Casey Bodley's activity
From 03/25/2019 to 04/23/2019
04/23/2019
04/22/2019
- 08:40 PM rgw Bug #22162: add roles_pool json decode and encode in RGWZoneParams
- 03:23 PM rgw Bug #39393: inefficient unordered bucket listing
- 03:20 PM Ceph Revision d37d0339 (ceph): rgw: cls_bucket_list_unordered lists a single shard
- CLSRGWIssueBucketList sends the request to every shard, but this loop
intended to list only the current_shard
Fixes:... - 03:20 PM Ceph Revision cd1fc96c (ceph): cls/rgw: expose cls_rgw_bucket_list_op for single shard
- Signed-off-by: Casey Bodley <cbodley@redhat.com>
04/20/2019
- 12:50 AM rgw Bug #39393 (Resolved): inefficient unordered bucket listing
- RGWRados::cls_bucket_list_unordered() loops over bucket index shards, but each loop uses CLSRGWIssueBucketList to sen...
04/18/2019
- 06:02 PM rgw Bug #39287 (Pending Backport): multisite: period pusher gets 403 Forbidden against other zonegroups
- 06:01 PM rgw Backport #23237 (Closed): Corrupted downloads from civetweb when using multipart with slow connec...
- i see this as a client bug more than anything - aria2c does not retry or reconnect on errors, and reports 'success' e...
- 05:16 PM Ceph Revision ac3d1a02 (ceph): rgw: move mdlog trimming into rgw_trim_mdlog.cc
- Signed-off-by: Casey Bodley <cbodley@redhat.com>
- 05:16 PM Ceph Revision e9628164 (ceph): rgw: move datalog trimming into rgw_trim_datalog.cc
- Signed-off-by: Casey Bodley <cbodley@redhat.com>
- 05:16 PM Ceph Revision 4184edfa (ceph): rgw: move rgw_sync_log_trim.* to rgw_trim_bilog.*
- Signed-off-by: Casey Bodley <cbodley@redhat.com>
- 04:42 PM Dashboard Feature #39369: mgr/dashboard: show RGW multi-site sync status info
- the intent is for rgw to provide a new admin api that returns a json representation of the information currently avai...
- 01:47 PM rgw Bug #39283 (Pending Backport): rgw multisite: bucket index logs do not get trimmed unless zones '...
- 12:42 AM Ceph Revision 4f813d2f (ceph): rgw: limit entries in remove_olh_pending_entries()
- If there are too many entries to send in a single osd op, the osd rejects
the request with EINVAL. This error happens...
04/17/2019
04/16/2019
- 05:14 PM Dashboard Bug #39295 (Fix Under Review): mgr/dashboard: RGW Bucket API should provide times in UTC that wil...
- discussed this in the rgw standup, and consensus was to return UTC unconditionally. i made this change in https://git...
- 02:41 PM Dashboard Bug #39295: mgr/dashboard: RGW Bucket API should provide times in UTC that will be converted into...
- the call path for the 'GET /admin/bucket' api looks like this:
rgw_rest_bucket.cc: RGWOp_Bucket_Info::execute()
r... - 05:10 PM Ceph Revision fd15ab09 (ceph): rgw: bucket stats report mtime in UTC
- Fixes: https://tracker.ceph.com/issues/39295
Signed-off-by: Casey Bodley <cbodley@redhat.com>
04/13/2019
- 05:06 PM Ceph Revision 59b9dad1 (ceph): rgw: silence a -Wunused-function warning in pubsub
- Signed-off-by: Casey Bodley <cbodley@redhat.com>
- 03:44 PM rgw Bug #39287 (Fix Under Review): multisite: period pusher gets 403 Forbidden against other zonegroups
- 03:33 PM rgw Bug #39287 (Resolved): multisite: period pusher gets 403 Forbidden against other zonegroups
- RGWPeriodPusher is responsible for sending new period configurations from the master zonegroup to other zonegroups, a...
- 03:34 PM Ceph Revision f97be8cb (ceph): rgw: RGWPeriodPusher uses zone system key for inter-zonegroup messages
- RGWPeriodPusher was using an empty RGWAccessKey for inter-zonegroup
messages, which were rejected as an anonymous use...
04/12/2019
- 09:06 PM rgw Bug #39283 (Fix Under Review): rgw multisite: bucket index logs do not get trimmed unless zones '...
- 06:48 PM rgw Bug #39283 (Resolved): rgw multisite: bucket index logs do not get trimmed unless zones 'sync_fro...
- BucketTrimInstanceCR fetches the sync status of all peer zones in its zonegroup, and only trims up to the minimum of ...
- 08:36 PM Ceph Revision 3d52f30e (ceph): rgw: multisite log trimming only checks peers that sync from us
- if there are peer zones in our zonegroup that do not sync from us, they
won't have any sync status that corresponds t... - 07:56 PM Ceph Revision ded54895 (ceph): rgw: use std::optional in bilog trim
- clean up an ugly workaround for a bug in boost::optional
Signed-off-by: Casey Bodley <cbodley@redhat.com> - 04:29 PM rgw Backport #39280 (In Progress): nautilus: rgw: dynamic large objects fail uploading manifest with ...
- https://github.com/ceph/ceph/pull/27561
merged - 04:26 PM rgw Backport #39280 (Resolved): nautilus: rgw: dynamic large objects fail uploading manifest with eta...
- https://github.com/ceph/ceph/pull/27561
- 04:27 PM Ceph Revision 64abb12e (ceph): rgw: don't recalculate etags for slo/dlo
- for slo, we've already calculated this as 'lo_etag' in get_params()
for dlo, the local 'etag' already contains the ha... - 04:25 PM rgw Bug #39160 (Pending Backport): rgw: dynamic large objects fail uploading manifest with etag with ...
04/11/2019
- 05:54 PM rgw Backport #38847 (In Progress): mimic: rgw admin api create user generates a segfault and coredump...
- 05:52 PM rgw Bug #39141 (Closed): sync_user failed
- these error messages don't indicate an actual failure; it's just recognizing that a new user does not have any bucket...
- 05:38 PM rgw Bug #38199 (Triaged): rgw: setacl on versioned object doesn't write correct bi log entries
- 05:31 PM rgw Bug #38638 (Pending Backport): S3 policy evaluated incorrectly
04/10/2019
- 09:30 PM Ceph Revision d637937d (ceph): rgw: data sync checks empty next_marker for datalog
- RGWReadRemoteDataLogShardCR tracks the marker and next_marker
separately, because next_marker will be empty when it r...
04/09/2019
- 07:34 PM rgw Bug #39160 (Fix Under Review): rgw: dynamic large objects fail uploading manifest with etag with ...
- 07:34 PM rgw Bug #39160: rgw: dynamic large objects fail uploading manifest with etag with nautilus
- i was able to reproduce this one. it looks like a regression from the switch to openssl for md5 hashes - reverting 13...
- 07:26 PM Ceph Revision 4b849a49 (ceph): rgw: don't recalculate etags for slo/dlo
- for slo, we've already calculated this as 'lo_etag' in get_params()
for dlo, the local 'etag' already contains the ha...
04/08/2019
- 03:08 PM rgw Bug #39142 (Resolved): multisite: test_versioned_object_incremental_sync sometimes fails in three...
- The three-zone multisite tests sometime fail to converge on the same version. This shows up in one of two failures:
...
04/05/2019
- 07:26 PM Ceph Revision 1cf90269 (ceph): rgw: move rgw_init_ioctx() to rgw_tools.cc
- move rgw_init_ioctx() out of rgw_rados.cc so that it doesn't need to be
duplicated by services/svc_rados.cc
Signed-o... - 06:32 PM Ceph Revision 7e6d5323 (ceph): vstart: only add --debug-ms=1 in RGWDEBUG
- Signed-off-by: Casey Bodley <cbodley@redhat.com>
- 04:02 PM rgw Bug #39118 (Pending Backport): rgw: remove_olh_pending_entries() does not limit the number of xat...
- 01:51 PM rgw Bug #39118: rgw: remove_olh_pending_entries() does not limit the number of xattrs to remove
- 01:44 PM rgw Bug #39118 (Fix Under Review): rgw: remove_olh_pending_entries() does not limit the number of xat...
- 01:56 PM rgw Backport #38081 (Resolved): luminous: multisite: overwrites in versioning-suspended buckets fail ...
- 01:26 PM Ceph Revision 3805ea63 (ceph): rgw: limit entries in remove_olh_pending_entries()
- If there are too many entries to send in a single osd op, the osd rejects
the request with EINVAL. This error happens...
04/04/2019
- 09:01 PM rgw Bug #39118: rgw: remove_olh_pending_entries() does not limit the number of xattrs to remove
- The bug in https://tracker.ceph.com/issues/37792 is one way to produce massive amounts of pending olh entries.
- 08:52 PM rgw Bug #39118 (Resolved): rgw: remove_olh_pending_entries() does not limit the number of xattrs to r...
- RGWRados::remove_olh_pending_entries() tries to remove all stale pending entries, but if there are too many entries t...
- 08:54 PM rgw Bug #39038: beast: multiple v4 and v6 endpoints with the same port will cause failure
- i've been backporting other beast features (ie ssl support) to luminous anyway to encourage early adoption - tagged f...
04/03/2019
- 03:01 AM Ceph Revision 2853bf5b (ceph): rgw: don't crash on missing /etc/mime.types
- lack of mime types is not a fatal error. when a Content-Type header
is not provided in swift's PutObj, it uses this m... - 12:29 AM Ceph Revision 583f9009 (ceph): rgw: don't crash on missing /etc/mime.types
- lack of mime types is not a fatal error. when a Content-Type header
is not provided in swift's PutObj, it uses this m...
04/02/2019
- 11:08 PM Ceph Revision b825870f (ceph): rgw: don't crash on missing /etc/mime.types
- lack of mime types is not a fatal error. when a Content-Type header
is not provided in swift's PutObj, it uses this m... - 04:53 PM rgw Feature #39084 (Fix Under Review): ability to control user op mask via admin apis
- 04:18 PM rgw Bug #38638: S3 policy evaluated incorrectly
- 07:46 AM Ceph Revision 258f3abc (ceph): rgw: fixup_range() clamps offsets to valid part range
- Signed-off-by: Adam Kupczyk <akupczyk@redhat.com>
Signed-off-by: Casey Bodley <cbodley@redhat.com>
(cherry picked fro...
04/01/2019
- 05:15 PM rgw Backport #39073 (In Progress): luminous: multisite: data sync loops back to the start of the data...
- https://github.com/ceph/ceph/pull/27299
- 04:57 PM rgw Backport #39073 (Resolved): luminous: multisite: data sync loops back to the start of the datalog...
- https://github.com/ceph/ceph/pull/27299
- 05:07 PM Ceph Revision a537029a (ceph): rgw: data sync checks empty next_marker for datalog
- RGWReadRemoteDataLogShardCR tracks the marker and next_marker
separately, because next_marker will be empty when it r... - 04:57 PM rgw Backport #39075 (Resolved): nautilus: multisite: data sync loops back to the start of the datalog...
- https://github.com/ceph/ceph/pull/27498
- 04:57 PM rgw Backport #39074 (Resolved): mimic: multisite: data sync loops back to the start of the datalog af...
- https://github.com/ceph/ceph/pull/29021
- 04:56 PM rgw Bug #39033 (Pending Backport): multisite: data sync loops back to the start of the datalog after ...
- 04:54 PM rgw Bug #38700: silent corruption using SSE-C on multi-part upload to S3 with non-default part size
- test cases in https://github.com/ceph/s3-tests/pull/266 can be backported as well
- 03:27 PM rgw Bug #38700 (Pending Backport): silent corruption using SSE-C on multi-part upload to S3 with non-...
- 03:58 PM Ceph Revision 325b3584 (ceph): rgw: fixup_range() clamps offsets to valid part range
- Signed-off-by: Adam Kupczyk <akupczyk@redhat.com>
Signed-off-by: Casey Bodley <cbodley@redhat.com>
(cherry picked fro... - 03:58 PM Ceph Revision 67f2b9cf (ceph): rgw: decrypt filter does not cross multipart boundaries
- multipart uploads with sse encrypts each part separately, using an
initialization vector based on the part offset
de... - 03:56 PM Ceph Revision 0a1c2bb1 (ceph): rgw: decrypt filter does not cross multipart boundaries
- multipart uploads with sse encrypts each part separately, using an
initialization vector based on the part offset
de... - 03:56 PM Ceph Revision d1bfa9d1 (ceph): rgw: fixup_range() clamps offsets to valid part range
- Signed-off-by: Adam Kupczyk <akupczyk@redhat.com>
Signed-off-by: Casey Bodley <cbodley@redhat.com>
(cherry picked fro... - 03:48 PM Ceph Revision c72b438d (ceph): rgw: decrypt filter does not cross multipart boundaries
- multipart uploads with sse encrypts each part separately, using an
initialization vector based on the part offset
de...
03/29/2019
- 06:18 PM Ceph Revision d0142a18 (ceph): DNM qa/rgw: disable debuginfo packages
- Signed-off-by: Casey Bodley <cbodley@redhat.com>
- 05:22 PM rgw Bug #39000 (Pending Backport): rgw: beast endpoint doesn't set a default port
- 04:26 PM rgw Documentation #39007 (Pending Backport): rgw: update resharding documentation
- 03:19 PM rgw Bug #39033 (Fix Under Review): multisite: data sync loops back to the start of the datalog after ...
- 03:06 PM rgw Bug #39033: multisite: data sync loops back to the start of the datalog after reaching the end
- this leads to the following behavior after multisite tests:...
- 02:29 PM rgw Bug #39033 (Resolved): multisite: data sync loops back to the start of the datalog after reaching...
- RGWDataSyncShardCR was changed in https://github.com/ceph/ceph/pull/23596 to no longer call RGWReadRemoteDataLogShard...
- 03:12 PM Ceph Revision 0f5065d2 (ceph): librados/asio: add async_notify() to wrap aio_notify()
- Signed-off-by: Casey Bodley <cbodley@redhat.com>
- 03:12 PM Ceph Revision 2fd33fec (ceph): rgw: add rgw_rados_notify() with optional_yield
- Signed-off-by: Casey Bodley <cbodley@redhat.com>
- 03:12 PM Ceph Revision c33eee4e (ceph): rgw: RGWSI_RADOS::Obj::notify() takes optional_yield
- Signed-off-by: Casey Bodley <cbodley@redhat.com>
- 03:12 PM Ceph Revision 15ec2fa3 (ceph): rgw: add optional_yield to RGWSI_SysObj_Core interface
- Signed-off-by: Casey Bodley <cbodley@redhat.com>
- 03:12 PM Ceph Revision 9e742a6b (ceph): rgw: add optional_yield to RGWSI_SysObj_Cache
- Signed-off-by: Casey Bodley <cbodley@redhat.com>
- 03:12 PM Ceph Revision 38ac4d17 (ceph): rgw: add optional_yield to RGWSI_SysObj
- Signed-off-by: Casey Bodley <cbodley@redhat.com>
- 03:12 PM Ceph Revision b16e5f3d (ceph): rgw: add optional_yield to RGWSI_Notify
- Signed-off-by: Casey Bodley <cbodley@redhat.com>
- 03:12 PM Ceph Revision 41806015 (ceph): rgw: callers pass null_yield to RGWSI_SysObj
- Signed-off-by: Casey Bodley <cbodley@redhat.com>
- 02:29 PM Ceph Revision 2ff63778 (ceph): rgw: data sync checks empty next_marker for datalog
- RGWReadRemoteDataLogShardCR tracks the marker and next_marker
separately, because next_marker will be empty when it r... - 03:39 AM Ceph Revision e0a2d1b8 (ceph): rgw: RGWHTTPClient::wait() warns if it blocks in an asio thread
- Signed-off-by: Casey Bodley <cbodley@redhat.com>
- 03:39 AM Ceph Revision 848f20cb (ceph): rgw: http interfaces take optional_yield
- Signed-off-by: Casey Bodley <cbodley@redhat.com>
03/28/2019
- 06:05 PM rgw Bug #37767 (Duplicate): librgw crash due to local variables deallocated
- fixed by https://github.com/ceph/ceph/pull/26981, with backports being tracked in https://tracker.ceph.com/issues/38769
- 06:01 PM rgw Bug #37754 (Need More Info): bucket metadata not deleted after placement and bucket deleted
- Is this a multisite configuration? Bucket metadata stays around so that sync can complete in all zones. This cleanup ...
- 05:51 PM rgw Bug #38119 (Triaged): rgw can't create bucket, because can't find zonegroup? location constraint ...
- The request to create a bucket is comparing the LocationConstraint against the zonegroup's api_name. It looks like yo...
- 05:31 PM rgw Bug #38700: silent corruption using SSE-C on multi-part upload to S3 with non-default part size
03/27/2019
- 06:02 PM Ceph Revision 4b3bac20 (ceph): rgw: fixup_range() clamps offsets to valid part range
- Signed-off-by: Adam Kupczyk <akupczyk@redhat.com>
Signed-off-by: Casey Bodley <cbodley@redhat.com>
03/26/2019
- 11:55 AM Ceph Revision b7829029 (ceph): rgw: decrypt filter does not cross multipart boundaries
- multipart uploads with sse encrypts each part separately, using an
initialization vector based on the part offset
de...
03/25/2019
- 08:42 PM rgw Bug #38938 (Pending Backport): multisite: sync status on master zone does not show "oldest increm...
- 08:39 PM rgw Bug #38938 (Resolved): multisite: sync status on master zone does not show "oldest incremental ch...
- 05:34 PM rgw Bug #38002 (Resolved): fix http 416(InvalidRange) when upload a super large file
- 03:32 PM rgw Bug #38722 (Pending Backport): rgw: fix RGWDeleteMultiObj::verify_permission
- 03:11 PM Ceph Revision 511cd89e (ceph): qa/rgw: point test repos at ceph-nautilus branch
- Signed-off-by: Casey Bodley <cbodley@redhat.com>
Also available in: Atom