Activity
From 07/20/2016 to 08/18/2016
08/18/2016
- 10:59 PM Bug #17076 (Resolved): AWS S3 Version 4 signatures fail sometimes.
- When you use Version 4 signatures from Java AWS client or from s3cmd some file transfers fail with a Signature Failur...
- 10:18 PM Bug #17051: period commit loses zonegroup changes: region_map converted repeatedly
- cbodley and I traced this down:
RGWRados::convert_regionmap runs every time RGWRados::init_complete runs, but does N... - 05:47 PM Bug #17047 (Fix Under Review): rgw multisite: doesn't retry RGWFetchAllMetaCR on failed lease
- https://github.com/ceph/ceph/pull/10778
- 05:47 PM Bug #17073 (Fix Under Review): multisite: StateBuildingFullSyncMaps doesn't check for errors befo...
- https://github.com/ceph/ceph/pull/10779
- 05:26 PM Bug #17073 (Resolved): multisite: StateBuildingFullSyncMaps doesn't check for errors before advan...
- During the StateBuildingFullSyncMaps state of RGWDataSyncCR, we try to fetch the list of bucket instances from the re...
- 05:59 AM Bug #17040: Files incorrectly uploaded with swift api
- I have found out what is causing behaviour described above:...
08/17/2016
- 07:46 PM Backport #17064 (Resolved): jewel: rgw: radosgw daemon core when reopen logs
- https://github.com/ceph/ceph/pull/10868
- 06:59 PM Bug #17051: period commit loses zonegroup changes: region_map converted repeatedly
- Reproduction instructions:...
- 06:40 PM Bug #17036 (Pending Backport): rgw: radosgw daemon core when reopen logs
- 05:19 PM Backport #16599 (Resolved): jewel: rgw: Swift API returns double space usage and objects of accou...
- 04:43 PM Backport #16182 (Resolved): jewel: backport static sites fixes master->jewel
- 04:17 PM Backport #17006 (In Progress): jewel: Increase log level for messages occuring while running rgw ...
- 04:15 PM Backport #17032 (In Progress): jewel: multisite: RGWPeriodPuller tries to pull from itself
- 04:14 PM Backport #17033 (In Progress): jewel: rgw: there may be some objects not delete in some circumsta...
- 04:11 PM Backport #17034 (In Progress): jewel: rgw: object expirer's hints might be trimmed without proces...
- 03:40 PM Bug #16126 (Resolved): selinux denials in RGW
- 11:54 AM Backport #16312 (Resolved): jewel: selinux denials in RGW
- 08:15 AM Bug #17040: Files incorrectly uploaded with swift api
- I am now able to reproduce it on a brand new installation of Ceph v.0.94.7, EC 5,2 and exactly same settings as prod ...
- 02:25 AM Bug #17052 (Resolved): unittest_http_manager times out
- ...
08/16/2016
- 11:51 PM Bug #17051 (Resolved): period commit loses zonegroup changes: region_map converted repeatedly
- Reproduced on both Jewel and Jewel-backports 10.2.2-432-gac44947 (commit ac44947). This is trying to upgrade a Hammer...
- 03:44 PM Bug #17047 (Resolved): rgw multisite: doesn't retry RGWFetchAllMetaCR on failed lease
- When starting multiple gateways in the same non-master zone, they race for the lease during StateBuildingFullSyncMaps...
- 02:48 PM Bug #17044 (Fix Under Review): multisite: assert(next) failed in RGWMetaSyncCR
- https://github.com/ceph/ceph/pull/10743
- 02:45 PM Bug #17044 (Resolved): multisite: assert(next) failed in RGWMetaSyncCR
- When testing with multiple gateways per zone, I'm seeing some omap-get-vals requests fail with EIO. RGWMetaSyncCR doe...
- 12:30 PM Backport #16448: hammer: default quota fixes
- The proper solution, I believe, is to include this commit as well:
e3ab3476146cf0fe604783e9e5a0d63237271735 - 09:01 AM Backport #16319 (Resolved): jewel: radosgw-admin: inconsistency in uid/email handling
- 08:47 AM Bug #17040: Files incorrectly uploaded with swift api
- Apologies for the formatting: EC pool is 5,2 in production case and 2,1 in testing case.
- 08:44 AM Bug #17040 (Duplicate): Files incorrectly uploaded with swift api
- Hello,
I am trying to upload a file with size=55042267 to an S3 bucket backed by EC(5,2) pool using SWIFT api. I h... - 06:59 AM Feature #16971 (Fix Under Review): HTTP Status count for RGW requests
- 06:47 AM Bug #17036 (Fix Under Review): rgw: radosgw daemon core when reopen logs
- 06:39 AM Bug #17036: rgw: radosgw daemon core when reopen logs
- PR: https://github.com/ceph/ceph/pull/10737
- 06:19 AM Bug #17036 (Resolved): rgw: radosgw daemon core when reopen logs
- in my environment, the radosgw daemon coredump several times within days. below is part of the radosgw daemon log :
...
08/15/2016
- 09:56 PM Documentation #16635 (Resolved): rgw: document multi tenancy
- 09:56 PM Bug #16627 (Resolved): failed to create bucket after upgrade from hammer to jewel
- 09:56 PM Bug #16603 (Resolved): rgw: segmentation fault on error_repo in data sync
- 09:56 PM Bug #16530 (Resolved): rgw: data sync stops after getting error in all data log sync shards
- 09:56 PM Bug #16327 (Resolved): comparing return code to ERR_NOT_MODIFIED in rgw_rest_s3.cc (needs minus s...
- 09:56 PM Bug #16286 (Resolved): LDAP bindpw parsing is broken
- 09:56 PM Bug #15348 (Resolved): CORS: Access-Control-Allow-Origin should return * when set that way
- 09:38 PM Backport #16393 (In Progress): jewel: rgw: Swift's object versioning doesn't support restoring a ...
- 09:30 PM Backport #17034 (Resolved): jewel: rgw: object expirer's hints might be trimmed without processin...
- https://github.com/ceph/ceph/pull/10763
- 09:30 PM Backport #17033 (Resolved): jewel: rgw: there may be some objects not delete in some circumstances
- https://github.com/ceph/ceph/pull/10763
- 09:30 PM Backport #17032 (Resolved): jewel: multisite: RGWPeriodPuller tries to pull from itself
- https://github.com/ceph/ceph/pull/10764
- 09:29 PM Backport #16112 (Resolved): jewel: CORS: Access-Control-Allow-Origin should return * when set tha...
- 09:29 PM Backport #16272 (Resolved): jewel: rgw ldap: fix ldap bindpw parsing
- 09:28 PM Backport #16565 (Resolved): jewel: rgw: data sync stops after getting error in all data log sync ...
- 09:28 PM Backport #16636 (Resolved): jewel: rgw: document multi tenancy
- 09:27 PM Backport #16700 (Resolved): jewel: rgw: segmentation fault on error_repo in data sync
- 09:27 PM Backport #16731 (Resolved): jewel: failed to create bucket after upgrade from hammer to jewel
- 06:02 PM Bug #16939 (Pending Backport): multisite: RGWPeriodPuller tries to pull from itself
- 03:12 PM Bug #16684 (Pending Backport): rgw: object expirer's hints might be trimmed without processing in...
- 03:12 PM Bug #16684 (Fix Under Review): rgw: object expirer's hints might be trimmed without processing in...
- PR: https://github.com/ceph/ceph/pull/10330.
- 03:11 PM Bug #16705: rgw: there may be some objects not delete in some circumstances
- PR: https://github.com/ceph/ceph/pull/10330.
- 03:10 PM Bug #16705 (Pending Backport): rgw: there may be some objects not delete in some circumstances
08/13/2016
- 06:33 PM Backport #16872 (Resolved): hammer: Have a flavor of bucket deletion in radosgw-admin to bypass g...
- 05:00 PM Bug #12890 (Resolved): rgw: the swift key remains after removing a subuser
- 05:00 PM Backport #14328 (Resolved): hammer: rgw: the swift key remains after removing a subuser
08/12/2016
- 01:12 PM Backport #16393 (Fix Under Review): jewel: rgw: Swift's object versioning doesn't support restori...
- PR: https://github.com/ceph/ceph/pull/10710.
- 10:24 AM Backport #16393 (In Progress): jewel: rgw: Swift's object versioning doesn't support restoring a ...
- No problem, Loic. I will take care of that.
- 10:44 AM Bug #17009 (Closed): rgw: leak in RGWLC::start_process
- 10:35 AM Bug #17009 (Closed): rgw: leak in RGWLC::start_process
- http://qa-proxy.ceph.com/teuthology/owasserm-2016-07-29_17:43:49-rgw-wip-orit-testing---basic-smithi/340269/remote/sm...
- 09:56 AM Backport #16317 (Resolved): hammer: rgw: bucket listing following object delete is partial
- 09:55 AM Backport #16734 (Resolved): hammer: Bucket index shards orphaned after bucket delete
- 09:51 AM Bug #15243 (Resolved): rgw: can't remove non-versioned object after enabling versioning on bucket
- 09:48 AM Backport #17006 (Resolved): jewel: Increase log level for messages occuring while running rgw adm...
- https://github.com/ceph/ceph/pull/10765
08/11/2016
- 02:33 PM Backport #16815 (Resolved): radosgw-admin region-map set is not reporting the bucket quota correc...
- 09:50 AM Backport #15254 (Resolved): hammer: rgw: can't remove non-versioned object after enabling version...
- 09:50 AM Backport #16381 (Resolved): jewel: comparing return code to ERR_NOT_MODIFIED in rgw_rest_s3.cc (n...
- 09:50 AM Backport #16392 (Resolved): jewel: rgw: master: build failures with boost > 1.58
- 09:25 AM Bug #16682 (Fix Under Review): ceph jewel failed create s3 tpye subuser from admin rest api
- https://github.com/ceph/ceph/pull/10325
- 02:30 AM Bug #16995 (New): rgw files can't upload
- When I upload a file。rgw response ERR_LENGTH_REQUIRED。
this means file length is empty。I see the rgw log,I find ngin...
08/10/2016
- 06:16 PM Bug #16935 (Pending Backport): Increase log level for messages occuring while running rgw admin c...
- 12:29 PM Backport #16958 (In Progress): jewel: Bug when using port 443s in rgw.
- 12:26 PM Backport #16934 (In Progress): jewel: Add zone rename to radosgw_admin
- 12:24 PM Backport #16901 (In Progress): jewel: segfault in RGWOp_MDLog_Notify
- 12:21 PM Backport #16871 (In Progress): jewel: Have a flavor of bucket deletion in radosgw-admin to bypass...
- 12:20 PM Backport #16864 (In Progress): jewel: multisite segfault on ~RGWRealmWatcher if realm was deleted
- 12:18 PM Backport #16863 (In Progress): jewel: use zone endpoints instead of zonegroup endpoints
- 12:10 PM Backport #16862 (In Progress): jewel: "default" zone and zonegroup cannot be added to a realm
- 12:05 PM Backport #16792 (In Progress): jewel: metadata sync can skip markers for failed/incomplete entries
- 12:04 PM Backport #16794: jewel: multisite bucket sync doesn't retry objects that it fails to fetch
- https://github.com/ceph/ceph/pull/10355/commits/aacc31f72cef6f151459e4a0543850edeeaf1938 does not apply cleanly and i...
- 12:00 PM Backport #16793 (Need More Info): jewel: upgrade from old multisite to new multisite fails
- 12:00 PM Bug #16751: upgrade from old multisite to new multisite fails
- Which pull request contains the associated fix ?
- 11:38 AM Backport #16564 (In Progress): jewel: cors auto memleak
- 11:03 AM Backport #16564: jewel: cors auto memleak
- https://github.com/ceph/ceph/pull/9979/commits/69c6cf551754e51debb4e645716a89e29517f4ee from https://github.com/ceph/...
- 11:36 AM Backport #16393: jewel: rgw: Swift's object versioning doesn't support restoring a previous versi...
- This does not cherry pick cleanly and the conflicts are non trivial. Since it's a feature, it should probably be done...
- 11:31 AM Bug #14090 (Resolved): rgw: the response XML format of admin policy API is error
- 11:22 AM Bug #15926 (Resolved): radosgw-admin period push --url fails with 'could not find connection'
- 11:21 AM Backport #15958 (Resolved): jewel: radosgw-admin period push --url fails with 'could not find con...
- 10:22 AM Backport #16702 (New): jewel: multisite: bucket sync failures with tenant users
- First backport attempt: https://github.com/ceph/ceph/pull/10522
This require a number of non trivial conflict reso... - 08:42 AM Bug #16968: synchonized bucket can't be used in other zonegroup
- Data synchronization happens between zones within same zonegroup. Buckets can only accessed on the zonegroup where th...
08/09/2016
- 08:54 PM Bug #16976 (Resolved): rgw file uses too much CPU in gc/idle thread
- In librgw/nfs-ganesha operation only, the periodic gc/idle thread is missing a delay step.
- 06:24 PM Bug #16935 (Fix Under Review): Increase log level for messages occuring while running rgw admin c...
- https://github.com/ceph/ceph/pull/10602
- 01:58 PM Feature #16971 (Fix Under Review): HTTP Status count for RGW requests
- We are capturing req, failed_req, get and put requests from every RGW node. It would be nice to have continuously inc...
- 08:32 AM Bug #16968 (Closed): synchonized bucket can't be used in other zonegroup
- When multisite is configured, a bucket created in a zonegroup would be synchonized to other zonegroups. But the bucke...
08/08/2016
- 06:44 PM Backport #16958 (Resolved): jewel: Bug when using port 443s in rgw.
- https://github.com/ceph/ceph/pull/10664
- 05:23 PM Bug #16957 (New): NO_SSL_DL is not passed to civetweb during compilation of radosgw
- RGW Makefile.am includes civetweb.c when compiling the radosgw binary. -DNO_SSL_DL is only provided to libcivetweb.
- 05:18 PM Bug #16548 (Pending Backport): Bug when using port 443s in rgw.
- 11:46 AM Bug #16624: register the correct handler for cls_user_complete_stats
- btw, in the original pr, Yehuda did mention that there isnt a need for backport
- 08:37 AM Bug #16624: register the correct handler for cls_user_complete_stats
- infernalis is EOL
- 08:44 AM Bug #15896 (Resolved): S3: set EncodingType in ListBucketResult
- 08:42 AM Bug #16464 (Resolved): multisite sync races with deletes
- 08:42 AM Feature #16004 (Resolved): rgw : support size suffixes for --max-size in radosgw-admin command
- 08:42 AM Bug #15940 (Resolved): rgw: aws4 parsing issue
- 08:42 AM Bug #15779 (Resolved): rgw: account/container metadata not actually present in a request are dele...
- 08:42 AM Bug #15720 (Resolved): most admin operations see a failed to create default zone error message
- 08:37 AM Bug #16412: Bucket index shards orphaned after bucket delete
- infernalis is EOL
- 08:36 AM Bug #13529: rgw: leak from RGWWatcher::handle_error
- infernalis is EOL
- 08:36 AM Bug #15445: multisite s3tests test_versioning_obj_create_versions_remove_special_names causes rad...
- infernalis is EOL
- 08:36 AM Bug #15243: rgw: can't remove non-versioned object after enabling versioning on bucket
- infernalis is EOL
- 08:36 AM Bug #14826: rgw: bucket listing following object delete is partial
- infernalis is EOL
- 08:35 AM Bug #12890: rgw: the swift key remains after removing a subuser
- infernalis is EOL
- 08:35 AM Bug #13817: RGW does not send HTTP header
- infernalis is EOL
- 08:35 AM Documentation #15444: orphan cleanup tool: document/test interaction w/ placement targets & multi...
- infernalis is EOL
- 08:34 AM Feature #15442: Orphan cleanup tool: Need to be able to list jobs
- infernalis is EOL
- 08:33 AM Bug #15745 (Resolved): RGW :: Subset of uploaded objects via radosgw are unretrievable when using...
- 08:33 AM Bug #14191 (Resolved): rgw: radosgw server abort when user passed bad parameters to set quota
- 08:32 AM Bug #13239 (Resolved): Seg fault 9.0.3-1845-gf1ead76 : RGWRESTSimpleRequest::forward_request(RGWA...
- 08:32 AM Bug #13140 (Resolved): rgw: swift API returns more than real object count and bytes used when ret...
- 08:32 AM Feature #12997 (Resolved): rgw: default quota params
- 08:32 AM Bug #12955 (Resolved): Setting ACL on Object removes ETag
- 08:32 AM Bug #11076 (Resolved): Bucket owner isn't changed after unlink/link
- 08:29 AM Backport #16733 (Rejected): infernalis: Bucket index shards orphaned after bucket delete
- 08:28 AM Backport #16316 (Rejected): infernalis: rgw: bucket listing following object delete is partial
- 08:28 AM Backport #15832 (Rejected): infernalis: RGW :: Subset of uploaded objects via radosgw are unretri...
- 08:28 AM Backport #15256 (Rejected): infernalis: rgw: radosgw server abort when user passed bad parameters...
- 08:28 AM Backport #15253 (Rejected): infernalis: rgw: can't remove non-versioned object after enabling ver...
- 08:28 AM Backport #15088 (Rejected): infernalis: Bucket owner isn't changed after unlink/link
- 08:28 AM Backport #14691 (Rejected): infernalis: Seg fault 9.0.3-1845-gf1ead76 : RGWRESTSimpleRequest::for...
- 08:28 AM Backport #14327 (Rejected): infernalis: rgw: the swift key remains after removing a subuser
- 08:28 AM Backport #14140 (Rejected): infernalis: rgw: leak from RGWWatcher::handle_error
- 08:28 AM Backport #13819 (Rejected): infernalis: Setting ACL on Object removes ETag
- 08:21 AM Backport #16339 (Resolved): jewel : rgw : support size suffixes for --max-size in radosgw-admin c...
- 08:20 AM Backport #16589 (Resolved): jewel: multisite sync races with deletes
08/05/2016
- 09:05 PM Backport #16945 (In Progress): jewel: RGW/civetweb no longer listens on IPv6: invalid port spec
- 09:04 PM Backport #16945 (Resolved): jewel: RGW/civetweb no longer listens on IPv6: invalid port spec
- https://github.com/ceph/ceph/pull/10580
- 09:03 PM Bug #16928 (Pending Backport): RGW/civetweb no longer listens on IPv6: invalid port spec
- 09:00 PM Bug #16928: RGW/civetweb no longer listens on IPv6: invalid port spec
- Um, wherever you can put in --include foo.h you can put in -DFOO. So I'm not sure why you think --include is minimal...
- 03:25 PM Bug #16928: RGW/civetweb no longer listens on IPv6: invalid port spec
- Marcus Watts wrote:
> should be a line like this,
> src/CMakeLists.txt: set_property(TARGET civetweb_common_objs
... - 04:58 AM Bug #16928: RGW/civetweb no longer listens on IPv6: invalid port spec
should be a line like this,
src/CMakeLists.txt: set_property(TARGET civetweb_common_objs
src/CMakeLists.txt: ...- 12:26 AM Bug #16928: RGW/civetweb no longer listens on IPv6: invalid port spec
- Master:
https://github.com/ceph/ceph/pull/10579
Jewel:
https://github.com/ceph/ceph/pull/10580 - 12:11 AM Bug #16928: RGW/civetweb no longer listens on IPv6: invalid port spec
- civetweb.c is included twice for compiling, but only the libcivetweb.la output passes "--include ./civetweb/include/c...
- 03:41 PM Bug #16939 (Fix Under Review): multisite: RGWPeriodPuller tries to pull from itself
- https://github.com/ceph/ceph/pull/10596
- 03:36 PM Bug #16939 (Resolved): multisite: RGWPeriodPuller tries to pull from itself
- when RGWPeriodPuller::pull() is called repeatedly (as in RGWPeriodHistory), the period argument is being reused witho...
- 01:10 PM Backport #16934: jewel: Add zone rename to radosgw_admin
- h3. original description
https://github.com/ceph/ceph/pull/9779 needs to be backported to jewel - 07:20 AM Backport #16934 (Resolved): jewel: Add zone rename to radosgw_admin
- https://github.com/ceph/ceph/pull/10663
- 09:54 AM Backport #16448 (Need More Info): hammer: default quota fixes
- 09:53 AM Backport #16448: hammer: default quota fixes
- the same problem with jewel
conflict with https://github.com/ceph/ceph/pull/9037
this introduce size unit is byte... - 07:39 AM Bug #16935 (Resolved): Increase log level for messages occuring while running rgw admin command
- # radosgw-admin period update --commit
2016-08-02 14:48:28.672409 7f9e630639c0 0 error in read_id for id : (2) No ... - 04:55 AM Bug #16813: "radosgw --show-config" and "radosgw-admin --show-config" do not display current conf...
- Hi James,
Actually, `--show-config` is a globally option that display the default code-level configurations and t...
08/04/2016
- 11:33 PM Bug #16928: RGW/civetweb no longer listens on IPv6: invalid port spec
- Ok, I can prove the USE_IPV6 is not defined when src/civetweb/src/civetweb.c is compiled, and it seems that civetweb/...
- 10:56 PM Bug #16928 (Resolved): RGW/civetweb no longer listens on IPv6: invalid port spec
- civetweb: 0x7f507d36cbe0: set_ports_option: [::]:7480: invalid port spec. Expecting list of: [IP_ADDRESS:]PORT[s|r]
... - 11:19 PM Bug #16930 (Resolved): s3tests-test-readwrite failing with 500
- sjust@teuthology:/a/dzafman-2016-08-03_16:16:17-rados-wip-zafman-testing-distro-basic-smithi/348084
- 01:07 PM Bug #16861 (Resolved): unittest_rgw_bencode failed during `make check`
08/03/2016
- 09:30 PM Bug #16813: "radosgw --show-config" and "radosgw-admin --show-config" do not display current conf...
- Hi Yan,
OK that works, but I don't see it documented anywhere. Can the `--show-config` be added to the `radosgw-a... - 02:17 PM Bug #16894 (Fix Under Review): multisite: metadata master can get the wrong value for 'oldest_log...
- https://github.com/ceph/ceph/pull/10558
- 12:31 PM Bug #16581: rgw: error is logged when binding civetweb to an ip
- Abhishek Lekshmanan wrote:
> While we're at it, there is an upstream support for ssl + civetweb, we could probably a... - 12:30 PM Bug #16581: rgw: error is logged when binding civetweb to an ip
- While we're at it, there is an upstream pr support for ssl + civetweb, we could probably address this one also as a p...
- 10:02 AM Bug #16581 (In Progress): rgw: error is logged when binding civetweb to an ip
- 11:13 AM Backport #16815 (In Progress): radosgw-admin region-map set is not reporting the bucket quota cor...
- 11:13 AM Backport #16317 (In Progress): hammer: rgw: bucket listing following object delete is partial
- 05:44 AM Backport #16901 (Resolved): jewel: segfault in RGWOp_MDLog_Notify
- https://github.com/ceph/ceph/pull/10662
08/02/2016
- 06:37 PM Bug #16894 (Resolved): multisite: metadata master can get the wrong value for 'oldest_log_period'
- When the metadata master starts up, it goes backwards through its per-period metadata logs until it finds a period wi...
- 03:11 PM Bug #16581: rgw: error is logged when binding civetweb to an ip
- Yeah, we seem to parse port as an integer early, though civetweb seperately gets the conf value as a string, so it do...
- 10:59 AM Bug #16581: rgw: error is logged when binding civetweb to an ip
- 01:22 PM Bug #16695 (Fix Under Review): radosgw Consumes too much CPU time to synchronize metadata or data...
- -https://github.com/ceph/ceph/pull/10467-
- 01:21 PM Bug #16666 (Pending Backport): segfault in RGWOp_MDLog_Notify
- 09:34 AM Bug #16388 (Resolved): Wrong Content-Range and Content-Length for zero size object with Range req...
- commit:4a90e648ed87433f026d03e6a499549c4bb8b710
- 08:32 AM Backport #16778 (In Progress): multisite: 400-error with certain complete multipart upload requests
- https://github.com/ceph/ceph/pull/10537
- 03:25 AM Bug #16888 (Resolved): [multisite] operating bucket's acl&cors is not restricted on slave zone
- modify bucket's acl cors on slave zone will not sync to master.so it should be restricted or redirected.
08/01/2016
- 11:29 PM Bug #16581: rgw: error is logged when binding civetweb to an ip
- +1. I've confirmed that the civetweb documentation says this is a valid syntax:
https://github.com/civetweb/cive... - 05:16 PM Backport #16734 (In Progress): hammer: Bucket index shards orphaned after bucket delete
- 05:05 PM Backport #16734: hammer: Bucket index shards orphaned after bucket delete
- https://github.com/ceph/ceph/pull/10530
- 03:46 PM Bug #16856 (In Progress): ACL request for objects with underscore at end and beginning
- 03:44 PM Bug #16856: ACL request for objects with underscore at end and beginning
- I've a wip-patch for this at https://github.com/ceph/ceph/pull/10528
- 02:37 PM Bug #16767: RadosGW Multipart Cleanup Failure
- With apologies for pestering, is there anyone looking into this? This bug affects the viability of Ceph as a backing...
- 09:10 AM Backport #16732 (In Progress): jewel: Bucket index shards orphaned after bucket delete
- 09:04 AM Backport #16731 (In Progress): jewel: failed to create bucket after upgrade from hammer to jewel
- 09:02 AM Backport #16700 (In Progress): jewel: rgw: segmentation fault on error_repo in data sync
- 08:59 AM Backport #16696: jewel: segfault in RGWBucketShardIncrementalSyncCR
- Waiting for http://tracker.ceph.com/issues/16702 to be resolved before backporting this
- 07:31 AM Backport #16702: jewel: multisite: bucket sync failures with tenant users
- Backported https://github.com/ceph/ceph/pull/10075 and https://github.com/ceph/ceph/pull/10252 in the PR
- 07:30 AM Backport #16702 (In Progress): jewel: multisite: bucket sync failures with tenant users
- 07:09 AM Backport #16447: jewel: default quota fixes
- It might be conflicting in rgw_quota.cc around the constructor for RGWQuotaHandlerImpl, which is not needed anymore. ...
- 06:47 AM Backport #16447 (Need More Info): jewel: default quota fixes
- This does not cleanly backport to jewel as it conflicts with https://github.com/ceph/ceph/pull/9037.
- 06:33 AM Backport #16319 (In Progress): jewel: radosgw-admin: inconsistency in uid/email handling
- 06:30 AM Backport #16312 (In Progress): jewel: selinux denials in RGW
- 06:27 AM Backport #16272 (In Progress): jewel: rgw ldap: fix ldap bindpw parsing
- 02:51 AM Bug #16877 (Duplicate): mulsites synchonize objects in limitless loop
- When multisite is configured, radosgw will update objects in limitless loop. For example, zone-a create an object nam...
07/30/2016
- 10:35 AM Bug #16813: "radosgw --show-config" and "radosgw-admin --show-config" do not display current conf...
- Hi, We can get the accurate configuration by specifying rgw instance name (eg. -n client.rgw.us-east-1). since we may...
- 03:32 AM Backport #16872 (In Progress): hammer: Have a flavor of bucket deletion in radosgw-admin to bypas...
- 03:27 AM Backport #16872 (Resolved): hammer: Have a flavor of bucket deletion in radosgw-admin to bypass g...
- https://github.com/ceph/ceph/pull/10509
- 03:27 AM Backport #16871 (Resolved): jewel: Have a flavor of bucket deletion in radosgw-admin to bypass ga...
- https://github.com/ceph/ceph/pull/10661
- 03:24 AM Backport #16864 (Resolved): jewel: multisite segfault on ~RGWRealmWatcher if realm was deleted
- https://github.com/ceph/ceph/pull/10660
- 03:23 AM Backport #16863 (Resolved): jewel: use zone endpoints instead of zonegroup endpoints
- https://github.com/ceph/ceph/pull/10659
- 03:23 AM Backport #16862 (Resolved): jewel: "default" zone and zonegroup cannot be added to a realm
- https://github.com/ceph/ceph/pull/10658
07/29/2016
- 11:07 PM Bug #16861: unittest_rgw_bencode failed during `make check`
- Here's a proposed fix:
https://github.com/ceph/ceph/pull/10510
(didn't even test compile, so need to do that be... - 08:42 PM Bug #16861 (Resolved): unittest_rgw_bencode failed during `make check`
- ...
- 10:41 PM Feature #15557 (Pending Backport): Have a flavor of bucket deletion in radosgw-admin to bypass ga...
- hammer backport PR:
https://github.com/ceph/ceph/pull/10509 - 04:24 PM Bug #16856: ACL request for objects with underscore at end and beginning
- I'm guessing we end up setting namespace for the object as temp here (rgw_common: from_index_key()) and this starts t...
- 08:49 AM Bug #16856: ACL request for objects with underscore at end and beginning
- ...
- 08:47 AM Bug #16856 (Resolved): ACL request for objects with underscore at end and beginning
- It seems that an ACL-request to an object starting with an underscore and ending with another underscore is
broken ... - 07:21 AM Backport #16698 (Resolved): hammer: multidelete query parameter not correctly parsed
- 07:20 AM Backport #16382 (Resolved): hammer: comparing return code to ERR_NOT_MODIFIED in rgw_rest_s3.cc (...
- 07:19 AM Bug #14005 (Resolved): RGW shouldn't send Content-Type nor Content-Length for 304 responses
- 07:18 AM Backport #15283 (Resolved): hammer: RGW shouldn't send Content-Type nor Content-Length for 304 re...
- 07:15 AM Backport #15257 (Resolved): hammer: rgw: radosgw server abort when user passed bad parameters to ...
- 07:13 AM Backport #15839 (Resolved): hammer: CORS: Access-Control-Allow-Origin should return * when set th...
- 07:11 AM Backport #16262 (Resolved): hammer: backport rgwx-copy-if-newer for radosgw-sync
- 07:10 AM Bug #14375 (Resolved): subuser rm fails with status 125
- 07:09 AM Backport #16190 (Resolved): hammer: subuser rm fails with status 125
- 07:08 AM Bug #15812 (Resolved): No Last-Modified, Content-Size and X-Object-Manifest headers if no segment...
- 07:08 AM Backport #15966 (Resolved): hammer: No Last-Modified, Content-Size and X-Object-Manifest headers ...
- 02:26 AM Bug #16854 (In Progress): the usage log record date is 16 hours later than the real operate time...
- hi all,
i want get the usage of user,so i use the command radosgw-admin usage
show ,but i can not get the usage w...
07/28/2016
- 02:58 PM Bug #16839 (Pending Backport): "default" zone and zonegroup cannot be added to a realm
- 02:26 PM Bug #16839: "default" zone and zonegroup cannot be added to a realm
- Orit Wasserman wrote:
> Creating a default realm will have backward compatibility issue as it doesn't exists in jewe... - 07:55 AM Bug #16839: "default" zone and zonegroup cannot be added to a realm
- Creating a default realm will have backward compatibility issue as it doesn't exists in jewel.
How about using zoneg... - 02:34 PM Bug #16323 (Resolved): terminate called after throwing an instance of 'std::out_of_range' in rgw-...
- No longer reproducible after https://github.com/ceph/ceph/pull/8528 was fixed.
- 07:50 AM Bug #16357: radosgw stats error
- Possibly https://github.com/ceph/ceph/pull/9584? I'm still able to see the issue in master though
- 07:47 AM Bug #13635: radosgw duplicate email address
- I would like to know the reasoning behind email address being unique. I understand that uid need to be unique, howeve...
07/27/2016
- 09:39 PM Bug #16836 (Can't reproduce): hammer-backports: cephtest/{swift,s3-tests} fail to bootstrap
- Not reproducible.
- 04:05 PM Bug #16836 (Can't reproduce): hammer-backports: cephtest/{swift,s3-tests} fail to bootstrap
- This happened in a bunch of jobs in hammer-backports testing - see
http://pulpito.ceph.com/smithfarm-2016-07-25_05:0... - 09:29 PM Bug #16839 (Resolved): "default" zone and zonegroup cannot be added to a realm
- Radosgw creates a zone and zonegroup by default, but does not create a default realm. The default zone and zonegroup ...
- 06:17 PM Bug #16834 (Pending Backport): use zone endpoints instead of zonegroup endpoints
- 05:31 PM Bug #16834 (Fix Under Review): use zone endpoints instead of zonegroup endpoints
- https://github.com/ceph/ceph/pull/10456
- 03:43 PM Bug #16834 (Resolved): use zone endpoints instead of zonegroup endpoints
- The command to change the master zone (radosgw-admin zone modify --master) does not change the zonegroup's endpoints ...
- 04:34 PM Bug #16357: radosgw stats error
- @Orit: does it happen in jewel?
- 08:34 AM Bug #16357: radosgw stats error
- This no longer happens in master (it still does in hammer).
We need to figure out what exactly fixed it and backport... - 03:46 AM Bug #16828 (Can't reproduce): radosgw swift interface put object response 500 error?
- hello,
I startup a radosgw (Civetweb embedded), i want use swift interface to put/get oject;
my radosgw's ceph c...
07/26/2016
- 11:27 PM Bug #16824 (New): Swift API - DLO manifest cannot be retrieved with 'multipart-manifest=get' query
- Suppose I have uploaded a 2G DLO to Radosgw (0.94.6):...
- 10:27 PM Bug #16822 (New): Swift API - DLO etag is computed from manifest rather than from etags of segments
- Consider uploading 2 multi-part dynamic large objects via the Swift API to Radosgw (version 0.94.6) and stat'ing them...
- 08:21 PM Feature #13718: rgw:CivetWeb http and https be support in the same time.
- *master PR* https://github.com/ceph/ceph/pull/10335
- 06:16 PM Bug #16817 (Pending Backport): multisite segfault on ~RGWRealmWatcher if realm was deleted
- 04:54 PM Bug #16817 (In Progress): multisite segfault on ~RGWRealmWatcher if realm was deleted
- https://github.com/ceph/ceph/pull/10446
- 04:39 PM Bug #16817 (Resolved): multisite segfault on ~RGWRealmWatcher if realm was deleted
- In a running multisite cluster:
$ radosgw-admin realm delete --rgw-realm=dev
the command succeeds, and deletes ... - 02:16 PM Feature #12790: rgw: multipart uploads copy part support
- The current test cases in s3-tests got old and they could be affected by some kind of regression. The attached script...
- 12:39 PM Backport #16815 (Resolved): radosgw-admin region-map set is not reporting the bucket quota correc...
- https://github.com/ceph/ceph/pull/10554
The output of the region-map command is not showing the correct bucket quo...
07/25/2016
- 09:06 PM Bug #16813 (New): "radosgw --show-config" and "radosgw-admin --show-config" do not display curren...
- These switches seem to show a mixture of the current configuration and the default configuration depending on the set...
- 05:11 PM Bug #11239: civetweb defaults to libssl.so and libcrypto.so when versions not passed
- Ah, never mind.
- 03:31 PM Bug #11239 (Fix Under Review): civetweb defaults to libssl.so and libcrypto.so when versions not ...
- https://github.com/ceph/ceph/pull/10335
- 04:43 PM Bug #16806: S3 API returns bad Canonical User ID
- I'd say it's a bug in the client library more than Ceph; and I've filed a bug with them
https://github.com/rustycono... - 11:07 AM Bug #16806 (New): S3 API returns bad Canonical User ID
- list all buckets request returs something like:...
- 08:25 AM Bug #16712 (Pending Backport): multisite: 400-error with certain complete multipart upload requests
07/23/2016
- 04:31 PM Feature #10342: rgw: tool to clean up clutter
- Hi Yehuda, where could the tool be found?
- 01:50 PM Bug #16444: rgw-admin: better error message when user has no buckets yet
- Pull request https://github.com/ceph/ceph/pull/10162 got merged.
07/22/2016
- 10:19 PM Documentation #13523: rgw: document civetweb ssl configuration
- All the steps for configuring multistage rgw are for http only. Could you please give us the steps for https along w...
- 08:36 PM Backport #16794 (Resolved): jewel: multisite bucket sync doesn't retry objects that it fails to f...
- https://github.com/ceph/ceph/pull/11021
- 08:36 PM Backport #16793 (Resolved): jewel: upgrade from old multisite to new multisite fails
- https://github.com/ceph/ceph/pull/10891
- 08:35 PM Backport #16792 (Resolved): jewel: metadata sync can skip markers for failed/incomplete entries
- https://github.com/ceph/ceph/pull/10657
- 06:38 PM Tasks #16784 (Fix Under Review): RADOSGW feature requires FCGI which no longer has an upstream so...
- RADOSGW of Ceph v10.2.2 relies on FCGI, but the upstream source for this no longer actually exists.
- 03:42 PM Backport #16778 (Resolved): multisite: 400-error with certain complete multipart upload requests
- https://github.com/ceph/ceph/pull/10537
- 09:39 AM Bug #16712: multisite: 400-error with certain complete multipart upload requests
- master pr: https://github.com/ceph/ceph/pull/10397
07/21/2016
- 10:02 PM Bug #16742 (Pending Backport): multisite bucket sync doesn't retry objects that it fails to fetch
- 10:01 PM Bug #16759 (Pending Backport): metadata sync can skip markers for failed/incomplete entries
- Fixed by commit:f5801c726efdc2f3067a071e6bb5ac83fd0cd147
- 08:42 PM Bug #15915: rgw command is consuming all the cpu time
> Good to know, thank you very much for testing. That means we can reproduce by running against older versions of l...- 08:29 PM Bug #15915: rgw command is consuming all the cpu time
- Russell Islam wrote:
> Tested with later version of curl. In my case 7.43. Got rid of this issue.
Good to know, t... - 08:21 PM Bug #15915: rgw command is consuming all the cpu time
- Tested with later version of curl. In my case 7.43. Got rid of this issue.
- 07:18 PM Bug #16767 (Resolved): RadosGW Multipart Cleanup Failure
- My current setup is a Ceph Hammer cluster running 0.94.6. The rest of the cluster details are irrelevant to this iss...
- 10:05 AM Bug #16751 (Pending Backport): upgrade from old multisite to new multisite fails
07/20/2016
- 09:08 PM Bug #16759 (Resolved): metadata sync can skip markers for failed/incomplete entries
- A metadata sync for 'bucket55' never completed, but a sync for a later entry in the same shard succeeded and updated ...
- 03:30 PM Bug #16712 (In Progress): multisite: 400-error with certain complete multipart upload requests
- I'm guessing this is happening when we update the bucket instance, we don't copy over the source data_extra pool and ...
- 10:58 AM Bug #16753 (Closed): radosgw-admin user list crashes
- seems to be fixed in latest mster
- 09:56 AM Bug #16753 (Closed): radosgw-admin user list crashes
- eph version v11.0.0-711-gd1f681a (d1f681a2741ac805d1a21087023147dddd218940)
1: ./bin/radosgw-admin() [0x6da51f]
2... - 09:48 AM Bug #16752 (Closed): Cannot start radosgw when there is a default zonegroup but no default zone w...
- 09:45 AM Bug #16751 (Resolved): upgrade from old multisite to new multisite fails
- Updated current period was not stored correctly
Also available in: Atom