Activity
From 06/25/2017 to 07/24/2017
07/24/2017
- 10:23 PM Bug #20373: "test_get_manifest (test.functional.tests.TestDlo)*" failed in rgw
- Building wip-yuri-master_2017_7_24
- 08:45 PM Bug #20527: v2 presigned URLs don't work with radosgw.
- Discovered that Hammer's auth_hdr debug output ends with:...
- 07:48 PM Bug #20527: v2 presigned URLs don't work with radosgw.
- I was able to reproduce this by running s3curl against demo Docker containers.
ceph/demo:tag-build-master-hammer-u... - 07:02 PM Bug #20527: v2 presigned URLs don't work with radosgw.
- Hello!
My team is trying to update from Hammer to Jewel, and we believe we have run into this issue. We are testi... - 05:21 PM Bug #20686 (Pending Backport): rgw hangs in RGWRealmReloader::reload on SIGHUP
- 03:18 PM Bug #20661: With RGWs configured in a load balancer, quota stats cache doesn't work
- OK we fetch the storage stats only for the bucket stats not the user stats.
The failure is for the user stats, makin... - 02:20 PM Bug #20661: With RGWs configured in a load balancer, quota stats cache doesn't work
- Can you provide rgw logs? (debug_rgw = 20)
From the code it should have return false on can_use_cache_stats and fetc... - 01:42 PM Bug #20661: With RGWs configured in a load balancer, quota stats cache doesn't work
- Yes it is reproducible on master as well, sorry if the affected versions field has misled you.
- 12:03 PM Bug #20661: With RGWs configured in a load balancer, quota stats cache doesn't work
- Dcan you reproduce it on newer ceph versions?
- 02:44 PM Bug #20289 (Resolved): "failed to list reshard log entries" spams log file
- 01:47 PM Bug #20757 (Fix Under Review): rgw: RadosGW leaks memory during Swift Static Website's error hand...
- PR: https://github.com/ceph/ceph/pull/16531.
- 01:01 PM Bug #20757 (Resolved): rgw: RadosGW leaks memory during Swift Static Website's error handling
- ...
- 01:10 PM Bug #20663: Segmentation fault when exporting rgw bucket in nfs-ganesha
- Matt Benjamin wrote:
> arg, it's...bitwise, sorry, will update
>
> I'm surprised this isn't universally hit.
> ... - 08:57 AM Bug #20756 (In Progress): radosgw daemon crash when service is restarted during lifecycle processing
- PR Created: https://github.com/ceph/ceph/pull/16495
- 08:50 AM Bug #20756 (Closed): radosgw daemon crash when service is restarted during lifecycle processing
- radosgw daemon crash when service is restarted during lifecycle processing:
2017-07-24 00:00:19.854742 7fd4ca9da70... - 08:47 AM Bug #20755 (Closed): radosgw crash when service is restarted during lifecycle processing
- radosgw crash when service is restarted during lifecycle processing:
2017-07-24 00:00:19.854742 7fd4ca9da700 -1 ...
07/22/2017
- 01:48 AM Bug #19956: radosgw Segmentation fault when service is restarted during lifecycle processing
- wei qiaomiao wrote:
> I think this PR: http://tracker.ceph.com/issues/19956 will help you.
The link just goes bac... - 12:31 AM Bug #19956: radosgw Segmentation fault when service is restarted during lifecycle processing
- I think this PR: http://tracker.ceph.com/issues/19956 will help you.
07/21/2017
- 02:36 PM Bug #20570: Ceph - Keystone Implicit Tenants
- Ross Martyn wrote:
> Discussions with a developer at my end suggested it could be possible to use something like the... - 02:00 PM Bug #20570: Ceph - Keystone Implicit Tenants
- Discussions with a developer at my end suggested it could be possible to use something like the following (based on a...
- 02:53 AM Feature #20733 (New): RGW bucket limits
- our reqirements:
10w small files/bucket, file size about 10Kb
We need 2w~20Wbucket satisfy our file store.
But... - 12:33 AM Bug #20612: radosgw ceases responding to list requests
- I've since found another solution for my storage needs and allocated my hardware to that, so I can't reproduce I'm af...
07/20/2017
- 08:34 PM Backport #20346: jewel: rgw: meta sync thread crash at RGWMetaSyncShardCR
- Backporting note: after cherry-picking the main master fix, also cherry-pick the follow-on fix https://github.com/cep...
- 08:30 PM Backport #20729 (Rejected): kraken: rgw: multipart parts on versioned bucket create versioned buc...
- 08:30 PM Backport #20728 (Resolved): jewel: rgw: multipart parts on versioned bucket create versioned buck...
- https://github.com/ceph/ceph/pull/17278
- 08:29 PM Backport #20727 (Rejected): kraken: rgw: user quota did not work well on multipart upload
- 08:29 PM Backport #20726 (Resolved): jewel: rgw: user quota did not work well on multipart upload
- https://github.com/ceph/ceph/pull/17277
- 08:29 PM Backport #20725 (Rejected): kraken: Content-MD5 header is not validated with POST uploads
- 08:29 PM Backport #20724 (Rejected): jewel: Content-MD5 header is not validated with POST uploads
- 08:28 PM Bug #20663: Segmentation fault when exporting rgw bucket in nfs-ganesha
- luminous backports haven't started yet
- 12:25 PM Bug #20663 (Pending Backport): Segmentation fault when exporting rgw bucket in nfs-ganesha
- arg, it's...bitwise, sorry, will update
I'm surprised this isn't universally hit.
https://github.com/ceph/ceph... - 08:28 PM Backport #20721 (Rejected): kraken: multisite: RGWPeriodPuller does not call RGWPeriod::reflect()...
- 08:28 PM Backport #20720 (Resolved): jewel: multisite: RGWPeriodPuller does not call RGWPeriod::reflect() ...
- https://github.com/ceph/ceph/pull/17167
- 08:28 PM Backport #20719 (Resolved): jewel: rgw: Truncated objects
- https://github.com/ceph/ceph/pull/16708
https://github.com/ceph/ceph/pull/17166 - 08:28 PM Backport #20718 (Rejected): kraken: rgw: Truncated objects
- 08:28 PM Backport #20717 (Rejected): hammer: rgw: Truncated objects
- 08:27 PM Backport #20716 (Rejected): kraken: rgw: radosgw-admin data sync run crash
- 08:27 PM Backport #20715 (Resolved): jewel: rgw: radosgw-admin data sync run crash
- https://github.com/ceph/ceph/pull/17165
- 08:27 PM Backport #20713 (Rejected): kraken: rgw: not initialized pointer cause rgw crash with ec data pool
- 08:27 PM Backport #20712 (Resolved): jewel: rgw: not initialized pointer cause rgw crash with ec data pool
- https://github.com/ceph/ceph/pull/17164
- 08:27 PM Backport #20711 (Rejected): kraken: Segmentation fault when exporting rgw bucket in nfs-ganesha
- 08:27 PM Backport #20710 (Rejected): kraken: radosgw-admin: bucket rm with --bypass-gc and without --purge...
- 08:27 PM Backport #20709 (Resolved): jewel: radosgw-admin: bucket rm with --bypass-gc and without --purge-...
- https://github.com/ceph/ceph/pull/17159
- 07:11 PM Bug #20708 (Resolved): rgw: policy cannot parse Condition with IfExists
- The s3test test_bucket_policy_set_condition_operator_end_with_IfExists from https://github.com/ceph/s3-tests/pull/167...
- 06:07 PM Bug #20612 (Need More Info): radosgw ceases responding to list requests
- 06:07 PM Bug #20612: radosgw ceases responding to list requests
- could we have radosgw logs at --debug-rgw=20 --debug-ms=1, please?
- 06:04 PM Bug #19285 (Pending Backport): rgw: user quota did not work well on multipart upload
- 06:01 PM Bug #19870 (In Progress): rgw: bytes_send and bytes_recv in the msg of usage show returning is 0 ...
- 06:01 PM Bug #17964 (Pending Backport): rgw: multipart parts on versioned bucket create versioned bucket i...
- 05:56 PM Bug #20423 (Pending Backport): rgw: radosgw-admin data sync run crash
- 05:56 PM Bug #20201 (In Progress): radosgw refuses upload when Content-Type missing from POST policy
- 05:54 PM Bug #19904: rgw: Objects of bucket with '_' fail to sync in multisite env
- normal prio because have workaround (relaxed names)
- 05:52 PM Bug #20668: rgw multisite: cannot sync objects encrypted with SSE-C
- 05:52 PM Bug #20671: rgw multisite: objects encrypted with SSE-KMS are stored unencrypted in target zone
- 05:52 PM Bug #20542 (Pending Backport): rgw: not initialized pointer cause rgw crash with ec data pool
- 05:50 PM Bug #19739 (Pending Backport): Content-MD5 header is not validated with POST uploads
- 05:48 PM Bug #19816 (Resolved): multisite: set_latest_epoch not atomic
- 05:46 PM Bug #19817 (Pending Backport): multisite: RGWPeriodPuller does not call RGWPeriod::reflect() on n...
- 05:43 PM Bug #20107 (Pending Backport): rgw: Truncated objects
- 03:35 PM Bug #20107: rgw: Truncated objects
- merged https://github.com/ceph/ceph/pull/15673
- 05:40 PM Bug #20490 (Need More Info): rgw: can not list buckets when enable s3 website hosting and use domain
- 05:40 PM Bug #20177 (In Progress): RGW lifecycle not expiring objects due to permissions on lc pool
- 05:38 PM Bug #20177: RGW lifecycle not expiring objects due to permissions on lc pool
- It is the same as the reshard error, we should init to namespace.
The same applys to the roles pool - 03:39 PM Feature #16191: rgw:get usage S3 api extend
- merged https://github.com/ceph/ceph/pull/14287
- 02:38 PM Bug #20100 (Resolved): get wrong content when download object with specific range when compressio...
- 02:38 PM Backport #20268 (Resolved): kraken: get wrong content when download object with specific range wh...
- 11:18 AM Bug #20541 (Closed): rgw: s3 put obj hangs with ec data pool on filestore
- 10:37 AM Feature #12102: Request for increased logging in RGW, for CivetWeb.
- The request for a status page is handled upstream at https://github.com/civetweb/civetweb/issues/243 and seems to hav...
- 09:06 AM Bug #20696 (Resolved): radosgw doesn't start with civetweb handlerin Debian Stretch
- Radosgw doesn't start due to incompatibility between civetweb and openssl version >= 1.1 . There is a issue in civetw...
07/19/2017
- 09:21 PM Bug #20663: Segmentation fault when exporting rgw bucket in nfs-ganesha
- supriti singh wrote:
> Somehow my original reply in comment#6 was deleted, restoring:
> """
> Hi Supriti,
>
> T... - 03:32 PM Bug #20663: Segmentation fault when exporting rgw bucket in nfs-ganesha
- Somehow my original reply in comment#6 was deleted, restoring:
"""
Hi Supriti,
The condition is "if (flags & (FL... - 03:10 PM Bug #20663: Segmentation fault when exporting rgw bucket in nfs-ganesha
- Matt Benjamin wrote:
> Hi Supriti,
>
> I think we -lock- lat.lock unconditionally in find_latch, but should NOT b... - 02:05 PM Bug #20663: Segmentation fault when exporting rgw bucket in nfs-ganesha
- Hi Supriti,
I think we -lock- lat.lock unconditionally in find_latch, but should NOT be unlocking it. we -would- ... - 02:01 PM Bug #20663: Segmentation fault when exporting rgw bucket in nfs-ganesha
- Matt Benjamin wrote:
> hi supriti,
>
> Releasing lat.lock having taken fh->mtx is not per-se a smoking gun, as we... - 07:48 PM Bug #20688 (Pending Backport): radosgw-admin: bucket rm with --bypass-gc and without --purge-data...
- 12:54 PM Bug #20688: radosgw-admin: bucket rm with --bypass-gc and without --purge-data doesn't throw erro...
- 12:38 PM Bug #20688: radosgw-admin: bucket rm with --bypass-gc and without --purge-data doesn't throw erro...
- Fixed here : https://github.com/ceph/ceph/pull/16419
- 12:23 PM Bug #20688 (Resolved): radosgw-admin: bucket rm with --bypass-gc and without --purge-data doesn't...
- When radosgw-admin bucket rm is invoked with _--bypass-gc_ and without _--purge-data_ flag, bucket deletion is not al...
- 02:51 PM Bug #19520 (Resolved): rgw: Swift's at-root features (/crossdomain.xml, /info, /healthcheck) are ...
- 02:51 PM Backport #20031 (Resolved): kraken: rgw: Swift's at-root features (/crossdomain.xml, /info, /heal...
- 02:15 PM Feature #9493 (Resolved): Ability to disable keystone revocation polling when using UUID keystone...
- 02:15 PM Backport #19777 (Resolved): kraken: rgw: implement support for OS-REVOKE extension of OpenStack I...
- 01:33 PM Bug #20285 (Resolved): Lifecycle thread will still handle the bucket even if it has been removed.
- 01:33 PM Backport #20405 (Resolved): kraken: Lifecycle thread will still handle the bucket even if it has ...
- 01:32 PM Backport #20363 (Resolved): kraken: VersionIdMarker and NextVersionIdMarker are not returned when...
- 01:31 PM Backport #20269 (Resolved): kraken: wrong object size after copy of uncompressed multipart objects
- 01:30 PM Bug #20225 (Resolved): 'radosgw-admin usage show' listing 0 bytes_sent/received
- 01:30 PM Backport #20261 (Resolved): kraken: 'radosgw-admin usage show' listing 0 bytes_sent/received
- 01:30 PM Backport #20156 (Resolved): kraken: fix: rgw crashed caused by shard id out of range when listing...
- 01:29 PM Backport #19839 (Resolved): kraken: reduce log level of 'storing entry at' in cls_log
- 01:28 PM Backport #19766 (Resolved): kraken: rgw: when uploading the objects continuesly in the versioned ...
- 01:28 PM Bug #18828 (Resolved): RGW S3 v4 authentication issue with X-Amz-Expires
- 01:27 PM Backport #19725 (Resolved): kraken: RGW S3 v4 authentication issue with X-Amz-Expires
- 01:27 PM Backport #20293 (Resolved): kraken: multisite: log_meta on secondary zone causes continuous loop ...
- 01:26 PM Backport #19776 (Resolved): kraken: multisite: realm rename does not propagate to other clusters
- 01:26 PM Backport #19616 (Resolved): kraken: multisite: bucket zonegroup redirect not working
- 01:26 PM Backport #19614 (Resolved): kraken: multisite: rest api fails to decode large period on 'period c...
- 01:25 PM Backport #20147 (Resolved): kraken: rgw: 'gc list --include-all' command infinite loop the first ...
- 12:47 PM Bug #20686 (Fix Under Review): rgw hangs in RGWRealmReloader::reload on SIGHUP
- 11:28 AM Bug #20686: rgw hangs in RGWRealmReloader::reload on SIGHUP
- maybe the guideline is inproper.
should be changed to : rgw hang in RGWRealmReloader::reload when got SIGHUP - 11:24 AM Bug #20686: rgw hangs in RGWRealmReloader::reload on SIGHUP
- https://github.com/ceph/ceph/pull/16417
- 11:17 AM Bug #20686: rgw hangs in RGWRealmReloader::reload on SIGHUP
- radosgw sometimes hang when got SIGHUP(mostly from logrotate)
once handle SIGHUP, RGWRealmReloader::reload will b... - 11:01 AM Bug #20686 (Resolved): rgw hangs in RGWRealmReloader::reload on SIGHUP
- 08:31 AM Bug #19056: radosgw S3 AWS4 signature and keystone integration broken
- Radoslaw Zarzynski wrote:
> PR: https://github.com/ceph/ceph/pull/14885. The code will be available starting from Lu...
07/18/2017
- 11:46 PM Bug #19056 (Resolved): radosgw S3 AWS4 signature and keystone integration broken
- PR: https://github.com/ceph/ceph/pull/14885. The code will be available starting from Luminous.
- 09:01 PM Backport #19704 (Resolved): civetweb-worker segmentation fault
- 08:04 PM Backport #20674 (Rejected): kraken: rgw:multisite: RGWRadosRemoveOmapKeysCR::request_complete ret...
- 08:04 PM Backport #20673 (Resolved): jewel: rgw:multisite: RGWRadosRemoveOmapKeysCR::request_complete retu...
- https://github.com/ceph/ceph/pull/17156
- 07:25 PM Bug #20671 (Resolved): rgw multisite: objects encrypted with SSE-KMS are stored unencrypted in ta...
- When SSE-KMS encryption is used, multisite sync is able to fetch the decrypted contents of the object. However, it st...
- 06:24 PM Bug #20663: Segmentation fault when exporting rgw bucket in nfs-ganesha
- @supriti, so far I'm not hitting this assert, mounting with pseudo /demo, at demo-demo alongside unmounted demo1-demo...
- 12:38 PM Bug #20663 (In Progress): Segmentation fault when exporting rgw bucket in nfs-ganesha
- hi supriti,
Releasing lat.lock having taken fh->mtx is not per-se a smoking gun, as we have taken lat.lock at find... - 11:42 AM Bug #20663 (Resolved): Segmentation fault when exporting rgw bucket in nfs-ganesha
- We see a segmentation fault when trying to export rgw bucket using nfs-ganesha (with latest master and ceph v 12.1.0)...
- 05:10 PM Bug #20539 (Pending Backport): rgw:multisite: RGWRadosRemoveOmapKeysCR::request_complete return v...
- 02:42 PM Bug #20668 (Resolved): rgw multisite: cannot sync objects encrypted with SSE-C
- GET requests for encrypted objects require special headers for SSE-C, enforced by rgw_s3_prepare_decrypt(). Multisite...
- 01:25 PM Bug #20665 (In Progress): STREAMING-AWS4-HMAC-SHA256-PAYLOAD fails with 0 byte upload
- @sendu, probably RGW but we will root cause
matt - 01:19 PM Bug #20665 (In Progress): STREAMING-AWS4-HMAC-SHA256-PAYLOAD fails with 0 byte upload
- I am using RGW Jewel, 10.2.3-0ubuntu0.16.04.2.
I believe this version came out after STREAMING-AWS4-HMAC-SHA256-PA... - 11:41 AM Cleanup #4057: Update Admin API spec with comments
- Where are received comments? Do we need to change ceph/admin
- 09:31 AM Bug #20661 (Fix Under Review): With RGWs configured in a load balancer, quota stats cache doesn't...
- https://github.com/ceph/ceph/pull/16389
- 09:09 AM Bug #20661 (Resolved): With RGWs configured in a load balancer, quota stats cache doesn't work
- With RGWs configured in a load balancer, quota stats cache can possibly run into unbound values. We have found errors...
- 04:11 AM Bug #20658: RGW: object copied from remote src acl permission become full-control issue
- Addressed by https://github.com/ceph/ceph/pull/16127
- 03:32 AM Bug #20658 (Resolved): RGW: object copied from remote src acl permission become full-control issue
- After copy object to remote destination, all grants' permission in the acl of the object in remote destination will b...
07/17/2017
- 04:02 PM Feature #20650 (New): Support webhook for authentication
- It's very good to RGW supports defining webhook for authentication.
- 01:03 PM Bug #20289: "failed to list reshard log entries" spams log file
- Possible fix: https://github.com/ceph/ceph/pull/16193
- 06:28 AM Support #20649 (New): Need limit bandwidth for s3 user
- Hi Team,
I want to limit bandwidth on uploading s ,downloading's and video streaming for rados gateway user bucket...
07/14/2017
- 08:12 PM Backport #20642 (Rejected): kraken: multisite: lock is not released when RGWMetaSyncShardCR::full...
- 08:12 PM Backport #20641 (Resolved): jewel: multisite: lock is not released when RGWMetaSyncShardCR::full_...
- https://github.com/ceph/ceph/pull/17155
- 03:53 PM Bug #19817: multisite: RGWPeriodPuller does not call RGWPeriod::reflect() on new period
- Casey Bodley wrote:
> https://github.com/ceph/ceph/pull/14915
merged - 03:53 PM Bug #19816: multisite: set_latest_epoch not atomic
- Casey Bodley wrote:
> https://github.com/ceph/ceph/pull/14915
>
> added backport tags to related issue http://tra...
07/13/2017
- 04:28 PM Bug #20620 (Closed): rgw: RGWRados::copy_obj_to_remote_dest() bad variable lifecycle
- Not sure that this code is ever called (most likely it doesn't), but out_stream_req can be deleted without prior init...
- 04:20 PM Bug #19602 (Resolved): Multipart upload may exceed the quota
- 04:15 PM Bug #18077 (Pending Backport): multisite: lock is not released when RGWMetaSyncShardCR::full_sync...
- 12:47 PM Bug #20418: rgw: URI with \0 in the middle is silently handled as its initial part
- https://github.com/ceph/ceph/pull/15953
- 12:46 PM Bug #20418 (Fix Under Review): rgw: URI with \0 in the middle is silently handled as its initial ...
- 09:05 AM Backport #20362 (In Progress): jewel: VersionIdMarker and NextVersionIdMarker are not returned wh...
- 03:50 AM Bug #20612: radosgw ceases responding to list requests
- @greg,
does this really look like a radosgw issue to you?
Matt - 01:44 AM Bug #20612 (Can't reproduce): radosgw ceases responding to list requests
- I have 4 OSDs set up with --dmcrypt and --bluestore. Occasionally, likely due to http://tracker.ceph.com/issues/20545...
07/12/2017
- 07:13 PM Backport #20262 (In Progress): jewel: "datalog trim" can't work as expected
- 05:05 PM Bug #17164: Orphan data gets leaked on Bucket deletion
- Abhishek Varshney wrote:
> @Pavan : I am not sure why this happens. Once in a while, we see buckets with stale index... - 04:17 PM Bug #20107: rgw: Truncated objects
- Would this issue affect a situation where a multi-part upload was partially completed, cleaned, and then successfully...
- 03:52 PM Bug #20377: Upon swift bucket list, do not fetch the bucket stats by default
- Pavan Rallabhandi wrote:
> https://github.com/ceph/ceph/pull/15834
tested and merged - 03:50 PM Bug #20394: rgw: /info lacks the tempauth.account_acls key
- Radoslaw Zarzynski wrote:
> PR: https://github.com/ceph/ceph/pull/15887.
merged - 03:47 PM Bug #19739: Content-MD5 header is not validated with POST uploads
- Orit Wasserman wrote:
> s3test tracker: http://tracker.ceph.com/issues/20213
merged - 02:00 PM Backport #20513 (In Progress): jewel: Stale bucket index entries are left over after object delet...
- 12:24 PM Backport #20513: jewel: Stale bucket index entries are left over after object deletions
- https://github.com/ceph/ceph/pull/16289
- 01:56 PM Backport #19613 (In Progress): jewel: multisite: rest api fails to decode large period on 'period...
- 01:27 PM Backport #19758 (In Progress): jewel: multisite: after CreateBucket is forwarded to master, local...
- 01:22 PM Backport #19759 (In Progress): kraken: multisite: after CreateBucket is forwarded to master, loca...
- 12:54 PM Bug #20542: rgw: not initialized pointer cause rgw crash with ec data pool
- Fixed in master https://github.com/ceph/ceph/pull/16177
- 11:11 AM Bug #20339: rgw stops serving requests after receiving SIGHUP
- Hi Casey,
No, I don't see the 'Resuming...' message when the frontend crashes. For now we've just resorted to rest... - 02:47 AM Backport #19563 (In Progress): jewel: rgw: folders starting with "_" underscore are not in bucket...
- 02:39 AM Backport #18903 (Need More Info): jewel: rgw: first write also tries to read object
- non-trivial backport; needs RGW developer
- 02:36 AM Backport #19532: jewel: radosgw-admin: unable to set global quotas in RGWPeriodConfig
- non-trivial backport; needs an RGW developer
- 02:31 AM Backport #19177 (Need More Info): jewel: anonymous user's error code of getting object is not con...
- Compilation error.
- 02:21 AM Feature #9493 (Pending Backport): Ability to disable keystone revocation polling when using UUID ...
- 02:21 AM Backport #19772 (Resolved): jewel: rgw: swift: disable revocation thread under certain circumstances
07/11/2017
- 08:20 PM Backport #20588 (In Progress): jewel: rgw: RGWHTTPArgs::sys_get() doesn't return exists correctly
- 08:19 PM Backport #20588 (Resolved): jewel: rgw: RGWHTTPArgs::sys_get() doesn't return exists correctly
- https://github.com/ceph/ceph/pull/16268
- 08:18 PM Backport #20587 (Rejected): jewel: radosgw/swift emulate split read/write acls?
- https://github.com/ceph/ceph/pull/20257
- 08:18 PM Backport #20586 (Rejected): kraken: radosgw/swift emulate split read/write acls?
- 08:15 PM Backport #20408 (In Progress): jewel: rgw: multipart copy-part encounts 404 error by using S3 jav...
- 06:15 PM Backport #20408: jewel: rgw: multipart copy-part encounts 404 error by using S3 java sdk
- https://github.com/ceph/ceph/pull/16266
- 07:02 PM Bug #16072 (Pending Backport): rgw: RGWHTTPArgs::sys_get() doesn't return exists correctly
- fix needed on jewel, where this causes swift requests to crash when issued with a system user. jewel backport pr: htt...
- 03:57 PM Backport #19736: radosgw/s3 chunked transfer encodings and fast_forward_request
- h3. description
fast_forward_request does not properly consume unread input when chunked transfer encodings are us... - 10:17 AM Support #20253: Ceph RGW Users (rgw keystone implicit tenants)
- This ticket has been resisted at http://tracker.ceph.com/issues/20570 and improved for clarity. please close this dup...
- 10:16 AM Bug #20570 (Closed): Ceph - Keystone Implicit Tenants
- With RGW configured with 'rgw keystone implicit tenants = true' and associated keystone options set, RGW checks a Key...
- 08:35 AM Bug #19289 (Pending Backport): radosgw/swift emulate split read/write acls?
07/10/2017
- 08:47 PM Backport #20565 (Rejected): kraken: rgw: folders starting with "_" underscore are not in bucket i...
- 01:27 PM Bug #20558 (New): rgw cls list_plain_entries() error code gets distorted by increment
- the result from list_plain_entries() is incremented by one for identifying truncation. This also happens for errors (...
- 08:40 AM Bug #20542: rgw: not initialized pointer cause rgw crash with ec data pool
- How to reproduce with vstart.sh environment (luminous branch v12.1.0):
$ ../src/vstart.sh -d -n -x -l --osd_num 5 ... - 07:41 AM Bug #20541: rgw: s3 put obj hangs with ec data pool on filestore
- Thanks, exactly, my mistake.
Not a bug, please, close ticket.
07/09/2017
07/08/2017
- 08:21 PM Bug #20463: rgw: AWSv4 encoding/signature problem, can happen with listobjects marker.
- Jewel has the same behavior - so yes it probably needs the same fix.
- 06:00 PM Backport #18287 (Closed): kraken: multisite: coroutine deadlock in RGWMetaSyncCR after ECANCELED ...
- Changing status to "Closed" so it doesn't turn up in the "missing target version" saved search.
- 05:14 PM Bug #20368 (Fix Under Review): rados/upgrade rgw swift test fails
- https://github.com/ceph/ceph/pull/16228
- 10:22 AM Bug #20541: rgw: s3 put obj hangs with ec data pool on filestore
- you have to set ruleset-failure-domain to osd. there are not enough hosts (3+2). your cluster is not in clean+active ...
- 06:43 AM Bug #20527: v2 presigned URLs don't work with radosgw.
- @mwatts I can't get it to fail in my production setups, maybe your setup had a weird rgw dns name or zone config?
07/07/2017
- 08:40 PM Bug #20368: rados/upgrade rgw swift test fails
- Sage is testing a fix in -https://github.com/ceph/ceph/pull/16222-
- 07:17 PM Bug #20373 (Need More Info): "test_get_manifest (test.functional.tests.TestDlo)*" failed in rgw
- Yuri, I checked and "master_2017_6_21" (the branch this test ran against) does not contain commit 9eec1b4 which fixes...
- 03:51 PM Bug #20373: "test_get_manifest (test.functional.tests.TestDlo)*" failed in rgw
- To the extent that the SLO/DLO swift tests are running on kraken (as part of whichever upgrade test), the bug should ...
- 05:35 PM Bug #20527: v2 presigned URLs don't work with radosgw.
- In my case v2 worked in both calling formats
- 07:12 AM Bug #20527: v2 presigned URLs don't work with radosgw.
- I checked: jewel has the same behavior.
I don't think Robin is describing different behavior than what I saw. We ... - 05:02 PM Bug #19432 (Pending Backport): rgw: folders starting with "_" underscore are not in bucket index
- 04:36 PM Bug #19432: rgw: folders starting with "_" underscore are not in bucket index
- https://github.com/ceph/ceph/pull/15916 merged 7/7/17
- 04:39 PM Bug #20307: website request URL without slashes (for folder) returns error
- Casey Bodley wrote:
> https://github.com/ceph/ceph/pull/15703
merged
- 03:03 PM Bug #20542 (In Progress): rgw: not initialized pointer cause rgw crash with ec data pool
- 02:44 PM Bug #20542: rgw: not initialized pointer cause rgw crash with ec data pool
- https://github.com/ceph/ceph/pull/16177
- 02:39 PM Bug #20542 (Resolved): rgw: not initialized pointer cause rgw crash with ec data pool
- In RGWPutObjProcessor_Atomic::complete_writing_data()
with pending_data_bl.length() > 0 and next_part_ofs==data_ofs
... - 02:13 PM Bug #20539: rgw:multisite: RGWRadosRemoveOmapKeysCR::request_complete return val is wrong
- 03:29 AM Bug #20539: rgw:multisite: RGWRadosRemoveOmapKeysCR::request_complete return val is wrong
- https://github.com/ceph/ceph/pull/16197
- 03:27 AM Bug #20539 (Resolved): rgw:multisite: RGWRadosRemoveOmapKeysCR::request_complete return val is wrong
- RGWRadosRemoveOmapKeysCR::request_complete return 'rval'. rval is a member variable, but it not initiated and used. S...
- 09:15 AM Bug #20541: rgw: s3 put obj hangs with ec data pool on filestore
- Not only on filestore, same on bluestore.
- 07:38 AM Bug #20541 (Closed): rgw: s3 put obj hangs with ec data pool on filestore
- luminous 12.1.0 and master(676e1c9b7cc81467c527e9eae118cbfb8b29ca34)
How to reproduce:
$ ../src/vstart.sh -d -n... - 07:50 AM Bug #17655 (Resolved): multisite: possible infinite loop in RGWFetchAllMetaCR
- 07:50 AM Backport #19322 (Resolved): kraken: multisite: possible infinite loop in RGWFetchAllMetaCR
- 07:49 AM Bug #18333 (Resolved): rgw: Realm set does not create a new period
- 07:49 AM Backport #18499 (Resolved): kraken: rgw: Realm set does not create a new period
- 07:48 AM Bug #18631 (Resolved): rgw crashes when updating period with placement group
- 07:48 AM Backport #18772 (Resolved): kraken: rgw crashes when updating period with placement group
- 07:45 AM Bug #18364 (Resolved): rgw: usage stats and quota are not operational for multi-tenant users
- 07:45 AM Backport #18843 (Resolved): kraken: rgw: usage stats and quota are not operational for multi-tena...
- 07:44 AM Backport #18904 (Resolved): kraken: rgw: first write also tries to read object
- 07:43 AM Backport #19180 (Resolved): kraken: rgw: 204 No Content is returned when putting illformed Swift'...
- 07:43 AM Bug #18484 (Resolved): the swift container acl does not support field ".ref"
- 07:42 AM Backport #18909 (Resolved): kraken: rgw: the swift container acl does not support field ".ref"
- 07:37 AM Bug #18331 (Resolved): RGW leaking data
- 07:36 AM Backport #19047 (Resolved): kraken: RGW leaking data
- 07:24 AM Bug #18829 (Resolved): RGW S3 v4 authentication issue with X-Amz-Expires
- 07:24 AM Backport #19477 (Resolved): kraken: rgw: S3 v4 authentication issue with X-Amz-Expires
- 07:23 AM Bug #18725 (Resolved): "zonegroupmap set" does not work
- 07:22 AM Backport #19479 (Resolved): kraken: rgw: "zonegroupmap set" does not work
- 07:21 AM Bug #19283 (Resolved): Error parsing xml when get bucket lifecycle
- 07:21 AM Backport #19534 (Resolved): kraken: rgw: Error parsing xml when get bucket lifecycle
- 07:19 AM Bug #19599 (Resolved): multisite: fetch_remote_obj() gets wrong version when copying from remote
- 07:19 AM Backport #19608 (Resolved): kraken: multisite: fetch_remote_obj() gets wrong version when copying...
- 07:18 AM Bug #19876 (Resolved): multisite: bi_list() decode failures
- 07:18 AM Backport #20015 (Resolved): kraken: multisite: bi_list() decode failures
- 07:03 AM Backport #19472 (Resolved): kraken: cannot cover the object expiration
- 07:01 AM Bug #19027 (Resolved): multisite: EPERM when trying to read SLO objects as system/admin user
- 07:01 AM Backport #19474 (Resolved): jewel: multisite: EPERM when trying to read SLO objects as system/adm...
- 07:01 AM Backport #19475 (Resolved): kraken: rgw: multisite: EPERM when trying to read SLO objects as syst...
- 06:59 AM Bug #18852 (Resolved): swift API: cannot disable object versioning with empty X-Versions-Location
- 06:59 AM Backport #19175 (Resolved): kraken: swift API: cannot disable object versioning with empty X-Vers...
- 06:57 AM Bug #19498 (Resolved): `radosgw-admin zone create` command with specified zone-id creates a zone ...
- 06:57 AM Backport #19524 (Resolved): kraken: rgw: 'radosgw-admin zone create' command with specified zone-...
- 06:56 AM Bug #19443 (Resolved): Response header of swift API returned by radosgw does not contain "x-opens...
- 06:55 AM Backport #19573 (Resolved): kraken: rgw: Response header of swift API returned by radosgw does no...
- 05:03 AM Bug #17164: Orphan data gets leaked on Bucket deletion
- @Pavan : I am not sure why this happens. Once in a while, we see buckets with stale index entries. This mostly happen...
07/06/2017
- 07:20 PM Backport #20531 (Rejected): kraken: default zone should not log meta
- 07:20 PM Backport #20530 (Resolved): jewel: default zone should not log meta
- https://github.com/ceph/ceph/pull/17148
- 06:19 PM Bug #17164: Orphan data gets leaked on Bucket deletion
- (ignore this one)
- 06:18 PM Bug #17164: Orphan data gets leaked on Bucket deletion
- 06:15 PM Bug #20284 (In Progress): 【rgw】multipart object leak when re-upload the same number part
- 06:12 PM Bug #20378: rgw: fix copy object err when bucket with specified placement rule
- 06:11 PM Bug #20378 (Fix Under Review): rgw: fix copy object err when bucket with specified placement rule
- 06:09 PM Bug #20307: website request URL without slashes (for folder) returns error
- 06:08 PM Bug #20308: implement stringlike wildcard for bucket policy
- (waiting for update from joke-lee)
- 06:08 PM Bug #20320 (Duplicate): avoid setting log_meta for single-zone configurations in RGWZoneGroup::po...
- 06:08 PM Bug #20357 (Pending Backport): default zone should not log meta
- 06:08 PM Bug #20357: default zone should not log meta
- https://github.com/ceph/ceph/pull/15777
- 06:01 PM Bug #20418: rgw: URI with \0 in the middle is silently handled as its initial part
- @radoslaw, is this in progress?
Matt - 06:01 PM Bug #19042: rgw: multiple zonegroups: bucket can't be created if the user name isn't registered o...
- We will need to forward all user creation operation to the master zonegroup.
At the moment all users needs to be cre... - 05:57 PM Bug #20289: "failed to list reshard log entries" spams log file
- We can use namespace instead of a new pool to avoid this
- 05:53 PM Bug #20466: testSmallUpload:111->writeRenameReadCompare:80 » AccessControl
- @marcus, could you have a look at this?
Matt - 05:51 PM Bug #20447: rgw: S3 object uploads using aws-sdk-java 1.11 to do chunked transfers fail.
- 05:50 PM Bug #20462: rgw_file: support compression
- 05:50 PM Bug #20468: rgw: admin API should return NoSuchUser for actions on non-existed user
- 05:49 PM Bug #19448: GET bucket policy response format
- @joke lee, not planned to backport bucket policy to jewel, L is nearly upon us
Matt - 05:47 PM Backport #20405 (In Progress): kraken: Lifecycle thread will still handle the bucket even if it h...
- 05:46 PM Bug #20463: rgw: AWSv4 encoding/signature problem, can happen with listobjects marker.
- 05:45 PM Bug #20234 (In Progress): rgw: shadow objects are sometimes not removed
- 05:45 PM Bug #20501 (Fix Under Review): S3 v4 auth fails when query string contains " "
- 05:44 PM Bug #20527 (In Progress): v2 presigned URLs don't work with radosgw.
- 05:43 PM Bug #20527: v2 presigned URLs don't work with radosgw.
- @mwatts, can you update with info from last night?
Matt - 06:13 AM Bug #20527: v2 presigned URLs don't work with radosgw.
- Either you're doing something weird, or it's broken between Jewel & master.
Here is working output on Jewel (not q... - 05:35 AM Bug #20527: v2 presigned URLs don't work with radosgw.
- I should mention: awscli does v4 signatures by default, but when properly coaxed (by giving it a specially prepared "...
- 05:27 AM Bug #20527 (Can't reproduce): v2 presigned URLs don't work with radosgw.
- s3cmd only supports v2 style presigned URLs (see https://github.com/s3tools/s3cmd/issues/765 ).
Also, it only suppor... - 05:42 PM Bug #20505 (Resolved): RGW: radosgw-admin bi list - ERROR: bi_list(): (4) Interrupted system call
- 05:36 PM Bug #20505: RGW: radosgw-admin bi list - ERROR: bi_list(): (4) Interrupted system call
- This issue seems to be fixed in 10.2.8
see http://tracker.ceph.com/issues/20014 - 11:53 AM Bug #20505: RGW: radosgw-admin bi list - ERROR: bi_list(): (4) Interrupted system call
- Hi,
I am not sure how you are using bi_list for backup, I would happy to hear more details to understand.
Is the ... - 05:41 PM Bug #19277: Legacy buckets cannot enable versioning
- @yehuda, move to WONTFIX?
Matt - 05:37 PM Backport #20363 (In Progress): kraken: VersionIdMarker and NextVersionIdMarker are not returned w...
- 05:36 PM Backport #20347 (In Progress): kraken: rgw: meta sync thread crash at RGWMetaSyncShardCR
- 05:18 PM Backport #20269 (In Progress): kraken: wrong object size after copy of uncompressed multipart obj...
- 05:16 PM Backport #20268 (In Progress): kraken: get wrong content when download object with specific range...
- 03:32 PM Bug #20368: rados/upgrade rgw swift test fails
- @Casey: Yes, jewel has a patched swift.py that simply (hard-coded) checks out the ceph-jewel branch. I have applied a...
- 02:33 PM Bug #20368: rados/upgrade rgw swift test fails
- Nathan Cutler wrote:
> @Yehuda: Thanks! Jewel already has a workaround in place, so kraken backport is sufficient I ... - 03:08 PM Backport #20520 (Need More Info): kraken: rados/upgrade rgw swift test fails
- 02:54 PM Backport #20263 (In Progress): kraken: "datalog trim" can't work as expected
- 02:53 PM Backport #20261 (In Progress): kraken: 'radosgw-admin usage show' listing 0 bytes_sent/received
- 02:51 PM Backport #20156 (In Progress): kraken: fix: rgw crashed caused by shard id out of range when list...
- 10:21 AM Backport #20031 (In Progress): kraken: rgw: Swift's at-root features (/crossdomain.xml, /info, /h...
- 09:40 AM Backport #19840 (In Progress): kraken: civetweb frontend segfaults in Luminous
- 09:36 AM Backport #19839 (In Progress): kraken: reduce log level of 'storing entry at' in cls_log
- 09:34 AM Feature #19499 (New): rgw: implement support for OS-REVOKE extension of OpenStack Identity API v3
- @Marcus, @Radoslaw: My understanding is that PR#14501 *fixes* #9493 and is only *related* to this tracker. Please cor...
- 09:27 AM Backport #19777 (In Progress): kraken: rgw: implement support for OS-REVOKE extension of OpenStac...
- 09:26 AM Backport #19766 (In Progress): kraken: rgw: when uploading the objects continuesly in the version...
- 09:24 AM Backport #19725 (In Progress): kraken: RGW S3 v4 authentication issue with X-Amz-Expires
- 09:20 AM Backport #20293 (In Progress): kraken: multisite: log_meta on secondary zone causes continuous lo...
- 09:20 AM Backport #19776 (In Progress): kraken: multisite: realm rename does not propagate to other clusters
- 09:19 AM Backport #19616 (In Progress): kraken: multisite: bucket zonegroup redirect not working
- 09:19 AM Backport #19614 (In Progress): kraken: multisite: rest api fails to decode large period on 'perio...
- 09:15 AM Backport #19759 (Need More Info): kraken: multisite: after CreateBucket is forwarded to master, l...
- Non-trivial backport
see also http://tracker.ceph.com/issues/19745#note-6 - 07:12 AM Bug #19754 (Resolved): rgw: S3 object uploads using the AWSv4's multi-chunk mode hang RadosGW
- 07:12 AM Backport #19837 (Resolved): kraken: rgw: S3 object uploads using the AWSv4's multi-chunk mode han...
- 07:08 AM Bug #19249 (Resolved): rgw: unsafe access in RGWListBucket_ObjStore_SWIFT::send_response()
- 06:54 AM Backport #19574 (Resolved): kraken: rgw: unsafe access in RGWListBucket_ObjStore_SWIFT::send_resp...
- 06:40 AM Bug #19019 (Resolved): multisite: RGWMetaSyncShardControlCR gives up on EIO
- 06:40 AM Backport #19160 (Resolved): kraken: multisite: RGWMetaSyncShardControlCR gives up on EIO
07/05/2017
- 11:15 PM Bug #20234: rgw: shadow objects are sometimes not removed
- Please see this PR:
https://github.com/ceph/ceph/pull/16145
Still needs more testing. - 08:59 AM Bug #20234: rgw: shadow objects are sometimes not removed
- It seems that this issue could be exploited to DoS a cluster, a client can circumvent their quota and fill up all spa...
- 09:33 PM Backport #20147 (In Progress): kraken: rgw: 'gc list --include-all' command infinite loop the fir...
- 06:13 PM Bug #19904 (In Progress): rgw: Objects of bucket with '_' fail to sync in multisite env
- 06:12 PM Bug #20490: rgw: can not list buckets when enable s3 website hosting and use domain
- Can you provide more logs?...
- 06:09 PM Bug #20368: rados/upgrade rgw swift test fails
- Oops, the kraken backport is non-trivial.
- 05:59 PM Bug #20368 (Pending Backport): rados/upgrade rgw swift test fails
- @Yehuda: Thanks! Jewel already has a workaround in place, so kraken backport is sufficient I guess.
- 05:47 PM Bug #20368: rados/upgrade rgw swift test fails
- Note original fix: https://github.com/ceph/ceph/pull/15060
- 05:46 PM Bug #20368: rados/upgrade rgw swift test fails
- Looking at the current failures, it's all happening when testing kraken (directly, or when doing an upgrade test). Cu...
- 06:25 AM Bug #20368: rados/upgrade rgw swift test fails
- /a/kchai-2017-07-05_03:23:34-rados-wip-kefu-testing-distro-basic-mira/1362323
- 05:59 PM Backport #20520 (Resolved): kraken: rados/upgrade rgw swift test fails
- https://github.com/ceph/ceph/pull/16131
- 05:55 PM Backport #20519 (Rejected): kraken: Stale bucket index entries are left over after object deletions
- 05:42 PM Backport #20513 (Resolved): jewel: Stale bucket index entries are left over after object deletions
- https://github.com/ceph/ceph/pull/16289
- 05:36 PM Bug #20380 (Pending Backport): Stale bucket index entries are left over after object deletions
- 09:13 AM Bug #20505 (Resolved): RGW: radosgw-admin bi list - ERROR: bi_list(): (4) Interrupted system call
- When trying to make a backup of a bucket index, it fails with
# radosgw-admin -n client.radosgw.be-west-3 bi list... - 01:15 AM Bug #20501: S3 v4 auth fails when query string contains " "
- https://github.com/ceph/ceph/pull/16116
- 01:12 AM Bug #20501 (Resolved): S3 v4 auth fails when query string contains " "
- If query string contains " ", the v4 auth will fail.
07/04/2017
- 05:56 PM Bug #20368: rados/upgrade rgw swift test fails
- http://pulpito.front.sepia.ceph.com/smithfarm-2017-07-04_10:46:32-rados-wip-kraken-backports-distro-basic-smithi/1359...
- 11:47 AM Bug #20368: rados/upgrade rgw swift test fails
- Also fails in kraken 11.2.1 integration testing: http://pulpito.front.sepia.ceph.com/smithfarm-2017-07-04_09:46:39-ra...
- 01:43 PM Feature #20496 (New): Support x-amz-replication-status for multisite
- - Support x-amz-replication-status for multisite
Currently, there is no way to verify replication status for an ob... - 11:38 AM Backport #20078: rgw jewel valgrind failure: make_params
- h3. description
http://pulpito.ceph.com/abhi-2017-05-23_12:16:16-rgw-wip-jewel-backports-mon-sortbitwise-distro-ba... - 11:38 AM Backport #20078 (Resolved): rgw jewel valgrind failure: make_params
- 11:33 AM Bug #18980 (Resolved): rgw: "cluster [WRN] bad locator @X on object @X...." in cluster log
- 11:33 AM Backport #19212 (Resolved): kraken: rgw: "cluster [WRN] bad locator @X on object @X...." in clust...
- 11:23 AM Backport #19574 (In Progress): kraken: rgw: unsafe access in RGWListBucket_ObjStore_SWIFT::send_r...
- 11:23 AM Backport #19533: kraken: rgw: radosgw-admin: unable to set global quotas in RGWPeriodConfig
- Needs an RGW developer
- 11:22 AM Backport #19475 (In Progress): kraken: rgw: multisite: EPERM when trying to read SLO objects as s...
- 09:47 AM Support #19308: 416 when Docker registry + Swift + rados gatway + ceph/daemon
- Hi
I can't help, but I can at least report that we are seeing the exact same problem.
Since we upgraded to runn... - 08:03 AM Backport #19661 (In Progress): kraken: rgw_file: fix readdir after dir-change
- 08:02 AM Backport #19663 (In Progress): kraken: rgw_file: fix event expire check, don't expire directories...
- 08:01 AM Backport #20158 (In Progress): kraken: rgw_file: handle chunked readdir
- 02:32 AM Bug #20490 (Closed): rgw: can not list buckets when enable s3 website hosting and use domain
- hi all,
i rencently test s3 website hosting on ceph radosgw, and i found some problem.
in my ceph.conf
...
07/03/2017
- 10:02 PM Feature #19804 (Resolved): APIs to support Ragweed suite
- 10:02 PM Backport #19809 (Resolved): kraken: APIs to support Ragweed suite
- 08:28 PM Feature #16602 (Resolved): rgw: sending Content-Length in 204 and 304 responses should be control...
- 08:28 PM Backport #18985 (Resolved): kraken: rgw: sending Content-Length in 204 and 304 responses should b...
07/02/2017
- 01:32 PM Bug #20463: rgw: AWSv4 encoding/signature problem, can happen with listobjects marker.
- do we need a fix for jewel?
07/01/2017
- 08:39 PM Backport #20483 (Rejected): jewel: rgw: s3 API does not honor rgw_keystone_implicit_tenants when ...
- 08:39 PM Backport #20482 (Rejected): kraken: rgw: s3 API does not honor rgw_keystone_implicit_tenants when...
- 08:37 PM Backport #20481 (Rejected): hammer: rgw log includes zero byte sometimes
- 08:37 PM Backport #20480 (Resolved): jewel: rgw log includes zero byte sometimes
- https://github.com/ceph/ceph/pull/17151
- 08:37 PM Backport #20479 (Rejected): kraken: rgw: connection reset or radosgw crashed when download large ...
- 01:44 PM Bug #20368: rados/upgrade rgw swift test fails
- This still fails every time. Latest:
/a/sage-2017-07-01_06:55:28-rados-wip-sage-testing2-distro-basic-smithi/1351...
06/30/2017
- 09:18 PM Feature #20473 (Resolved): RGW Multi-site radosgw-admin sync status and radosgw-admin bucket sync...
- - RGW Multi-site radosgw-admin sync status and radosgw-admin bucket sync status commands addition of 'detail' flag
... - 08:55 PM Bug #20449: tests: rados/thrash...rgw_snaps.yaml transient failure: Timeout exception after green...
- So this is a bug in the the qa-suite python code, right? That's not a RADOS issue... :)
- 07:29 AM Bug #20449: tests: rados/thrash...rgw_snaps.yaml transient failure: Timeout exception after green...
- ...
- 06:01 AM Bug #20449: tests: rados/thrash...rgw_snaps.yaml transient failure: Timeout exception after green...
- The failure is transient.
- 07:31 PM Bug #20201: radosgw refuses upload when Content-Type missing from POST policy
- Have a working setup for testing this. Proposed quick-fix by Y. doesn't seem quite right, but working through it. W...
- 05:47 PM Bug #20380: Stale bucket index entries are left over after object deletions
- Yes, I think follows that there could be other ceph::timespan problems. I have so far grepped through the Ceph sourc...
- 09:50 AM Bug #20380: Stale bucket index entries are left over after object deletions
- Yes Matt, the fix seems to take effect. Thanks for the quick turnaround.
I was going through the commit where the ... - 02:35 PM Bug #20468 (In Progress): rgw: admin API should return NoSuchUser for actions on non-existed user
- 06:08 AM Bug #20468: rgw: admin API should return NoSuchUser for actions on non-existed user
- pull request:
https://github.com/ceph/ceph/pull/16033 - 06:02 AM Bug #20468 (Resolved): rgw: admin API should return NoSuchUser for actions on non-existed user
- According to http://docs.ceph.com/docs/master/radosgw/adminops/#standard-error-responses, there shoud be "NoSuchUser"...
- 10:22 AM Bug #19448: GET bucket policy response format
- Casey Bodley wrote:
> there is work in progress at https://github.com/ceph/ceph/pull/14307 that implements this GetB...
06/29/2017
- 11:53 PM Bug #20466 (Closed): testSmallUpload:111->writeRenameReadCompare:80 » AccessControl
- I am seeing 2 s3a hadoop intergration tests that fail with 2.7.3, previously based on the command line options these ...
- 07:31 PM Bug #20380: Stale bucket index entries are left over after object deletions
- Hi Pavan,
I haven't proved this resolves, but I -did- prove (in a non-reproducer scenario) that the tag_timeout va... - 05:54 PM Bug #20380 (In Progress): Stale bucket index entries are left over after object deletions
- 12:43 PM Bug #20380: Stale bucket index entries are left over after object deletions
- Orit Wasserman wrote:
> Pavan, are you using versioning?
No Orit, we are not using object versioning. One interes... - 10:45 AM Bug #20380: Stale bucket index entries are left over after object deletions
- Pavan, are you using versioning?
- 07:15 PM Backport #19563 (New): jewel: rgw: folders starting with "_" underscore are not in bucket index
- Attempted backport -https://github.com/ceph/ceph/pull/14368- was closed because it wasn't a cherry-pick.
- 06:35 PM Bug #20037 (Pending Backport): rgw log includes zero byte sometimes
- 03:20 PM Bug #20037: rgw log includes zero byte sometimes
- Fix has been merged, please backport to Jewel.
- 06:25 PM Bug #17779 (Pending Backport): rgw: s3 API does not honor rgw_keystone_implicit_tenants when keys...
- Backport is not trivial
- 06:17 PM Bug #20098 (Pending Backport): rgw: connection reset or radosgw crashed when download large zero ...
- 05:56 PM Bug #19904 (Fix Under Review): rgw: Objects of bucket with '_' fail to sync in multisite env
- 05:55 PM Bug #19739: Content-MD5 header is not validated with POST uploads
- 05:54 PM Bug #20423: rgw: radosgw-admin data sync run crash
- 05:43 PM Bug #20177: RGW lifecycle not expiring objects due to permissions on lc pool
- http://lists.ceph.com/pipermail/ceph-users-ceph.com/2017-June/018933.html
Need to make sure that we log errors, so... - 09:27 AM Feature #19917: radosgw access log is lacking useful information
- https://github.com/ceph/ceph/pull/15263
- 06:31 AM Bug #20463: rgw: AWSv4 encoding/signature problem, can happen with listobjects marker.
- I've made this PR for ceph: https://github.com/ceph/ceph/pull/15998 - contains just the patch I posted above.
- 12:46 AM Bug #20384 (Resolved): rgw: TempURL returns wrong error code for expired links
- Merged: bdf7fa8279a5ae1287592e714439f9faef8f5a60
06/28/2017
- 10:23 PM Bug #20463 (Resolved): rgw: AWSv4 encoding/signature problem, can happen with listobjects marker.
- For buckets containing object names that contain / and %, it is
possible in some circumstances (at least with boto)... - 09:59 PM Bug #20447: rgw: S3 object uploads using aws-sdk-java 1.11 to do chunked transfers fail.
- I've submitted a PR with the patch I posted here. https://github.com/ceph/ceph/pull/15965
- 07:53 AM Bug #20447 (Resolved): rgw: S3 object uploads using aws-sdk-java 1.11 to do chunked transfers fail.
- As of 1.11 aws-sdk-java uses v4 signatures by default, and optionally does
chunked transfers. When it decides to do... - 09:57 PM Bug #17076: AWS S3 Version 4 signatures fail sometimes.
- I just fixed another problem with chunked transfers and AWSv4 signatures - see http://tracker.ceph.com/issues/20447 ....
- 09:31 PM Bug #20462: rgw_file: support compression
- PR added: https://github.com/ceph/ceph/pull/15989
- 09:25 PM Bug #20462 (Resolved): rgw_file: support compression
- The compression change did not add support for initializing compression filters in NFS write path.
- 08:13 PM Backport #20458 (In Progress): jewel: infinite loop in rest api for datalog list
- 08:12 PM Backport #20458 (Resolved): jewel: infinite loop in rest api for datalog list
- https://github.com/ceph/ceph/pull/15988
- 08:12 PM Backport #20459 (Rejected): kraken: infinite loop in rest api for datalog list
- 08:10 PM Bug #20386: infinite loop in rest api for datalog list
- added trivial jewel backport at https://github.com/ceph/ceph/pull/15988
- 08:07 PM Bug #20386 (Pending Backport): infinite loop in rest api for datalog list
- 07:43 PM Bug #20386: infinite loop in rest api for datalog list
- opened https://github.com/ceph/ceph/pull/15983 with some additional changes
- 06:41 PM Backport #19843 (In Progress): kraken: Add custom user data support in bucket index
- 06:21 PM Backport #19843 (Fix Under Review): kraken: Add custom user data support in bucket index
- https://github.com/ceph/ceph/pull/15985
- 03:41 PM Bug #20395: rgw: X-Account-Access-Control of Swift API should be omitted if there is no ACL
- pr merged - does it need backport? if not, please resolve
- 02:12 PM Bug #20423 (Fix Under Review): rgw: radosgw-admin data sync run crash
- 10:38 AM Bug #20449 (Closed): tests: rados/thrash...rgw_snaps.yaml transient failure: Timeout exception af...
- This happened in jewel 10.2.8 integration testing.
Test: ... - 10:18 AM Backport #19844 (In Progress): jewel: Add custom user data support in bucket index
- 10:13 AM Bug #20448 (Resolved): rgw: crash with rgw_run_sync_thread=false
- Segfault at:
main()
rest_filter()
sync_module is empty because it was not created in RGWRados::init_compl...
06/27/2017
- 11:47 PM Bug #20289: "failed to list reshard log entries" spams log file
- I encountered this same issue when upgrading from Jewel to Luminous 12.1.0 RC1.
I was able to correct it by fixing... - 03:09 PM Bug #20372 (Resolved): "s3tests.functional.test_s3.test_post_object*" failed in rgw
- 12:10 PM Bug #19739: Content-MD5 header is not validated with POST uploads
- s3test tracker: http://tracker.ceph.com/issues/20213
- 11:55 AM rgw-testing Feature #20429 (New): Add haproxy setup to rgw teuthology suite
- We need to add the ability to test rgw (including multisite) with haproxy
- 09:30 AM Backport #19844: jewel: Add custom user data support in bucket index
- @Pavan: Please ask the RGW developers (Yehuda et al.)
- 08:46 AM Backport #19844: jewel: Add custom user data support in bucket index
- Nathan Cutler wrote:
> Ordinarily we don't backport features to stable releases, but assigning to Pavan if he wants ... - 03:02 AM Bug #20423: rgw: radosgw-admin data sync run crash
- https://github.com/ceph/ceph/pull/15938
- 02:52 AM Bug #20423 (Resolved): rgw: radosgw-admin data sync run crash
- steps to reproduce:
1、setup a multisite env with two zones in a zonegroup
2、put some buckets and objects in master ... - 12:47 AM Bug #19042: rgw: multiple zonegroups: bucket can't be created if the user name isn't registered o...
- Is this related to #19756 / #19757 ?
06/26/2017
- 04:25 PM Backport #20014 (Resolved): jewel: multisite: bi_list() decode failures
- 02:38 PM Bug #20418 (Resolved): rgw: URI with \0 in the middle is silently handled as its initial part
- > $ curl -i "$publicURL/abc%00de" -X PUT -H "X-Auth-Token: $token"
> HTTP/1.1 201 Created
> Content-Length: 0
> X... - 11:54 AM Bug #19432 (Fix Under Review): rgw: folders starting with "_" underscore are not in bucket index
- 11:53 AM Bug #19432: rgw: folders starting with "_" underscore are not in bucket index
- https://github.com/ceph/ceph/pull/15916
- 11:23 AM Bug #20166: RGW: Bad error handling when tail object is missing
- No, I tried to dig a bit deeper into how the error is handled, but no success yet.
- 11:19 AM Bug #20107: rgw: Truncated objects
- Added a PR for my patch, too, assuming both are needed for a complete solution.
https://github.com/ceph/ceph/pull/...
Also available in: Atom