Revision 8b7b23db (ceph): test/librados: util to set pg_num
- Because some unit tests benefit from having a
non-power-of-two pg_num.
Signed-off-by: John Spray <john.spray@redhat.... - 02:57 PM Revision 5555f9f4 (ceph): test: tests for new-style pgls
- Signed-off-by: John Spray <john.spray@redhat.com>
Signed-off-by: Sage Weil <sage@redhat.com> - 02:57 PM Revision d70d3c10 (ceph): osdc/Objecter: direct enumerate request at hash, not pg
- The existing request machinery already maps the hash to a pg; use
it. Among other things, it means that a split will... - 02:57 PM Revision dbc489c8 (ceph): librados: update naming for new object list methods
- Signed-off-by: John Spray <john.spray@redhat.com>
- 02:57 PM Revision 532254aa (ceph): librados: C++ bindings for new object_list methods
- Signed-off-by: John Spray <john.spray@redhat.com>
- 02:57 PM Revision 60361457 (ceph): osdc/Objecter: end of pg if next pointer is MAX, not 0 items
- It is possible in pathological cases for the OSD to return 0
items with a non-MAX next value (indicating the objecter... - 02:57 PM Revision fc8ea3f2 (ceph): osd/osd_types: add pg_t::get_hobj_{start,end} methods
- Convenience methods to easily get the bounds of the object range
that are managed by a given PG. Note that this is a... - 02:57 PM Revision 3ee77318 (ceph): osd: set handle to start of next PG on pgnls end of pg
- For the namespaced PGNLS op, set the handle to the start
of the next PG when we reach the end of the current PG.
We ... - 02:55 PM Revision 42c6ef44 (ceph): Merge pull request #6789 from dillaman/wip-13969
- librbd: partial revert of commit 9b0e359
Reviewed-by: Josh Durgin <jdurgin@redhat.com> - 02:40 PM Bug #13939: PGs stuck in "creating" when created just after deleting another pool
- Updated: after rebasing test branch on latest master I saw a run where we didn't hit the issue, so either fixed or in...
- 02:27 PM Backport #12949: Race condition in rados bench
- Created https://github.com/ceph/ceph/pull/6791 which replaces https://github.com/ceph/ceph/pull/5812.
- 02:23 PM Revision ed6792ba (ceph): Merge pull request #6748 from liewegas/wip-fix-cephtool
- qa/workunits/cephtool/test.sh: no ./
Reviewed-by: Loic Dachary <ldachary@redhat.com> - 02:10 PM Revision 72691dd9 (ceph): Merge pull request #5941 from jcsp/wip-12137
- #12137: cephfs-data-scan scan_frags
- 01:38 PM Revision 157ef2e8 (ceph): Merge pull request #6777 from dillaman/wip-librbd-lockdep
- librbd: journaling-related lock dependency cleanup
Reviewed-by: Josh Durgin <jdurgin@redhat.com> - 01:26 PM Revision e2b7bc52 (ceph): Merge pull request #6762 from dillaman/wip-13924
- journal: fire replay complete event after reading last object
Reviewed-by: Josh Durgin <jdurgin@redhat.com> - 01:23 PM Revision 111f6a6f (ceph): Merge pull request #6703 from dillaman/wip-tests-librbd-gmock-state-mac...
- librbd: initial collection of state machine unit tests
Reviewed-by: Josh Durgin <jdurgin@redhat.com> - 01:18 PM Revision c2c6d025 (ceph): tools: fix race condition in seq/rand bench
- Under certain conditions (like bench seq/rand -b 1024 -t 128) it is
possible that aio_read reads data into destinatio... - 01:18 PM Revision a619b621 (ceph): tools: add --no-verify option to rados bench
- When doing seq and rand read benchmarks using rados bench, a quite large
portion of cpu time is consumed by doing obj... - 12:55 PM Revision 91f01bdb (ceph): librbd: partial revert of commit 9b0e359
- Fixes: #13969
Signed-off-by: Jason Dillaman <dillaman@redhat.com> - 12:40 PM Revision 4e15c039 (ceph): pybind/rbd.pyx: Return something useful from str() and repr()
- str() defaults to repr(), so implement __repr__ instead.
Signed-off-by: Hector Martin <marcan@marcan.st> - 12:31 PM Revision 6e7f9c8e (ceph): doc/dev/index.rst: start writing Bugfixing chapter
- Signed-off-by: Nathan Cutler <ncutler@suse.com>
- 12:31 PM Bug #13878: test/librados/tier.cc doesn't completely clean up EC pools
- continuing at https://github.com/ceph/ceph/pull/6788
- 12:14 PM Bug #13453: backfill pushes do not appear to increment the recovery rate counters
- Sam included it yesterday in his testing branch.
- 12:14 PM Bug #13857 (Need More Info): tests: make check timeout occasionaly
- 11:40 AM Revision 91e4a1c7 (ceph): Merge pull request #6690 from branch-predictor/bp-smaller-pglog-2
- osd: slightly reduce actual size of pg_log_entry_t
Reviewed-by: Sage Weil <sage@redhat.com> - 11:40 AM Revision a93fdb6e (ceph): Merge pull request #6689 from 0003ydlom/doubled_cond
- osd: avoid duplicate op->mark_started in ReplicatedBackend
Reviewed-by: xinxin shu <xinxin.shu@intel.com> - 11:39 AM Bug #13516 (Resolved): ObjectStore_Bench failed to rerun
- 11:39 AM Revision 71e172fb (ceph): Merge pull request #6681 from ifed01/wip-objstore-bench-stabilization
- ceph_objectstore_bench: fix race condition, bugs
Reviewed-by: Kefu Chai <kchai@redhat.com> - 11:38 AM Revision c80945fb (ceph): Merge pull request #6677 from jack-changtao/test-bench
- rados: bench: fix off-by-one to avoid writing past object_size
Reviewed-by: Kefu Chai <kchai@redhat.com> - 11:37 AM Revision eb82a852 (ceph): Merge pull request #6555 from mslovy/wip-reduce-omap-transaction
- osd: don't update epoch and rollback_info objects attrs if there is no need
Reviewed-by: Sage Weil <sage@redhat.com> - 10:58 AM Revision b3281316 (ceph): cmake: add rgw_basic_types.cc to librgw.a
- Signed-off-by: Orit Wasserman <owasserm@redhat.com>
- 10:00 AM Bug #13970 (Resolved): ceph-disk list fails on /dev/cciss!c0d0
- Reported by Stolte, Felix...
- 09:28 AM Revision 722705b8 (ceph): Merge pull request #6773 from ceph/wip-11104-firefly-add-epoch
- Revert "Revert "ceph.spec.: add epoch""
Reviewed-by: Boris Ranto <branto@redhat.com> - 08:53 AM Revision 002d26b0 (ceph): doc/dev/index.rst: rewrite introductory material
- Signed-off-by: Nathan Cutler <ncutler@suse.com>
- 06:52 AM Revision e4218a2d (ceph): KeyValueStore: Don't queue NULL context under test purpose
- Fix: #13961
Signed-off-by: Haomai Wang <haomai@xsky.com> - 05:52 AM Revision 39ba840f (ceph): Merge pull request #6784 from ceph/wip-5073
- rgw multitenancy
Buckets and users can now have a tenant associated with them. Buckets can be specified as [<tenant>... - 03:22 AM Revision 341e3f95 (ceph): osd: do evict after flush
- When the evict mode is not idle, after successfully flushing an object
to the base pool, we can try to evict it. We d... - 02:59 AM Bug #13968 (Rejected): OSD:lock PG when get_pg_or_queue_for_pg method successfully return a PG
- Otherwise caller shall race with other PG users, especially with PG removal.
Also, simplify wake_pg_waiters method a... - 02:16 AM Revision ba346bf4 (ceph): qa: add a test for the recovery_priority/recovery_op_priority pool sett...
- Signed-off-by: Guang Yang <yguang@yahoo-inc.com>
- 02:10 AM Revision f0cc7220 (ceph): mon: support recovery_priority and recovery_op_priority with monitor co...
- Fixes: 13121
Signed-off-by: Guang Yang yguang@yahoo-inc.com - 01:49 AM Bug #13967 (Resolved): PG::choose_acting valgrind error or ./common/hobject.h: 182: FAILED assert...
- <error>
<unique>0x0</unique>
<tid>47</tid>
<kind>UninitCondition</kind>
<what>Conditional jump or move de... - 01:31 AM Revision cd2bc413 (ceph): Objecter: clean up Objecter.h/ObjectOperation
- Signed-off-by: Jie Wang <jie.wang@kylin-cloud.com>
- 01:20 AM Bug #13965: jewel: osd/PGLog.h: 88: FAILED assert(rollback_info_trimmed_to == head)
- 12:38 AM Bug #13965 (Resolved): jewel: osd/PGLog.h: 88: FAILED assert(rollback_info_trimmed_to == head)
- -3> 2015-11-30 23:08:38.896573 7f6339553700 15 journal prepare_single_write 1 will write 34803712 : seq 46551 len...
- 01:20 AM Revision 053ee919 (ceph): PGLog::rewind_divergent_log: fix rollback_info_trimmed_to before index()
- Fixes: #13965
Signed-off-by: Samuel Just <sjust@redhat.com> - 01:17 AM Revision 66c72464 (ceph): TestPGLog: add test for 13965
- Signed-off-by: Samuel Just <sjust@redhat.com>
- 01:02 AM Bug #13966 (Resolved): memory leak in PG/replica_scrub
- PG::replica_scrub
if (last_update_applied < msg->scrub_to) {
dout(10) << "waiting for last_update_applied t...
Also available in: Atom