Project

General

Profile

Activity

From 10/25/2016 to 11/23/2016

11/23/2016

08:54 PM Backport #17840 (Resolved): hammer: rgw: the value of total_time is wrong in the result of 'rados...
Nathan Cutler
07:47 PM Bug #17897 (Pending Backport): rgw: gc records empty entries
Orit Wasserman
11:20 AM Bug #17897 (In Progress): rgw: gc records empty entries
Would like to propose a PR for this unless someone else already did. Pavan Rallabhandi
03:09 PM Bug #16597 (Pending Backport): the response element "X-Timestamp" of swift stat bucket api is zero
Loïc Dachary
09:41 AM Bug #16435 (Resolved): COPY broke multipart files uploaded under dumpling
Nathan Cutler
09:41 AM Backport #17123 (Resolved): hammer: COPY broke multipart files uploaded under dumpling
Nathan Cutler
09:40 AM Bug #13452 (Resolved): RGW Swift API: ranged request on a DLO provides wrong values in Content-Ra...
Nathan Cutler
09:40 AM Backport #16594 (Resolved): hammer: RGW Swift API: ranged request on a DLO provides wrong values ...
Nathan Cutler
09:17 AM Bug #13598 (Resolved): radosgw-admin: inconsistency in uid/email handling
Nathan Cutler
09:17 AM Backport #16318 (Resolved): hammer: radosgw-admin: inconsistency in uid/email handling
Nathan Cutler
09:08 AM Bug #17987: incorrect entry is removed when a part is upload more than once during multipart-upload
https://github.com/ceph/ceph/pull/12152 wangsu-os wangsu-os
08:16 AM Backport #17969 (In Progress): jewel: multisite upgrade from hammer -> jewel ignores rgw_region_r...
Loïc Dachary
07:35 AM Backport #17961 (In Progress): rgw, jewel: TempURL fails if rgw_keystone_implicit_tenants is enabled
h3. initial description
This is because in Jewel in the TempURL implementation we aren't trying to _rgw_get_user_i...
Loïc Dachary
05:42 AM Bug #13207 (Resolved): CVE-2016-7031 rgw: Anonymous user is able to read bucket with authenticate...
Nathan Cutler
05:42 AM Backport #17150 (Resolved): hammer: rgw: Anonymous user is able to read bucket with authenticated...
Nathan Cutler
05:34 AM Bug #17635 (Resolved): rgw crash when client post object with null conditions
Nathan Cutler
05:33 AM Backport #17631 (Resolved): hammer: Fix rgw crash when client post object with null condition
Nathan Cutler
01:43 AM Bug #18003 (Resolved): slave zonegroup cannot enable the bucket versioning
The master zone in slave zonegroup cannot enable the bucket versioning, created in slave zonegroup, and it will retur... wenjun jing

11/22/2016

09:56 PM Bug #17788: rgw: failing to authenticate aws4 with boto and with port specified
Can I reproduce this w/boto? Could someone paste a quick reproducer?
Thanks!
Matt Benjamin
06:28 PM Bug #17996 (Resolved): rgw: ECANCELED in rgw_get_system_obj() leads to infinite loop
Yehuda Sadeh
04:52 PM Bug #17995 (Resolved): rgw: zonegroup add cmd didn't update zone's realm_id
before zone add to a zonegroup, it's realm_id may be empty or an old realm_id, so when it join another zonegroup, it ... Tianshan Qu
04:12 PM Bug #17976: multisite upgrade from hammer->jewel still doesn't handle different rgw_region_root_pool
simple workaround:
add rgw_zonegroup_root_pool, rgw_realm_root_pool and rgw_period_root_pool entries in the config w...
Orit Wasserman
02:59 AM Bug #17987 (New): incorrect entry is removed when a part is upload more than once during multipar...
Here is the discussion
2016-11-19 2:25 GMT+08:00 Yehuda Sadeh-Weinraub <ysadehwe@redhat.com>:
I understand. Yeah,...
wangsu-os wangsu-os

11/21/2016

11:39 PM Bug #17985 (Fix Under Review): rgw sends omap_getvals with (u64)-1 limit
-https://github.com/ceph/ceph/pull/12115- (closed) Sage Weil
11:38 PM Bug #17985 (Resolved): rgw sends omap_getvals with (u64)-1 limit
This makes the OSD fall over in jewel (and presumably hammer). In master we impose a limit on the OSD. Sage Weil
06:55 PM Bug #17935 (Fix Under Review): rgw: wrong error code is returned when putting container metadata ...
https://github.com/ceph/ceph/pull/12109 Radoslaw Zarzynski
06:07 PM Bug #17935 (In Progress): rgw: wrong error code is returned when putting container metadata with ...
Radoslaw Zarzynski
04:31 PM Bug #17822 (Fix Under Review): multisite requests failing with '400 Bad Request' with civetweb 1.8
https://github.com/ceph/ceph/pull/12105 Casey Bodley
04:16 PM Bug #17822: multisite requests failing with '400 Bad Request' with civetweb 1.8
Thanks Yehuda. With 'tcpdump -A -r http.multi.tcpdump', I'm not seeing two different '200 OK' replies - I'm just seei... Casey Bodley
03:16 PM Bug #17938 (Fix Under Review): rgw: wrong error message is returned when putting container with a...
https://github.com/ceph/ceph/pull/12089 Radoslaw Zarzynski
01:50 PM Bug #17963: multisite upgrade from hammer -> jewel ignores rgw_region_root_pool
If the rgw_region_root_pool and rgw_zone_root_pool are the same, we have a simple workaround:
The user need to add r...
Orit Wasserman
12:58 PM Bug #17976 (Won't Fix): multisite upgrade from hammer->jewel still doesn't handle different rgw_r...
Orit Wasserman
11:46 AM Bug #17975 (Fix Under Review): rgw:can get bucket's location in slave zonegroup,not returned 301
get bucket's location in different zonegroup ,the original codes will return 301 ,restricting
the op_get type . we s...
洁 李
07:44 AM Bug #17971 (Closed): bucket acl all user write
When I set the bucket all user acl permission write.
Then use the postman to send a put request to the bucket, to ov...
Jiang Yu
01:05 AM Bug #17215: Unable to suspend a user that is part of a tenant
https://github.com/ceph/ceph/pull/12081 wenjun jing

11/20/2016

09:32 PM Backport #17969 (Resolved): jewel: multisite upgrade from hammer -> jewel ignores rgw_region_root...
https://github.com/ceph/ceph/pull/12156 Nathan Cutler

11/19/2016

09:36 PM Bug #17963 (Pending Backport): multisite upgrade from hammer -> jewel ignores rgw_region_root_pool
Orit Wasserman
07:57 PM Backport #17961 (Fix Under Review): rgw, jewel: TempURL fails if rgw_keystone_implicit_tenants is...
https://github.com/ceph/ceph/pull/12079 Radoslaw Zarzynski
02:53 AM Bug #16408: Creating Bucket failed and can't be deleted.
I haven't seen it with newer buckets, just old buckets.
I do confirm in this case, that your fix worked: @radosgw-...
Robin Johnson

11/18/2016

11:08 PM Bug #17822: multisite requests failing with '400 Bad Request' with civetweb 1.8
just note that a single 200 response is also valid. Not sure why it would return two 200s, and I can't reproduce it l... Yehuda Sadeh
10:45 PM Bug #17822: multisite requests failing with '400 Bad Request' with civetweb 1.8
And now I also notice this:... Yehuda Sadeh
10:42 PM Bug #17822: multisite requests failing with '400 Bad Request' with civetweb 1.8
What's also interesting is that it's a GET request with Transfer-Encoding header. Yehuda Sadeh
10:39 PM Bug #17822: multisite requests failing with '400 Bad Request' with civetweb 1.8
Seems that it's a chunked request without any data markers (need to specify a zero to end content). Yehuda Sadeh
10:01 PM Bug #17963: multisite upgrade from hammer -> jewel ignores rgw_region_root_pool
Nathan Cutler wrote:
> Does the backport field need "hammer,jewel" or just "jewel"?
Just jewel, thanks.
Casey Bodley
09:51 PM Bug #17963: multisite upgrade from hammer -> jewel ignores rgw_region_root_pool
Does the backport field need "hammer,jewel" or just "jewel"? Nathan Cutler
09:11 PM Bug #17963 (Fix Under Review): multisite upgrade from hammer -> jewel ignores rgw_region_root_pool
https://github.com/ceph/ceph/pull/12076 Casey Bodley
09:05 PM Bug #17963 (Resolved): multisite upgrade from hammer -> jewel ignores rgw_region_root_pool
If preexisting multisite configuration is in non-default pools (via config variables rgw_region_root_pool and rgw_zon... Casey Bodley
09:59 PM Bug #17965 (Resolved): rgw: lifecycle worker thread does not set lock duration
lifecycle lock does not expire which requires manual intervention in case of process dying while processing entries. Yehuda Sadeh
09:43 PM Bug #17964 (Resolved): rgw: multipart parts on versioned bucket create versioned bucket index ent...
No need to create versioned entried on bucket index for multipart part objects. Need to treat these objects as unvers... Yehuda Sadeh
09:06 PM Bug #17962: RGWRados::replace_region_with_zonegroup() calls unsafe conf->set_val("rgw_zonegroup")
actual pr https://github.com/ceph/ceph/pull/12075 Casey Bodley
08:55 PM Bug #17962 (Resolved): RGWRados::replace_region_with_zonegroup() calls unsafe conf->set_val("rgw_...
This modification is not thread-safe.
https://github.com/ceph/ceph/pull/12051
Casey Bodley
08:18 PM Tasks #16784: RADOSGW feature requires FCGI which no longer has an upstream source
Confirmed, the latest fcgi upstream URLs are completely gone.
http://fastcgi.com/dist/fcgi-2.4.0.tar.gz is that l...
Ken Dreyer
06:14 PM Backport #17961 (Resolved): rgw, jewel: TempURL fails if rgw_keystone_implicit_tenants is enabled
https://github.com/ceph/ceph/pull/12079 Radoslaw Zarzynski

11/17/2016

10:02 PM Bug #17822: multisite requests failing with '400 Bad Request' with civetweb 1.8
Attaching a raw tcpdump that captures the configuration (realm pull, period get, and period commit) and startup of tw... Casey Bodley
06:22 PM Feature #17931 (Fix Under Review): rgw: Swift's "prefix" parameter is not supported on account li...
PR: https://github.com/ceph/ceph/pull/12047. Radoslaw Zarzynski
04:06 PM Feature #17931 (In Progress): rgw: Swift's "prefix" parameter is not supported on account listing
Radoslaw Zarzynski
01:09 PM Feature #17931 (Resolved): rgw: Swift's "prefix" parameter is not supported on account listing
RadosGW doesn't support the _prefix_ parameter on account listing through the Swift API. Output from ngrep:... Radoslaw Zarzynski
03:05 PM Bug #17930: failure to run any task in rgw suite on vps
from http://qa-proxy.ceph.com/teuthology/owasserm-2016-11-17_09:14:20-rgw-wip-orit-testing---basic-vps/555640/teuthol... Casey Bodley
11:33 AM Bug #17930 (Rejected): failure to run any task in rgw suite on vps
http://pulpito.ceph.com/owasserm-2016-11-17_09:14:20-rgw-wip-orit-testing---basic-vps/
HTTPConnectionPool(host='vp...
Orit Wasserman
02:42 PM Backport #17875 (In Progress): jewel: rgw file: remove spurious mount entries for RGW buckets
Abhishek Varshney
02:39 PM Backport #17886 (In Progress): jewel: multisite: ECANCELED & 500 error on bucket delete
Abhishek Varshney
02:04 PM Bug #17938 (Resolved): rgw: wrong error message is returned when putting container with a name th...
Recently Tempest got a few negative tests for verifying interactions with misbehaving clients. RadosGW wrongly respon... Radoslaw Zarzynski
01:54 PM Bug #17936 (Resolved): rgw: /info lacks swift.max_meta_value_length
Recently Tempest got a few negative tests which depend on _/info_. Unfortunately, the RadosGW's implementation doesn'... Radoslaw Zarzynski
01:49 PM Bug #17935 (Resolved): rgw: wrong error code is returned when putting container metadata with too...
Recently Tempest got couple of negative tests for verifying interactions with misbehaving clients. RadosGW wrongly re... Radoslaw Zarzynski
01:31 PM Bug #17934 (Resolved): rgw: /info lacks swift.max_meta_count
Recently Swift got couple of negative tests which depends _/info_. Unfortunately, the RadosGW's implementation doesn'... Radoslaw Zarzynski
01:23 PM Bug #17932: rgw: missing support for per storage-policy usage statistics
Dump of */info*:... Radoslaw Zarzynski
01:16 PM Bug #17932 (Resolved): rgw: missing support for per storage-policy usage statistics
Following headers of the Swift API aren't sent nor even calculated for GET/HEAD responses on account:... Radoslaw Zarzynski
08:12 AM Backport #17686 (Resolved): jewel: don't loop forever when reading data from 0 sized segment
Nathan Cutler
02:05 AM Bug #16463: AWS4 Presigned URL not accepted by RGW
I got same results with Nginx as load balancer and TLS termination for Ceph RGWs v10.2.3. Looking forward for the bug... Charles Alva

11/16/2016

11:44 PM Bug #17387 (Resolved): multisite: 'radosgw-admin period prepare' is obsolete
Nathan Cutler
11:19 PM Feature #17925 (Resolved): rgw: add radosgw-admin command to check progress toward bucket shardin...
To assist users with bucket sharding decisions, add a command to list current bucket shard-fill percentage, and warn ... Matt Benjamin
05:52 PM Bug #17924: valgrind failures due to leaks of std::string memory in md_config_t
https://github.com/ceph/ceph/pull/12026 Casey Bodley
05:47 PM Bug #17924 (New): valgrind failures due to leaks of std::string memory in md_config_t
Example run: http://qa-proxy.ceph.com/teuthology/teuthology-2016-11-06_11:05:02-rgw-master-distro-basic-smithi/527300... Casey Bodley
12:52 PM Bug #17920 (Closed): upgrade from hammer only convert first zone
Orit Wasserman
12:28 PM Bug #17920 (Closed): upgrade from hammer only convert first zone
Orit Wasserman

11/15/2016

09:57 PM Bug #16408: Creating Bucket failed and can't be deleted.
Yeah, bucket entrypoint is broken. Does it happen also with newer buckets? This one seems a bit old. We can try ident... Yehuda Sadeh
05:27 AM Bug #16408: Creating Bucket failed and can't be deleted.
... Robin Johnson
01:12 AM Bug #16408: Creating Bucket failed and can't be deleted.
This seems broken:... Yehuda Sadeh
12:05 AM Bug #16408: Creating Bucket failed and can't be deleted.
Attached is the log from:
@
b=bucket
sudo radosgw-admin --bucket $b bucket rm --debug-rgw=20 --debug-ms 20 2>&1 ...
Robin Johnson
09:48 PM Backport #17908 (In Progress): jewel: rgw: the response element "X-Timestamp" of swift stat bucke...
h3. original description
when we get the stat of a bucket with swift stat bucket api, the response element "X-Time...
Nathan Cutler
09:31 AM Backport #17908: jewel: rgw: the response element "X-Timestamp" of swift stat bucket api is zero
https://github.com/ceph/ceph/pull/11990 Alexey Sheplyakov
09:23 AM Backport #17908 (Resolved): jewel: rgw: the response element "X-Timestamp" of swift stat bucket a...
https://github.com/ceph/ceph/pull/11990 Alexey Sheplyakov
08:28 AM Bug #17907 (New): multisite: unlock failure
when set rgw_num_rados_handles>1 and use rgw_num_async_rados_threads default 32 will cause unlock failure.
since m...
Tianshan Qu
12:11 AM Backport #15652: hammer: check-head-obj-locator fix failing & bucket non-deletable
I don't have any buckets immediately that still exhibit this problem in Jewel, sorry :-(. Robin Johnson

11/14/2016

07:28 PM Bug #17794: realm pull failure in teuthology with Unknown error 2202
So, despite the 'command crashed with signal 11' and valgrind output, I don't think we're actually segfaulting here. ... Casey Bodley
06:34 PM Bug #17657: Ceph / Civetweb doesn't accept Absolute URI
This issue is critical to us at this point. Will this be backported to Jewel in the upcoming 10.2.4? Vaibhav Bhembre
06:16 PM Bug #17897 (Resolved): rgw: gc records empty entries
When removing objects that only have heads we don't need to record these objects in the gc as they don't have tails. Yehuda Sadeh
01:45 PM Bug #16408: Creating Bucket failed and can't be deleted.
Can you add a log at least for failed delete? Orit Wasserman
01:44 PM Bug #17364 (Fix Under Review): rgw: Unable to commit period zonegroup change
https://github.com/ceph/ceph/pull/11965 Orit Wasserman
02:38 AM Bug #17888 (Resolved): rgw:fix for deleting objects name beginning and ending with underscores of...
those objects name beginning and ending with underscores can not be deleted using POST method of js sdk
body of ht...
ruisong zhou

11/13/2016

12:26 PM Backport #17886 (Resolved): jewel: multisite: ECANCELED & 500 error on bucket delete
https://github.com/ceph/ceph/pull/12044 Nathan Cutler
10:28 AM Backport #15652 (Need More Info): hammer: check-head-obj-locator fix failing & bucket non-deletable
Robin, is there a master PR we can cherry-pick, or is this a hammer-only bugfix? Nathan Cutler
10:25 AM Backport #16318 (In Progress): hammer: radosgw-admin: inconsistency in uid/email handling
Nathan Cutler
10:06 AM Backport #16594 (In Progress): hammer: RGW Swift API: ranged request on a DLO provides wrong valu...
Nathan Cutler
10:04 AM Backport #17123 (In Progress): hammer: COPY broke multipart files uploaded under dumpling
Nathan Cutler
10:02 AM Backport #17677 (In Progress): hammer: swift: Problems with DLO containing 0 length segments
@Orit Thanks for the backport! Nathan Cutler
06:21 AM Bug #17882 (Resolved): s3tests-test-readwrite failed in rados run (Connection refused)
... Kefu Chai

11/12/2016

10:45 AM Backport #17336 (Resolved): hammer: radosgw-admin(8) does not describe "--job-id" or "--max-concu...
Nathan Cutler
10:44 AM Documentation #17281 (Resolved): radosgw-admin(8) does not describe "--job-id" or "--max-concurre...
Nathan Cutler
10:40 AM Documentation #17280 (Resolved): radosgw-admin lacks docs for "--orphan-stale-secs"
Nathan Cutler
10:38 AM Backport #17338 (Resolved): hammer: radosgw-admin lacks docs for "--orphan-stale-secs"
Nathan Cutler
09:57 AM Backport #17875 (Resolved): jewel: rgw file: remove spurious mount entries for RGW buckets
https://github.com/ceph/ceph/pull/12045 Nathan Cutler

11/11/2016

08:35 PM Bug #17872: all swift tests fail with '[Errno 111] Connection refused' because radosgw instance i...
Later in that teuthology.log:... Casey Bodley
08:24 PM Bug #17872 (Resolved): all swift tests fail with '[Errno 111] Connection refused' because radosgw...
From http://qa-proxy.ceph.com/teuthology/cbodley-2016-11-11_12:15:00-rgw:verify-wip-rgw-compression-type-random---bas... Casey Bodley
08:33 PM Bug #17794: realm pull failure in teuthology with Unknown error 2202
That fix to retry the 'realm pull' does not appear to solve the issue. It reproduced again [1] and all 10 tries faile... Casey Bodley
04:28 PM Bug #17794: realm pull failure in teuthology with Unknown error 2202
I actually saw this once in my own local testing when I started the gateways under valgrind - valgrind apparently ret... Casey Bodley
05:28 PM Bug #16854: the usage log record date is 16 hours later than the real operate time in jewel1 10....
Revised status:
The obvious reason the highest epoch value is not stored is that currently the epoch value is rounde...
Matt Benjamin
03:00 AM Bug #16854: the usage log record date is 16 hours later than the real operate time in jewel1 10....
Status update:
I'm still confirming invariants in the code. 1) By instrumenting UsageLogger::add_user, I caused a lo...
Matt Benjamin
03:50 PM Backport #17507 (Resolved): jewel: multisite: 'radosgw-admin period prepare' is obsolete
https://github.com/ceph/ceph/pull/11574 got merged. Gaurav Garg

11/10/2016

06:59 PM Bug #17788 (Fix Under Review): rgw: failing to authenticate aws4 with boto and with port specified
Orit Wasserman
06:57 PM Bug #17818 (Fix Under Review): The user interface created by S3 does not return a Content-Type
Orit Wasserman
06:56 PM Bug #17818: The user interface created by S3 does not return a Content-Type
new pr https://github.com/ceph/ceph/pull/11828 Casey Bodley
06:56 PM Backport #17840 (In Progress): hammer: rgw: the value of total_time is wrong in the result of 'ra...
Nathan Cutler
06:55 PM Bug #16888 (Fix Under Review): [multisite] operating bucket's acl&cors is not restricted on slave...
-https://github.com/ceph/ceph/pull/11891- closed Orit Wasserman
06:55 PM Bug #17850 (Pending Backport): rgw file: remove spurious mount entries for RGW buckets
Matt Benjamin
06:52 PM Bug #17465 (In Progress): multisite: coroutine deadlock in RGWMetaSyncCR after ECANCELED errors
Casey Bodley
06:51 PM Bug #17239 (In Progress): radosgw-admin period update reverts deleted zonegroup
Orit Wasserman
06:46 PM Bug #17698 (Pending Backport): multisite: ECANCELED & 500 error on bucket delete
Casey Bodley
06:36 PM Bug #17803 (Resolved): rgw: usage statistics are broken for containers archiving versioned object...
Orit Wasserman
11:14 AM Bug #17111 (Resolved): ERROR: got unexpected error when trying to read object: -2
Loïc Dachary

11/09/2016

11:20 PM Bug #16408: Creating Bucket failed and can't be deleted.
@owasserm:
Do you have progress/insights on this? I'm seeing it on a lot of Dreamhost buckets, but it looks like it'...
Robin Johnson
09:44 PM Bug #17850 (Fix Under Review): rgw file: remove spurious mount entries for RGW buckets
Ken Dreyer
07:31 PM Bug #17850: rgw file: remove spurious mount entries for RGW buckets
Fixed in https://github.com/ceph/ceph/pull/11881 Matt Benjamin
07:24 PM Bug #17850 (Resolved): rgw file: remove spurious mount entries for RGW buckets
The Linux kernel NFS client creates a new mount entry whenever a new fsid is seen in a subordinate directory--even on... Matt Benjamin
03:24 PM Backport #17839 (In Progress): jewel: rgw: the value of total_time is wrong in the result of 'rad...
Loïc Dachary
02:47 PM Backport #17839 (Resolved): jewel: rgw: the value of total_time is wrong in the result of 'radosg...
https://github.com/ceph/ceph/pull/11876 Loïc Dachary
03:08 PM Backport #17838 (In Progress): jewel: leak in RGWFetchAllMetaCR
Loïc Dachary
02:39 PM Backport #17838 (Resolved): jewel: leak in RGWFetchAllMetaCR
https://github.com/ceph/ceph/pull/11872 Loïc Dachary
03:02 PM Backport #17313: jewel: rgw-ldap: add ldap lib to rgw lib deps based on build config
b3b3185008a0a2149dcba59813e0f0400d2e47de conflicts as of 0c38c464fff2280a6345b470f1c83aa6229776cc Loïc Dachary
03:00 PM Backport #17208: jewel: rgw: setting rgw_swift_url_prefix = "/" doesn't work as expected
dbf5b5b73dc5ceb20959a4a55dd695cd5f95d034 conflicts as of 0c38c464fff2280a6345b470f1c83aa6229776cc Loïc Dachary
02:52 PM Backport #17846 (In Progress): jewel: RGWHTTPManager deadlock when HAVE_CURL_MULTI_WAIT=0
Loïc Dachary
02:51 PM Backport #17846 (Resolved): jewel: RGWHTTPManager deadlock when HAVE_CURL_MULTI_WAIT=0
https://github.com/ceph/ceph/pull/11627 Loïc Dachary
02:47 PM Backport #17840 (Resolved): hammer: rgw: the value of total_time is wrong in the result of 'rados...
https://github.com/ceph/ceph/pull/11899 Loïc Dachary
02:38 PM Bug #16368 (Pending Backport): RGWHTTPManager deadlock when HAVE_CURL_MULTI_WAIT=0
Loïc Dachary
02:37 PM Backport #17343 (In Progress): jewel: radosgw Consumes too much CPU time to synchronize metadata ...
Loïc Dachary
02:37 PM Backport #17512 (In Progress): jewel: multisite: metadata master can get the wrong value for 'old...
Loïc Dachary
02:30 PM Backport #17709 (In Progress): jewel: multisite: coroutine deadlock assertion on error in FetchAl...
Loïc Dachary
02:29 PM Backport #17710 (In Progress): jewel: multisite: race between ReadSyncStatus and InitSyncStatus l...
Loïc Dachary
02:26 PM Backport #17708 (In Progress): jewel: 'radosgw-admin bucket sync init' crashes
Loïc Dachary
02:23 PM Backport #17733 (In Progress): jewel: multisite: after finishing full sync on a bucket, increment...
Loïc Dachary
02:20 PM Backport #17735 (In Progress): jewel: RGW will not list Argonaut-era bucket via HTTP (but radosgw...
Loïc Dachary
02:18 PM Backport #17756: jewel: rgw: bucket resharding
conflicts at b01e732f3d597670f4f781be3db81786e63d5053 from https://github.com/ceph/ceph/pull/11230/commits Loïc Dachary
11:42 AM Backport #17786: jewel: multisite: assertion in RGWRados::wakeup_data_sync_shards
unassigning so that someone else can grab it Loïc Dachary

11/08/2016

10:46 PM Bug #17822: multisite requests failing with '400 Bad Request' with civetweb 1.8
That's weird. Need to look at the previous request and what happens there. That commit fixes a similar issue, so I'm ... Yehuda Sadeh
10:24 PM Bug #17822: multisite requests failing with '400 Bad Request' with civetweb 1.8
Managed to catch the failure in parse_http_message():... Casey Bodley
09:57 PM Bug #17822: multisite requests failing with '400 Bad Request' with civetweb 1.8
When sending an empty chunked request there still needs to be a marker that specifies end of data. Can you check to s... Yehuda Sadeh
08:08 PM Bug #17822: multisite requests failing with '400 Bad Request' with civetweb 1.8
i captured a failing request using tshark:... Casey Bodley
04:13 PM Bug #17822 (Resolved): multisite requests failing with '400 Bad Request' with civetweb 1.8
The multisite test script test/rgw/test_multi.py is getting stuck due to rejected requests. On the sending side:
<pr...
Casey Bodley
05:41 PM Bug #17812 (Pending Backport): leak in RGWFetchAllMetaCR
Casey Bodley
12:57 PM Documentation #17820: Add workaround for upgrade issues for older jewel versions
https://github.com/ceph/ceph/pull/11827 Orit Wasserman
12:57 PM Documentation #17820 (Resolved): Add workaround for upgrade issues for older jewel versions
Orit Wasserman
09:54 AM Bug #17818: The user interface created by S3 does not return a Content-Type
https://github.com/ceph/ceph/pull/11777 Jiang Yu
09:54 AM Bug #17818 (Fix Under Review): The user interface created by S3 does not return a Content-Type
the Content-Type is not returned when the s3 API user create.
Causing the client to not be able to parse the data fr...
Jiang Yu

11/07/2016

07:07 PM Bug #17793 (Fix Under Review): leak in RGWAsyncRadosProcessor::handle_request
https://github.com/ceph/ceph/pull/11815 Casey Bodley
03:31 PM Bug #17793: leak in RGWAsyncRadosProcessor::handle_request
There's a circular reference between RGWMetaStoreEntryCR and RGWAsyncMetaStoreEntry. I'll try removing the reference ... Casey Bodley
07:06 PM Bug #17812 (Fix Under Review): leak in RGWFetchAllMetaCR
https://github.com/ceph/ceph/pull/11816 Casey Bodley
05:46 PM Bug #17812 (Resolved): leak in RGWFetchAllMetaCR
Running valgrind on a local mstart cluster produces leaks in RGWFetchAllMetaCR. The allocation at rgw_sync.cc:810 is ... Casey Bodley
06:45 PM Bug #17598 (Pending Backport): rgw: the value of total_time is wrong in the result of 'radosgw-ad...
https://github.com/ceph/ceph/pull/11425 Yehuda Sadeh
10:00 AM Backport #17631 (In Progress): hammer: Fix rgw crash when client post object with null condition
Nathan Cutler

11/05/2016

12:14 PM Bug #17803 (Fix Under Review): rgw: usage statistics are broken for containers archiving versione...
https://github.com/ceph/ceph/pull/11794 Radoslaw Zarzynski
11:41 AM Bug #17803 (In Progress): rgw: usage statistics are broken for containers archiving versioned obj...
Urgent because the failed test is a part of RefStack. Radoslaw Zarzynski
04:57 AM Bug #17803 (Resolved): rgw: usage statistics are broken for containers archiving versioned object...
*cver-store* used in the listings is a container supposed to store non-current versions of objects belonging to a Swi... Radoslaw Zarzynski

11/04/2016

08:01 AM Bug #11239: civetweb defaults to libssl.so and libcrypto.so when versions not passed
I've got yet another version of the ssl patch: these PRs: https://github.com/ceph/ceph/pull/11776 https://github.com/... Marcus Watts
01:10 AM Feature #17795 (New): rgw: provide sync info about pending objects
Yehuda Sadeh

11/03/2016

04:43 PM Bug #17794 (New): realm pull failure in teuthology with Unknown error 2202
http://qa-proxy.ceph.com/teuthology/owasserm-2016-11-02_22:33:41-rgw-wip-orit-testing---basic-mira/512520/teuthology.... Orit Wasserman
10:35 AM Backport #17786 (New): jewel: multisite: assertion in RGWRados::wakeup_data_sync_shards
Looks like a non-trivial conflict due to commit : b974dac04e03eed08ece95484b57471772eacbf4 Abhishek Varshney
09:37 AM Backport #17786 (In Progress): jewel: multisite: assertion in RGWRados::wakeup_data_sync_shards
Abhishek Varshney
10:32 AM Bug #17793 (Resolved): leak in RGWAsyncRadosProcessor::handle_request
http://qa-proxy.ceph.com/teuthology/owasserm-2016-11-02_22:33:41-rgw-wip-orit-testing---basic-mira/512511/remote/mira... Orit Wasserman
10:26 AM Bug #17792: leak in RGWMetaStoreEntryCR
see http://pulpito.ceph.com/owasserm-2016-11-02_22:33:41-rgw-wip-orit-testing---basic-mira/512511 Orit Wasserman
10:25 AM Bug #17792 (Duplicate): leak in RGWMetaStoreEntryCR
<error>
<unique>0x10</unique>
<tid>1</tid>
<kind>Leak_PossiblyLost</kind>
<xwhat>
<text>8 bytes in 1...
Orit Wasserman
10:08 AM Feature #17790: rgw: support for X-Robots-Tag header
https://github.com/ceph/ceph/pull/11753 Orit Wasserman
10:07 AM Feature #17790 (Resolved): rgw: support for X-Robots-Tag header
Orit Wasserman
10:00 AM Backport #17769 (In Progress): jewel: disable virtual hosting of buckets when no hostnames are co...
Abhishek Varshney
09:56 AM Backport #17783 (In Progress): jewel: rgw: json encode/decode of RGWBucketInfo missing index_type...
Abhishek Varshney
09:54 AM Backport #17784 (In Progress): jewel: osd crashes when "radosgw-admin bi list --max-entries=1" co...
Abhishek Varshney
09:47 AM Backport #17785 (In Progress): jewel: ldap: unhandled exception from rgw::from_base64() in RGW_Au...
Abhishek Varshney

11/02/2016

10:46 PM Feature #17790 (Resolved): rgw: support for X-Robots-Tag header
Yehuda Sadeh
06:52 PM Bug #17788: rgw: failing to authenticate aws4 with boto and with port specified
https://github.com/ceph/ceph/pull/11751 Yehuda Sadeh
06:36 PM Bug #17788 (Resolved): rgw: failing to authenticate aws4 with boto and with port specified
in current master Yehuda Sadeh
03:09 PM Backport #17786 (Resolved): jewel: multisite: assertion in RGWRados::wakeup_data_sync_shards
https://github.com/ceph/ceph/pull/13886 Loïc Dachary
03:09 PM Backport #17785 (Resolved): jewel: ldap: unhandled exception from rgw::from_base64() in RGW_Auth_...
https://github.com/ceph/ceph/pull/11672 Loïc Dachary
03:08 PM Backport #17784 (Resolved): jewel: osd crashes when "radosgw-admin bi list --max-entries=1" comma...
https://github.com/ceph/ceph/pull/11758 Loïc Dachary
03:08 PM Backport #17783 (Resolved): jewel: rgw: json encode/decode of RGWBucketInfo missing index_type field
https://github.com/ceph/ceph/pull/11759 Loïc Dachary
01:51 PM Bug #17779: rgw: s3 API does not honor rgw_keystone_implicit_tenants when keystone integration is...
http://docs.ceph.com/docs/master/radosgw/multitenancy/
Well, I read the doc again, and it says "When a client appl...
Yiu Chung Lee
09:51 AM Bug #17779: rgw: s3 API does not honor rgw_keystone_implicit_tenants when keystone integration is...
Just realized that S3 bucket namespace is supposed to be global. I think this ticket can be closed.... Yiu Chung Lee
09:36 AM Bug #17779: rgw: s3 API does not honor rgw_keystone_implicit_tenants when keystone integration is...
Note that you need to create the EC2 credentials in openstack (openstack ec2 credentials create) to replicate this be... Yiu Chung Lee
09:31 AM Bug #17779: rgw: s3 API does not honor rgw_keystone_implicit_tenants when keystone integration is...
The problem here seems to be the user auto-created by S3 API seems does not honour rgw_keystone_implicit_tenants ceph... Yiu Chung Lee
09:24 AM Bug #17779: rgw: s3 API does not honor rgw_keystone_implicit_tenants when keystone integration is...
attached the code I used to create buckets using S3 API. For Swift API the standard openstack swift command is used (... Yiu Chung Lee
09:22 AM Bug #17779 (Resolved): rgw: s3 API does not honor rgw_keystone_implicit_tenants when keystone int...
When I tried to access rgw configured with keystone integration using S3 API with a new user , It appears that the ne... Yiu Chung Lee
08:28 AM Bug #17777: Document that erasure pools cannot be used for RGW
For the reason stated in http://tracker.ceph.com/issues/17736#note-2 Nathan Cutler
03:07 AM Bug #17777 (New): Document that erasure pools cannot be used for RGW
Ref bug http://tracker.ceph.com/issues/17736
In the object gateway examples, it is sometimes suggested that the po...
Richard Chan
08:07 AM Bug #11239: civetweb defaults to libssl.so and libcrypto.so when versions not passed
I've updated https://github.com/ceph/ceph/pull/11571
It now needs to be applied after https://github.com/ceph/civetw...
Marcus Watts
03:09 AM Bug #17736: Pure EC pools, RGW cannot create system user
Done! - thank you for this information.
http://tracker.ceph.com/issues/17777
Richard Chan

11/01/2016

11:36 PM rgw-testing Bug #17776 (New): rgw: test aws4
Yehuda Sadeh
08:57 PM Bug #17745 (Pending Backport): osd crashes when "radosgw-admin bi list --max-entries=1" command r...
Yehuda Sadeh
01:54 AM Bug #17745: osd crashes when "radosgw-admin bi list --max-entries=1" command runing
PR: https://github.com/ceph/ceph/pull/11697 wei qiaomiao
06:20 PM Backport #17342: jewel: teuthology: assertion failure in a radosgw-admin related task
https://github.com/ceph/ceph/pull/11132 should be backported Orit Wasserman
06:19 PM Bug #17167: teuthology: assertion failure in a radosgw-admin related task
https://github.com/ceph/ceph/pull/11132 Orit Wasserman
06:16 PM Bug #17569 (Pending Backport): multisite: assertion in RGWRados::wakeup_data_sync_shards
Yehuda Sadeh
06:05 PM Bug #17663 (Pending Backport): ldap: unhandled exception from rgw::from_base64() in RGW_Auth_S3::...
Matt Benjamin
06:04 PM Bug #17663: ldap: unhandled exception from rgw::from_base64() in RGW_Auth_S3::authorize_v2()
PR https://github.com/ceph/ceph/pull/10777 fixes the issue on master.
Matt Benjamin
05:59 PM Bug #17736 (Won't Fix): Pure EC pools, RGW cannot create system user
rgw cannot use ec pool for metadata as it lacks some of the functionality that it requires. Please open a doc bug if ... Yehuda Sadeh
05:54 PM Bug #17755 (Pending Backport): rgw: json encode/decode of RGWBucketInfo missing index_type field
Yehuda Sadeh
05:53 PM rgw-testing Bug #17771 (New): rgw: missing a test for enabling/disabling indexless buckets
Yehuda Sadeh
02:44 PM Backport #17769 (Resolved): jewel: disable virtual hosting of buckets when no hostnames are confi...
https://github.com/ceph/ceph/pull/11760 Loïc Dachary
09:15 AM Bug #17440 (Pending Backport): disable virtual hosting of buckets when no hostnames are configured
Orit Wasserman

10/31/2016

06:38 PM Backport #17756 (Resolved): jewel: rgw: bucket resharding
https://github.com/ceph/ceph/pull/13341 Loïc Dachary
06:21 PM Feature #17550 (Pending Backport): rgw: bucket resharding
From an IRC conversation today, Yehuda confirmed we should backport this to jewel. Ken Dreyer
05:59 PM Bug #17755 (Fix Under Review): rgw: json encode/decode of RGWBucketInfo missing index_type field
PR for master: https://github.com/ceph/ceph/pull/11707 Ken Dreyer
05:38 PM Bug #17755 (Resolved): rgw: json encode/decode of RGWBucketInfo missing index_type field
We can't modify index type on existing buckets. Yehuda Sadeh
06:22 AM Bug #17745 (Resolved): osd crashes when "radosgw-admin bi list --max-entries=1" command runing
osd log:
2016-10-29 15:00:39.220298 7f696f8d4800 -1 osd.0 86 log_to_monitors {default=true}
2016-10-29 15:00:39.2...
wei qiaomiao

10/28/2016

02:17 PM Bug #17736: Pure EC pools, RGW cannot create system user
1. If us-east.rgw.log pool is replicated then system user can be created
2. us-east.rgw.buckets.index us-east.rgw....
Richard Chan
01:40 PM Bug #17736 (Won't Fix): Pure EC pools, RGW cannot create system user
I create all EC pools to begin a multisite configuration in Jewel 10.2.3 on CentOS 7
[default]
rgw_realm_root_poo...
Richard Chan
12:53 PM Backport #17735 (Resolved): jewel: RGW will not list Argonaut-era bucket via HTTP (but radosgw-ad...
https://github.com/ceph/ceph/pull/11863 Loïc Dachary
12:51 PM Backport #17733 (Resolved): jewel: multisite: after finishing full sync on a bucket, incremental ...
https://github.com/ceph/ceph/pull/11864 Loïc Dachary
10:41 AM Backport #17674 (In Progress): jewel: set_acl fails for object beginning and ending in underscore
Loïc Dachary
09:47 AM Backport #17674: jewel: set_acl fails for object beginning and ending in underscore
https://github.com/ceph/ceph/pull/11675 Alexey Sheplyakov

10/27/2016

08:58 PM Backport #17673 (Resolved): jewel: rgw crash when client post object with null conditions
Loïc Dachary
07:48 PM Backport #17510 (Resolved): jewel: ERROR: got unexpected error when trying to read object: -2
Loïc Dachary
06:20 PM Backport #17676 (Resolved): jewel: swift: Problems with DLO containing 0 length segments
Orit Wasserman
06:03 PM Bug #17372 (Pending Backport): RGW will not list Argonaut-era bucket via HTTP (but radosgw-admin ...
Yehuda Sadeh
06:01 PM Bug #17633: Source Package radosgw file has authentication issues
did you follow installation instructions? it looks like misconfiguration. Yehuda Sadeh
05:59 PM Bug #17624 (Pending Backport): multisite: after finishing full sync on a bucket, incremental sync...
Yehuda Sadeh
03:27 PM Bug #17635: rgw crash when client post object with null conditions
CVE-2016-8626 tracks this issue. Ken Dreyer
07:49 AM Bug #17692 (Resolved): rgw: don't loop forever when reading data from 0 sized segment
Loïc Dachary
07:48 AM Bug #17256 (Resolved): objects in the metadata_heap pool are created, but never read or removed
Loïc Dachary

10/26/2016

08:52 PM Bug #17635: rgw crash when client post object with null conditions
The Red Hat Security team is going to assign a CVE to this issue. Ken Dreyer
06:07 PM Backport #17673 (In Progress): jewel: rgw crash when client post object with null conditions
Pushed:
https://github.com/ceph/ceph/pull/11662
Matt Benjamin
03:59 PM Bug #17574: multisite: many duplicate mdlog entries cause race to sync and result in ECANCELED
This issue reproduced in job 494193. The osd log [1] indicates that the majority of ECANCELED errors are coming from ... Casey Bodley
02:31 PM Backport #17343: jewel: radosgw Consumes too much CPU time to synchronize metadata or data betwee...
backport pr reopened with an extra commit requested by Ken Dreyer Casey Bodley
12:06 PM Backport #17339 (Resolved): jewel: objects in the metadata_heap pool are created, but never read ...
Loïc Dachary
09:57 AM Backport #17693 (Resolved): jewel: rgw: don't loop forever when reading data from 0 sized segment
Loïc Dachary
09:41 AM Backport #17312 (Resolved): jewel: build/ops: allow building RGW with LDAP disabled
Loïc Dachary
07:08 AM Backport #17710 (Resolved): jewel: multisite: race between ReadSyncStatus and InitSyncStatus lead...
https://github.com/ceph/ceph/pull/11865 Loïc Dachary
07:08 AM Backport #17709 (Resolved): jewel: multisite: coroutine deadlock assertion on error in FetchAllMe...
https://github.com/ceph/ceph/pull/11866 Loïc Dachary
07:08 AM Backport #17708 (Resolved): jewel: 'radosgw-admin bucket sync init' crashes
https://github.com/ceph/ceph/pull/11864 Loïc Dachary

10/25/2016

10:07 PM Bug #17661 (Pending Backport): 'radosgw-admin bucket sync init' crashes
Yehuda Sadeh
10:05 PM Bug #17571 (Pending Backport): multisite: coroutine deadlock assertion on error in FetchAllMetaCR
Yehuda Sadeh
10:04 PM Bug #17568 (Pending Backport): multisite: race between ReadSyncStatus and InitSyncStatus leads to...
Yehuda Sadeh
05:04 PM Bug #17698: multisite: ECANCELED & 500 error on bucket delete
proposed fix at https://github.com/ceph/ceph/pull/11648 Casey Bodley
04:59 PM Bug #17698: multisite: ECANCELED & 500 error on bucket delete
Sounds like secondary is racing with the request forwarded from the master (which the secondary itself initiated). We... Yehuda Sadeh
04:35 PM Bug #17698: multisite: ECANCELED & 500 error on bucket delete
Adding the rgw log of the relevant bucket, it seems we're issuing delete on the bucket object twice, once at
2016-1...
Abhishek Lekshmanan
04:28 PM Bug #17698 (Resolved): multisite: ECANCELED & 500 error on bucket delete
sometimes on bucket delete at the secondary, we're seeing secondary returning 500 on a bucket delete (caused by rerai... Abhishek Lekshmanan
03:38 PM Bug #17574: multisite: many duplicate mdlog entries cause race to sync and result in ECANCELED
I pushed a branch to ceph-qa-suite [1] that sets debug_objclass=20 and scheduled a run of the rgw/verify suite [2].
...
Casey Bodley
10:31 AM Backport #17693 (In Progress): jewel: rgw: don't loop forever when reading data from 0 sized segment
Loïc Dachary
10:30 AM Backport #17693 (Resolved): jewel: rgw: don't loop forever when reading data from 0 sized segment
https://github.com/ceph/ceph/pull/11626 Loïc Dachary
10:29 AM Bug #17692 (Resolved): rgw: don't loop forever when reading data from 0 sized segment
https://github.com/ceph/ceph/pull/11567 Loïc Dachary
 

Also available in: Atom