X-Git-Url: https://gerrit.opnfv.org/gerrit/gitweb?a=blobdiff_plain;f=src%2Fceph%2Fdoc%2Fchangelog%2Fv10.2.8.txt;fp=src%2Fceph%2Fdoc%2Fchangelog%2Fv10.2.8.txt;h=0000000000000000000000000000000000000000;hb=7da45d65be36d36b880cc55c5036e96c24b53f00;hp=575a7b062e35239facb3329f13e2fb11b4c4e30b;hpb=691462d09d0987b47e112d6ee8740375df3c51b2;p=stor4nfv.git diff --git a/src/ceph/doc/changelog/v10.2.8.txt b/src/ceph/doc/changelog/v10.2.8.txt deleted file mode 100644 index 575a7b0..0000000 --- a/src/ceph/doc/changelog/v10.2.8.txt +++ /dev/null @@ -1,5435 +0,0 @@ -commit f5b1f1fd7c0be0506ba73502a675de9d048b744e -Author: Jenkins Build Slave User -Date: Thu Jul 6 14:56:18 2017 +0000 - - 10.2.8 - -commit 66dbf9beef04988dbd3653591e51afa6d84e3990 -Merge: 2f491b2e5e f46ccf2cb4 -Author: Nathan Cutler -Date: Tue Jul 4 17:43:57 2017 +0200 - - Merge pull request #14710 from smithfarm/wip-start-race - - tests: rados: sleep before ceph tell osd.0 flush_pg_stats after restart - - Reviewed-by: Sage Weil - Reviewed-by: Kefu Chai - Reviewed-by: David Zafman - -commit 2f491b2e5e8b1f340b28415c3bd3d9628603c377 -Merge: 552a573f84 a372b4eca1 -Author: Yuri Weinstein -Date: Tue Jul 4 07:31:50 2017 -0700 - - Merge pull request #16089 from ceph/wip_fix_point_jewel - - qa/Fixed upgrade sequence to 10.2.0 -> 10.2.7 -> latest -x (10.2.8) - - Reviewed-by: Nathan Cutler - -commit 552a573f8426ecfec1a0df21a6c3941afd4e460c -Merge: 53a3be7261 55eeaadfc4 -Author: Yuri Weinstein -Date: Mon Jul 3 17:23:43 2017 -0700 - - Merge pull request #16088 from smithfarm/wip-fix-client-upgrade-centos - - tests: run upgrade/client-upgrade on latest CentOS 7.3 - - Reviewed-by: Yuri Weinstein - -commit a372b4eca1f25647541943918ae737f20783db11 -Author: Yuri Weinstein -Date: Mon Jul 3 14:18:14 2017 -0700 - - Fixed upgrade sequence to 10.2.0 -> 10.2.7 -> latest -x (10.2.8) - - Signed-off-by: Yuri Weinstein - -commit 55eeaadfc4025c83cb63c951265710868df0325f -Author: Nathan Cutler -Date: Mon Jul 3 22:55:21 2017 +0200 - - tests: run upgrade/client-upgrade on latest CentOS 7.3 - - Before this patch, all centos jobs were failing because there are no longer any - CentOS 7.2 machines in Sepia. - - Signed-off-by: Nathan Cutler - -commit 53a3be7261cfeb12445fbdba8238eefa40ed09f5 -Merge: 84bd162978 d33b30cdb0 -Author: Nathan Cutler -Date: Fri Jun 30 16:43:38 2017 +0200 - - Merge pull request #15504 from Vicente-Cheng/wip-20151-jewel - - jewel: ceph-disk: do not setup_statedir on trigger - - Reviewed-by: Josh Durgin - -commit 84bd162978e48eead40335bcbd92e4ab18e9c590 -Merge: c710689109 8e0e4a0ce7 -Author: Nathan Cutler -Date: Wed Jun 28 10:20:23 2017 +0200 - - Merge pull request #15904 from smithfarm/wip-20413-jewel - - jewel: tests: upgrade:hammer-x/stress-split-erasure-code-x86_64 fails in 10.2.8 integration testing - - Reviewed-by: Brad Hubbard - -commit c7106891096c895f0cc5c2cef438078ea48de95d -Merge: 5c6cb14806 9d3110c276 -Author: Nathan Cutler -Date: Wed Jun 28 08:16:33 2017 +0200 - - Merge pull request #14930 from smithfarm/wip-19829-jewel - - jewel: tests: New upgrade test for #14930 - - Reviewed-by: Josh Durgin - -commit 5c6cb1480699f5ce464e25b9cacdda770ce3660d -Merge: bdc085d02a d43e19d886 -Author: Nathan Cutler -Date: Wed Jun 28 08:16:04 2017 +0200 - - Merge pull request #14392 from asheplyakov/19508-jewel - - jewel: osd: pg_pool_t::encode(): be compatible with Hammer <= 0.94.6 - - Reviewed-by: Josh Durgin - -commit 9d3110c276917055b078cd14c181b2bda2625821 -Author: Nathan Cutler -Date: Sun Jun 25 10:32:16 2017 +0200 - - tests: upgrade/hammer-x/v0-94-6-mon-overload: tweak packages list - - Include some hammer dependencies that aren't in the jewel default packages - list, and exclude some java packages that may not be in the hammer repo and are - not needed for the upgrade test in any case. - - N.B.: This cannot be cherry-picked from master because upgrade/hammer-x was - dropped in master. - - Signed-off-by: Nathan Cutler - -commit 6a64f8901bb3b218a8dc58b11d6c13033d45f067 -Author: Nathan Cutler -Date: Wed May 3 11:39:27 2017 +0200 - - tests: upgrade/hammer-x: new v0-94-6-mon-overload subsuite - - This is not a cherry-pick from master because direct upgrades - from hammer to kraken+ are not supported. - - Fixes: http://tracker.ceph.com/issues/19829 - References: http://tracker.ceph.com/issues/19508 - Signed-off-by: Nathan Cutler - -commit bdc085d02ab9723f6b90b6a3047bc51cf224b930 -Merge: e41ae4a7b1 d2d4b7202d -Author: Nathan Cutler -Date: Tue Jun 27 12:41:59 2017 +0200 - - Merge pull request #15936 from batrick/i20412 - - qa: enable quotas for pre-luminous quota tests - - Reviewed-by: John Spray - Reviewed-by: Nathan Cutler - -commit d2d4b7202d77e5696eb18c4da4f7d614116ced36 -Author: Patrick Donnelly -Date: Mon Jun 26 19:04:48 2017 -0700 - - qa: enable quotas for pre-luminous quota tests - - This cannot be cherry-picked from master because the config option is removed - since 0f250a889dba2100d3afcea0a18e4f6a8d086b86. - - Fixes: http://tracker.ceph.com/issues/20412 - - Signed-off-by: Patrick Donnelly - -commit e41ae4a7b1b9cc4394473e21f6e6d6ef9cab1d59 -Merge: a21af3b7a1 682b4d717c -Author: Sage Weil -Date: Mon Jun 26 21:24:11 2017 -0500 - - Merge pull request #15933 from smithfarm/wip-hammer-jewel-x - - jewel: tests: drop upgrade/hammer-jewel-x - -commit 682b4d717c96b516c315a01b1174af3503dedba6 -Author: Nathan Cutler -Date: Tue Jun 27 02:27:22 2017 +0200 - - tests: drop upgrade/hammer-jewel-x - - This suite doesn't have any test logic in it. Its existence in the jewel branch - appears to be an oversight. - - This cannot be cherry-picked from master because the upgrade/hammer-jewel-x - suite is present (and justified) in master and is not currently being dropped - there. - - Signed-off-by: Nathan Cutler - -commit a21af3b7a1ea5abfa4f344800e413d6249824204 -Merge: 615a6ab9b7 a744340790 -Author: John Spray -Date: Mon Jun 26 16:25:12 2017 -0400 - - Merge pull request #15438 from Vicente-Cheng/wip-20027-jewel - - jewel: mds: issue new caps when sending reply to client - - Reviewed-by: John Spray - -commit 615a6ab9b723d204c6de567750fe9450742fcedb -Merge: 9b13b48b3e 8ac0e5c363 -Author: John Spray -Date: Mon Jun 26 16:23:48 2017 -0400 - - Merge pull request #15000 from jan--f/wip-19846-jewel - - jewel: cephfs: normalize file open flags internally used by cephfs - - Reviewed-by: John Spray - -commit 9b13b48b3ed919340789a41d065eb4a9a27110de -Merge: d217da1742 b429fa1807 -Author: Nathan Cutler -Date: Mon Jun 26 18:25:03 2017 +0200 - - Merge pull request #15383 from asheplyakov/20014-bp-jewel - - jewel: cls/rgw: list_plain_entries() stops before bi_log entries - - Reviewed-by: Orit Wasserman - -commit d217da174252f9126d530868aa472230ecba31ca -Merge: e520040ece 4028774122 -Author: Zack Cerza -Date: Mon Jun 26 10:17:03 2017 -0600 - - Merge pull request #15870 from smithfarm/wip-swift-task-move-jewel - - tests: move swift.py task from teuthology to ceph, phase one (jewel) - -commit e520040ecec756ce181f716dad1c0bad41c77a7d -Merge: dde8656e6b 1c0c9093ab -Author: Nathan Cutler -Date: Mon Jun 26 09:24:04 2017 +0200 - - Merge pull request #15842 from smithfarm/wip-sortbitwise-jewel - - qa/suites/upgrade/hammer-x: set "sortbitwise" for jewel clusters - - Reviewed-by: Kefu Chai - -commit dde8656e6b5e7ffe66a6fd695cbc17dfb18fb43e -Merge: 498c96e66a 06cf9f3650 -Author: John Spray -Date: Sun Jun 25 19:59:40 2017 -0400 - - Merge pull request #15468 from smithfarm/wip-20140-jewel - - jewel: cephfs: Journaler may execute on_safe contexts prematurely - - Reviewed-by: John Spray - -commit 4028774122954023265d7825fbf9e91dc526fdee -Author: Nathan Cutler -Date: Sun Jun 25 12:42:36 2017 +0200 - - tests: swift.py: tweak imports - - The ".." form only works within the teuthology repo. With swift.py now in the - Ceph repo, we have to be explicit. - - Error message was: "ValueError: Attempted relative import beyond toplevel - package - - Signed-off-by: Nathan Cutler - -commit 8e0e4a0ce7489542f47522e0a5161a5bf123c744 -Author: Nathan Cutler -Date: Sun Jun 25 10:27:58 2017 +0200 - - tests: upgrade/hammer-x/stress-split: tweak packages list - - Include some hammer dependencies that aren't in the jewel default packages - list, and exclude some java packages that may not be in the hammer repo and are - not needed for the upgrade test in any case. - - N.B.: This cannot be cherry-picked from master because upgrade/hammer-x was - dropped in master. - - Signed-off-by: Nathan Cutler - -commit a86ce728954a765797ce634025d43650d990e480 -Author: Nathan Cutler -Date: Sun Jun 25 09:27:47 2017 +0200 - - tests: swift.py: clone the ceph-jewel branch - - The master branch of ceph/swift.git contains tests that are incompatible with - Jewel and Hammer. The ceph-jewel branch omits these tests. - - Signed-off-by: Nathan Cutler - -commit 498c96e66a91edc8bd614cfc8fc5a14b3d210a76 -Merge: 38af498f9f cda721bbbf -Author: John Spray -Date: Fri Jun 23 08:02:48 2017 -0400 - - Merge pull request #15472 from smithfarm/wip-20148-jewel - - jewel: mds: Too many stat ops when trying to probe a large file - - Reviewed-by: John Spray - -commit 3d5b489369bb2cecccb1f36347654c0a37069d1c -Merge: 38af498f9f 7b58ac97e9 -Author: Nathan Cutler -Date: Fri Jun 23 08:35:27 2017 +0200 - - Merge branch 'master' of /home/smithfarm/src/ceph/upstream/teuthology into wip-swift-task-move-jewel - -commit 7b58ac97e9dd195f4170e9e0ea00bae76d1f3ccd -Author: Nathan Cutler -Date: Fri Jun 23 08:27:42 2017 +0200 - - tests: move swift.py task to qa/tasks - - In preparation for moving this task from ceph/teuthology.git into ceph/ceph.git - - The move is necessary because jewel-specific changes are needed, yet teuthology - does not maintain a separate branch for jewel. Also, swift.py is a - Ceph-specific task so it makes more sense to have it in Ceph. - - Signed-off-by: Nathan Cutler - -commit 38af498f9f7b62f9f851364ae7f2691832423198 -Merge: d0ae1de51f aa0cd461df -Author: Nathan Cutler -Date: Thu Jun 22 22:07:23 2017 +0200 - - Merge pull request #15529 from badone/wip-async-sleep-timer-fix-jewel - - jewel: osd: Implement asynchronous scrub sleep - - Reviewed-by: Josh Durgin - -commit 1c0c9093ab913a82c1dc5656a54b4009bdc35c9c -Author: Nathan Cutler -Date: Thu Jun 22 11:32:42 2017 +0200 - - qa/suites/upgrade/hammer-x: set "sortbitwise" for jewel clusters - - Inspired by 3734280522a913ca8340ebc98b80978f63bade6f - - This cannot be cherry-picked from master because master does not have - qa/suites/upgrade/hammer-x - - Signed-off-by: Nathan Cutler - -commit d0ae1de51f5faf26a2f4b0d5b7f494a4923f870d -Merge: 64c011a8c4 de76fdbb9f -Author: Kefu Chai -Date: Thu Jun 22 11:59:51 2017 +0800 - - Merge pull request #15824 from tchaikov/jewel - - qa/workunits/rados/test-upgrade-*: whitelist tests the right way - - Reviewed-by: Sage Weil - -commit de76fdbb9f435652e2c15326d00b01d26ab007a7 -Author: Kefu Chai -Date: Thu Jun 22 08:06:43 2017 +0800 - - qa/workunits/rados/test-upgrade-*: whitelist tests the right way - - --gtest_filter=POSTIVE_PATTERNS[-NEGATIVE_PATTERNS], so we cannot add - multiple exclusive patterns using -pattern:-pattern, instead, we should - use: -pattern:pattern - - Signed-off-by: Kefu Chai - Conflicts: - qa/workunits/rados/test-upgrade-v11.0.0.sh: this change is not - cherry-picked from master, because the clone-range op was removed - from master. and only supported in pre-luminous releases. - -commit 64c011a8c4af27dc095b1a9190ccf1ca76d2cc8f -Merge: e8da5e376f ab78cd040f -Author: Nathan Cutler -Date: Tue Jun 20 22:54:00 2017 +0200 - - Merge pull request #14661 from smithfarm/wip-19575-jewel - - jewel: rgw: unsafe access in RGWListBucket_ObjStore_SWIFT::send_response() - - Reviewed-by: Casey Bodley - -commit e8da5e376fc426f85fcab84a5ae71b3c17ed0068 -Merge: 5a1e849ecf 1af6781d3c -Author: Kefu Chai -Date: Tue Jun 20 22:42:47 2017 +0800 - - Merge pull request #15778 from tchaikov/wip-upgrade-without-clone-range-jewel - - qa/workunits/rados/test-upgrade-*: whitelist tests for master - - Reviewed-by: Sage Weil - -commit d43e19d88692bd318f0569559867df919c26d8db -Author: Alexey Sheplyakov -Date: Fri Apr 7 12:34:20 2017 +0400 - - jewel: osd: pg_pool_t::encode(): be compatible with Hammer <= 0.94.6 - - This patch is necessary for Jewel only since direct upgrades from Hammer - to Kraken and newer are not supported. - - Fixes: http://tracker.ceph.com/issues/19508 - - Signed-off-by: Alexey Sheplyakov - -commit 1af6781d3c60421930087d31124e62cae530ca24 -Author: Kefu Chai -Date: Tue Jun 20 19:49:14 2017 +0800 - - qa/workunits/rados/test-upgrade-*: whitelist tests for master - - The jewel-x upgrade test now runs this script against a mixed cluster on - a machine with code from master installed. That means we have to - skip any new tests that will fail on a mixed cluster. CloneRange was - removed in 0d7b0b7. - - Signed-off-by: Kefu Chai - Conflicts: - qa/workunits/rados/test-upgrade-v11.0.0.sh: this change is not - cherry-picked from master, because the clone-range op was removed from - master. and only supported in pre-luminous releases. - -commit a74434079088129244b7aae6ccc6df7094282eba -Author: Yan, Zheng -Date: Sat Apr 22 12:27:12 2017 +0800 - - mds: issue new caps when sending reply to client - - After Locker::issue_new_caps() adds new Capability data struct, - do not issue caps immediately. Let CInode::encode_inodestate() - do the job instead. This can avoid various races that early reply - is not allowed, caps that haven't been sent to client gets revoked. - - Fixes: http://tracker.ceph.com/issues/19635 - Signed-off-by: "Yan, Zheng" - (cherry picked from commit 799703a4acb49db0b6cc99a23e4326767e694c3a) - -commit d33b30cdb019937ff88f9724599f52f4e00d37cf -Author: Loic Dachary -Date: Thu Jun 1 11:37:20 2017 +0200 - - ceph-disk: do not setup_statedir on trigger - - trigger may run when statedir is unavailable and does not use it. - - Fixes: http://tracker.ceph.com/issues/19941 - - Signed-off-by: Loic Dachary - (cherry picked from commit 16bfbdd3d9988523bba31aace516c303057daa58) - -commit f46ccf2cb4701cd93cd9b15a4e57b5b97798b947 -Author: Nathan Cutler -Date: Fri Apr 21 11:05:05 2017 +0200 - - tests: rados: sleep before ceph tell osd.0 flush_pg_stats after restart - - Even though we wait for HEALTH_OK after restarting the daemons, they are not - ready to respond to flush_pg_stats. - - The reason why the osd is not ready for "tell" command after "ceph health" - shows that the cluster is "HEALTH_OK" is that the monitor fails to be notified - that the osd in question is not up in "heatbeat_interval". Because infernalis - does not have the osd_fast_fail_on_connection_refused support, the monitor - needs longer to detect that an osd is down, and osd_heartbeat_grace is used to - determine if an osd is down. - - References: http://tracker.ceph.com/issues/16239 - Signed-off-by: Nathan Cutler - Signed-off-by: Kefu Chai - -commit ab78cd040f6d3946ed40b6638ebcf52969a7cbb6 -Author: Yehuda Sadeh -Date: Wed Mar 8 14:52:34 2017 -0800 - - rgw: fix crash when listing objects via swift - - Fixes: http://tracker.ceph.com/issues/19249 - - Signed-off-by: Yehuda Sadeh - (cherry picked from commit a9ec5e8ce184e19c009863db4d3519f9d8af91bd) - - Conflicts: - src/rgw/rgw_rest_swift.cc ("key" element of RGWObjEnt struct - is not a reference; fix) - - (cherry picked from commit 92b35155ff7b7492f3c50bf4f2ff0ffef2ca1c55) - -commit 5a1e849ecf215d82e31b9bdd0970cb04200de2c9 -Merge: 2469085d57 66c3db7aee -Author: Nathan Cutler -Date: Mon Jun 19 22:46:47 2017 +0200 - - Merge pull request #14752 from cbodley/wip-19474 - - jewel: rgw: allow system users to read SLO parts - - Reviewed-by: Casey Bodley - -commit cda721bbbfae00ec4244718ae20cbd9ae914c630 -Author: Yan, Zheng -Date: Fri May 19 09:37:15 2017 +0800 - - client: update the 'approaching max_size' code - - The old 'approaching max_size' code expects MDS set max_size to - '2 x reported_size'. This is no longer true. The new code reports - file size when half of previous max_size increment has been used. - - Signed-off-by: "Yan, Zheng" - (cherry picked from commit 9316b0442c6f828dcf8da952e4c7a63c4db1398d) - - Conflicts: - src/client/Client.cc - in jewel, second argument to check_caps() is - a bool (see 0df562a8e13 which is not in jewel) - -commit 439f39128ec278ce78139d4b96ed098c68efa3f5 -Author: Yan, Zheng -Date: Wed May 17 19:08:37 2017 +0800 - - mds: limit client writable range increment - - For very large file, setting the writable range to '2 * file_size' - causes file recovery to run a long time. To recover a 1T file, Filer - needs to probe 2T~1T range. - - Fixes: http://tracker.ceph.com/issues/19955 - Signed-off-by: "Yan, Zheng" - (cherry picked from commit 538f35bef944b18e9bca2b15ed7f4e8807ef0554) - - Conflicts: - src/mds/Locker.h - in jewel, file_update_finish() has different - arguments than it does in master - -commit 06cf9f365033f7913051bdf4060f0bc6fc0444d7 -Author: Yan, Zheng -Date: Tue May 23 21:46:54 2017 +0800 - - osdc/Journaler: avoid executing on_safe contexts prematurely - - Journaler::_do_flush() can skip flushing some data when prezered - journal space isn't enough. Before updating Journaler::next_safe_pos, - we need to check if Journaler::_do_flush() has flushed enough data. - - Fixes: http://tracker.ceph.com/issues/20055 - Signed-off-by: "Yan, Zheng" - (cherry picked from commit 6511e7a9e35a14216c03cd6921ca4d232274f953) - -commit 2e299b50de4a297fee2aec21290632336d239857 -Author: Yan, Zheng -Date: Wed Apr 12 16:00:18 2017 +0800 - - osdc/Journaler: make header write_pos align to boundary of flushed entry - - This can speed up the process that detects and drops partial written - entry in the log tail. - - Signed-off-by: "Yan, Zheng" - (cherry picked from commit 8ae2962b79903e217fda83cea4140af64b5d6883) - - Conflicts: - src/osdc/Journaler.cc - 8d4f6b92cba is not being backported to jewel - src/osdc/Journaler.h - Journaler::Journaler initializer list is different in jewel, compared to master - -commit 2469085d57a05933589165f6f99a67b2e28c7022 -Merge: 42c3fbc129 d57437e338 -Author: John Spray -Date: Wed Jun 14 10:01:27 2017 -0400 - - Merge pull request #14672 from smithfarm/wip-19334-jewel - - jewel: MDS heartbeat timeout during rejoin, when working with large amount of caps/inodes - - Reviewed-by: John Spray - -commit 8ac0e5c363bd6439071d26874b6714cf2376736f -Author: Yan, Zheng -Date: Fri May 12 10:38:51 2017 +0800 - - pybind: fix cephfs.OSError initialization - - Traceback (most recent call last): - File "", line 1, in - File "cephfs.pyx", line 672, in cephfs.LibCephFS.open (/home/zhyan/Ceph/ceph-2/build/src/pybind/cephfs/pyrex/cephfs.c:10160) - File "cephfs.pyx", line 155, in cephfs.OSError.__init__ (/home/zhyan/Ceph/ceph-2/build/src/pybind/cephfs/pyrex/cephfs.c:1889) - TypeError: __init__() takes exactly 3 positional arguments (2 given) - - Signed-off-by: "Yan, Zheng" - (cherry picked from commit e6493f64ba4592b8dca54ece4464efa6c7f331a7) - -commit 09b9410c2e69a466b001d92fc14eb44d768009f1 -Author: Yan, Zheng -Date: Wed May 10 08:13:52 2017 +0800 - - pybind: fix open flags calculation - - (O_WRONLY | O_RDWR) is invaild open flags - - Fixes: http://tracker.ceph.com/issues/19890 - Signed-off-by: "Yan, Zheng" - (cherry picked from commit 2c25c99cb4572ffae97555a56b24a4c4097dcdec) - -commit 42c3fbc129cbb60d447c1a6b0402a9def1656446 -Merge: 71d45e1905 7347f11939 -Author: John Spray -Date: Wed Jun 14 09:43:05 2017 -0400 - - Merge pull request #14677 from smithfarm/wip-19665-jewel - - jewel: mds: C_MDSInternalNoop::complete doesn't free itself - - Reviewed-by: John Spray - -commit 71d45e190528124a8ff7e4674f7bfb7c340f80ee -Merge: 27c915f5d2 e6daee8a9f -Author: John Spray -Date: Wed Jun 14 09:35:08 2017 -0400 - - Merge pull request #15466 from smithfarm/wip-19762-jewel - - jewel: cephfs: non-local quota changes not visible until some IO is done - - Reviewed-by: John Spray - -commit 27c915f5d21bd84502e35eb269d955fafc47de0b -Merge: a76357622a db053da618 -Author: John Spray -Date: Wed Jun 14 09:32:47 2017 -0400 - - Merge pull request #14700 from smithfarm/wip-19709-jewel - - jewel: mds: enable start when session ino info is corrupt - - Reviewed-by: John Spray - -commit a76357622a2773850153aa1f6ea02b1737942c4b -Merge: 85aab833f7 db86a24e79 -Author: John Spray -Date: Wed Jun 14 09:31:22 2017 -0400 - - Merge pull request #14685 from smithfarm/wip-19675-jewel - - jewel: cephfs: Test failure: test_data_isolated (tasks.cephfs.test_volume_client.TestVolumeClient) - - Reviewed-by: John Spray - -commit 85aab833f7f66ac81d8b12f4203fa215787f8d2a -Merge: f6b395115e 7b9283beec -Author: John Spray -Date: Wed Jun 14 09:30:53 2017 -0400 - - Merge pull request #14684 from smithfarm/wip-19673-jewel - - jewel: cephfs: mds is crushed, after I set about 400 64KB xattr kv pairs to a file - - Reviewed-by: John Spray - -commit f6b395115e4de15d73269ff6b96f2ee0fd0ea9c3 -Merge: ff60fceb5f b52c508861 -Author: John Spray -Date: Wed Jun 14 09:30:13 2017 -0400 - - Merge pull request #14683 from smithfarm/wip-19671-jewel - - jewel: cephfs: MDS assert failed when shutting down - - Reviewed-by: John Spray - -commit ff60fceb5f5f7f9df4a48da4ad8bd863b9a04e2d -Merge: 6a6d57d2de 96e801fb53 -Author: John Spray -Date: Wed Jun 14 09:29:50 2017 -0400 - - Merge pull request #14682 from smithfarm/wip-19668-jewel - - jewel: cephfs: MDS goes readonly writing backtrace for a file whose data pool has been removed - - Reviewed-by: John Spray - -commit 6a6d57d2de674c5a971e999cb2731b2d8ae1b523 -Merge: 8260669efa f34489dd52 -Author: John Spray -Date: Wed Jun 14 09:27:13 2017 -0400 - - Merge pull request #14679 from smithfarm/wip-19666-jewel - - jewel: cephfs: The mount point break off when mds switch hanppened. - - Reviewed-by: John Spray - -commit 8260669efa74f41d3b1b1039fb7b34e070951c34 -Merge: 013529b61f 824b19a9a6 -Author: John Spray -Date: Wed Jun 14 09:26:40 2017 -0400 - - Merge pull request #14676 from smithfarm/wip-19619-jewel - - jewel: cephfs: MDS server crashes due to inconsistent metadata. - - Reviewed-by: John Spray - -commit 013529b61fc4fbf02656f7c6cb0baa1bc6004758 -Merge: b518522f64 eab56dae67 -Author: John Spray -Date: Wed Jun 14 09:24:38 2017 -0400 - - Merge pull request #14674 from smithfarm/wip-19482-jewel - - jewel: cephfs: No output for ceph mds rmfailed 0 --yes-i-really-mean-it command - - Reviewed-by: John Spray - -commit b518522f64b66f4a14618a9345b6314ca0f2c54c -Merge: 388e0d1bc3 63f41d543f -Author: John Spray -Date: Wed Jun 14 09:23:27 2017 -0400 - - Merge pull request #14671 from smithfarm/wip-19044-jewel - - jewel: tests: buffer overflow in test LibCephFS.DirLs - - Reviewed-by: John Spray - -commit 388e0d1bc35985a6916d3eb1ca5184a0907b6e8b -Merge: d7c7ce7ebb 7146816065 -Author: John Spray -Date: Wed Jun 14 09:23:13 2017 -0400 - - Merge pull request #14670 from smithfarm/wip-18949-jewel - - jewel: mds: avoid reusing deleted inode in StrayManager::_purge_stray_logged - - Reviewed-by: John Spray - -commit d7c7ce7ebbb663dab0dfa8058c845d494d7615cc -Merge: d717ef73bc d8b139b584 -Author: John Spray -Date: Wed Jun 14 09:22:28 2017 -0400 - - Merge pull request #14669 from smithfarm/wip-18900-jewel - - jewel: cephfs: Test failure: test_open_inode - - Reviewed-by: John Spray - -commit d717ef73bc0d5ef24551ec9157385c8b0521380b -Merge: c2a3b7567f 36c86f71ef -Author: John Spray -Date: Wed Jun 14 09:21:43 2017 -0400 - - Merge pull request #14668 from smithfarm/wip-18705-jewel - - jewel: mds: fragment space check can cause replayed request fail - - Reviewed-by: John Spray - -commit c2a3b7567fbe0b2f62bcd38cd9bb9a5a2a238743 -Merge: fd9256b770 5b56214519 -Author: John Spray -Date: Tue Jun 13 19:05:55 2017 +0100 - - Merge pull request #14698 from smithfarm/wip-19677-jewel - - jewel: cephfs: ceph-fuse does not recover after lost connection to MDS - - Reviewed-by: John Spray - -commit fd9256b77010066c934fd0016eb6d3f9c1fb54e1 -Merge: 26ada59c81 c49b114e8d -Author: Abhishek L -Date: Fri Jun 9 19:39:17 2017 +0200 - - Merge pull request #14766 from smithfarm/wip-19757-jewel - - jewel: rgw: fix failed to create bucket if a non-master zonegroup has a single zone - - Reviewed-by: Abhishek Lekshmanan - Reviewed-by: Casey Bodley - -commit 26ada59c810bfda5c16f2d935a2b98711e7b2c76 -Merge: fdd25c2bdb e552d91f73 -Author: Abhishek L -Date: Fri Jun 9 19:38:46 2017 +0200 - - Merge pull request #14787 from linuxbox2/jewel-rgw-shard-limit-ck - - jewel: rgw: add bucket size limit check to radosgw-admin - - - Reviewed-by: Abhishek Lekshmanan - Reviewed-by: Casey Bodley - -commit fdd25c2bdb9ec6bb8e3061088b8782bfb2331bc4 -Merge: cfd6750416 fb3ee2efcc -Author: Abhishek L -Date: Fri Jun 9 19:38:23 2017 +0200 - - Merge pull request #14789 from mdw-at-linuxbox/wip-jewel-rgw-rvk - - jewel: rgw: swift: disable revocation thread if sleep == 0 || cache_size == 0 - - - Reviewed-by: Abhishek Lekshmanan - Reviewed-by: Casey Bodley - -commit cfd6750416baeb2bbd3836bc7dc55d83fc214c20 -Merge: ce1fc3492e 86980a045b -Author: Abhishek L -Date: Fri Jun 9 19:36:02 2017 +0200 - - Merge pull request #14815 from smithfarm/wip-19786-jewel - - jewel: rgw: failure to create s3 type subuser from admin rest api - - Reviewed-by: Abhishek Lekshmanan - Reviewed-by: Casey Bodley - -commit ce1fc3492e87c669f7059c2047a3bed077418a89 -Merge: 7ca0252560 aa99558934 -Author: Nathan Cutler -Date: Wed Jun 7 20:13:15 2017 +0200 - - Merge pull request #15312 from theanalyst/wip-20078 - - jewel: rgw: only append zonegroups to rest params if not empty - - Reviewed-by: Casey Bodley - -commit 7ca0252560d1ff23384afb50a9c2ae2aad2ce85c -Merge: 62c500f522 59bd6711a4 -Author: Nathan Cutler -Date: Wed Jun 7 13:48:06 2017 +0200 - - Merge pull request #15382 from theanalyst/wip-mem-leak2 - - jewel: rgw:fix memory leaks in data/md sync - - Reviewed-by: Casey Bodley - -commit aa0cd461df5fee6d143bc07440ec6de829761cef -Author: Brad Hubbard -Date: Mon May 22 13:21:25 2017 +1000 - - osd: Move scrub sleep timer to osdservice - - PR 14886 erroneously creates a scrub sleep timer for every pg resulting - in a proliferation of threads. Move the timer to the osd service so - there can be only one. - - Fixes: http://tracker.ceph.com/issues/19986 - - Signed-off-by: Brad Hubbard - (cherry picked from commit f110a82437df79dc20207d296e8229fc0e9ce18b) - - Conflicts: - src/osd/PG.cc - ceph_clock_now requires a CephContext argmunent - in Jewel - -commit c47bd0562b1187ffb0b1b2c1ef5f105aa7951d10 -Author: Brad Hubbard -Date: Mon Apr 24 14:10:47 2017 +1000 - - osd: Implement asynchronous scrub sleep - - Rather than blocking the main op queue just do an async sleep. - - Fixes: http://tracker.ceph.com/issues/19497 - - Signed-off-by: Brad Hubbard - (cherry picked from commit 7af3e86c2e4992db35637864b83832535c94d0e6) - -commit e6daee8a9fbc576da2a03550a81056d093a516c9 -Author: Nathan Cutler -Date: Sun Jun 4 20:39:58 2017 +0200 - - Client.cc: adjust Client::_getattr calls - - Signed-off-by: Nathan Cutler - -commit a2c7a2262ac8ecbea78f09e6e8e6a37498568d57 -Author: John Spray -Date: Wed Mar 15 19:36:08 2017 +0000 - - qa/cephfs: use getfattr/setfattr helpers - - Signed-off-by: John Spray - (cherry picked from commit dd43d3bc646aeab88486b0963fc83de0b18800c4) - - Conflicts: - qa/tasks/cephfs/test_data_scan.py: difference in the - self._mount.run_shell() call in NonDefaultLayout::write (which is - being dropped by this commit) - in jewel it has "sudo", and in - master it doesn't - -commit 12aa35a6d50f612df77199ac4f35c7baeed0583e -Author: John Spray -Date: Wed Jun 22 13:00:44 2016 +0100 - - tasks/cephfs: fix race while mounting - - This could fail if the mount hadn't finished - coming up. - - Signed-off-by: John Spray - (cherry picked from commit adfb757c898a80f18c15dafd02e29840c5931c87) - -commit a7b699269b65c76361fcb1d10593812be40c7612 -Author: John Spray -Date: Wed Mar 15 19:26:30 2017 +0000 - - qa: add test for reading quotas from different clients - - Fixes: http://tracker.ceph.com/issues/17939 - Signed-off-by: John Spray - (cherry picked from commit 61617f8f10a6322603a9add77980865cd972ef97) - -commit 8b8ee392b8093b9b140dbbe895691f69ae40440f -Author: John Spray -Date: Wed Mar 15 17:51:44 2017 +0000 - - client: _getattr on quota_root before using in statfs - - ...so that after someone adjusts the quota settings - on an inode that another client is using as its mount root, - the change is visible immediately on the other client. - - Signed-off-by: John Spray - (cherry picked from commit 3d25941aadd223669448d0f5d3c0bd1fefa72308) - -commit dd7d59a08141d6a24b172c22f5e27c8962e25fb9 -Author: John Spray -Date: Wed Mar 15 15:32:47 2017 +0000 - - client: getattr before read on ceph.* xattrs - - Previously we were returning values for quota, layout - xattrs without any kind of update -- the user just got - whatever happened to be in our cache. - - Clearly this extra round trip has a cost, but reads of - these xattrs are fairly rare, happening on admin - intervention rather than in normal operation. - - Fixes: http://tracker.ceph.com/issues/17939 - Signed-off-by: John Spray - (cherry picked from commit 532dc4b68e538c189ef828f67cecd0d647a62250) - -commit 62c500f52240eaa5faadd3795bd9ec84bdcbc6c7 -Merge: 2badc2416c 8dd93cabd5 -Author: David Zafman -Date: Fri Jun 2 09:54:22 2017 -0700 - - Merge pull request #15416 from dzafman/wip-20126 - - Reviewed-by: Josh Durgin - -commit 8dd93cabd52cbafc29a47862f343431eb6f1cfe3 -Author: David Zafman -Date: Wed May 31 15:39:19 2017 -0700 - - osd: Object level shard errors are tracked and used if no auth available - - Shards with object mismatch are tracked to mark them inconsistent - Fix test because storing omap_digest in object_info not behaving as before - - Fixes: http://tracker.ceph.com/issues/20089 - - Signed-off-by: David Zafman - - (cherry picked from commit 1cacbea763c7aabfeaaf4bd5e878301044184117) - - Conflicts: - src/test/osd/osd-scrub-repair.sh (no alloc_hint in object_info) - -commit 59bd6711a47c354117a612dd4ef033d70d449383 -Author: weiqiaomiao -Date: Wed Jun 1 17:20:49 2016 +0800 - - rgw:fix memory leaks - - Signed-off-by: weiqiaomiao - (cherry picked from commit 73e5be2b6133cf4caa0e5e5c8c9eae748b785dbf) - -commit b429fa1807062716c9705ddcf316ed9b2741cc43 -Author: Casey Bodley -Date: Fri May 5 14:56:40 2017 -0400 - - cls/rgw: list_plain_entries() stops before bi_log entries - - list_plain_entries() was using encode_obj_versioned_data_key() to set - its end_key, which gives a prefix of BI_BUCKET_OBJ_INSTANCE_INDEX[=2] - - that range between start_key and end_key would not only span the - BI_BUCKET_OBJS_INDEX[=0] prefixes, but BI_BUCKET_LOG_INDEX[=1] prefixes - as well. this can result in list_plain_entries() trying and failing to - decode a rgw_bi_log_entry as a rgw_bucket_dir_entry - - Fixes: http://tracker.ceph.com/issues/19876 - - Signed-off-by: Casey Bodley - (cherry picked from commit b29a1633a57abf443d5790c13d680d2917f86037) - -commit 2badc2416c9e16babbc91364502dcb03877e88bf -Merge: 5d2a68eb90 1f895c2403 -Author: Sage Weil -Date: Tue May 30 09:39:47 2017 -0500 - - Merge pull request #15360 from liewegas/wip-jewel-master-mixed - - qa/workunits/rados/test-upgrade-*: whitelist tests for master - - Reviewed-by: Kefu Chai - -commit 1f895c24030b79ae9f2eae16b09582fc72928b6a -Author: Sage Weil -Date: Tue May 30 09:58:09 2017 -0400 - - qa/workunits/rados/test-upgrade-*: whitelist tests for master - - The jewel-x upgrade test now runs this script against a mixed cluster on - a machine with code from master installed. That means we have to skip - any new tests that will fail on a mixed cluster. - - Signed-off-by: Sage Weil - -commit 5d2a68eb903f533689d1a1cbc410ca940baeaff3 -Merge: 998d1ee4f5 81e35b9416 -Author: Nathan Cutler -Date: Sat May 27 09:21:44 2017 +0200 - - Merge pull request #15208 from liewegas/wip-sortbitwise-jewel - - mon: fix 'sortbitwise' warning on jewel - - Reviewed-by: Josh Durgin - -commit 998d1ee4f503cb4875283fe8d2140f3b10a26643 -Merge: 54bc1e13ea 99c65bbc18 -Author: Nathan Cutler -Date: Sat May 27 02:46:55 2017 +0200 - - Merge pull request #14851 from yehudasa/wip-rgw-support-ragweed-jewel - - jewel: rgw: add apis to support ragweed suite - - Reviewed-by: Nathan Cutler - -commit aa99558934c6143cb35e722148b87f35ccecc397 -Author: Yehuda Sadeh -Date: Tue Aug 23 10:22:42 2016 -0700 - - rgw: rest conn functions cleanup, only append zonegroup if not empty - - Signed-off-by: Yehuda Sadeh - (cherry picked from commit 4e41af19846db75081cb0ddb7b33dc2bb9321ace) - -commit 5d90798432a6446636699fc03b2f548010b1212f -Author: Karol Mroz -Date: Thu Mar 17 10:32:14 2016 +0100 - - rgw: rest and http client code to use param vectors - - Replaces param/header lists with vectors. In these cases, we're only ever - adding to the back of the list, so a vector should be more efficient. - Also moves param_pair_t/param_vec_t higher up the include chain for - cleaner function signatures. - - Signed-off-by: Karol Mroz - (cherry picked from commit d4a2527872e0f5c3ae2874bb7d0ff459ae83cfd4) - - Conflicts: - src/rgw/rgw_http_client.cc - trivial ws conflict on rebase - -commit 54bc1e13ea19642c8d5893f29e96ee5c053f9c59 -Merge: 966f222917 43327f83ef -Author: Alfredo Deza -Date: Thu May 25 12:45:30 2017 -0400 - - Merge pull request #14765 from smithfarm/wip-18972-jewel - - jewel: ceph-disk does not support cluster names different than 'ceph' - - Reviewed-by: Alfredo Deza - -commit 81e35b941659a0f6dff8a935c27c7dd6d5cc4213 -Author: Sage Weil -Date: Wed May 24 09:48:11 2017 -0400 - - qa/suites/rados/singleton-nomsgr/*: set sortbitwise after upgrade - - Signed-off-by: Sage Weil - -commit f2814e4dbb2821a415310559fce405c2eab23947 -Author: huanwen ren -Date: Tue Dec 27 10:54:45 2016 +0000 - - mon/OSDMonitor: fixup sortbitwise flag warning - - "ceph -s" does not report warning when using - command "ceph osd unset sortbitwise" to drop - sortbitwise flag. - we should use "osdmap.get_up_osd_features() & - CEPH_FEATURE_OSD_BITWISE_HOBJ_SORT" - instead of "(osdmap.get_features(CEPH_ENTITY_TYPE_OSD, NULL) & - CEPH_FEATURE_OSD_BITWISE_HOBJ_SORT)", - because osdmap.get_features only get local "features" - - Signed-off-by: huanwen ren - (cherry picked from commit c25ee187e28724846d0011cd8145e16956d3636e) - -commit 02617188688eebde759c375a2257e076e4538491 -Author: Sage Weil -Date: Tue Nov 1 12:09:57 2016 -0400 - - mon: remove config option to disable no sortbitwise warning - - We'll require this soon. - - Signed-off-by: Sage Weil - (cherry picked from commit 04e3319e8aecde9ca58ccb7c89016f9079c7d657) - -commit 966f2229178c5d72722982d0f6a40f4d01210d9a -Merge: 6537fc741c 4ceaa7cce9 -Author: Nathan Cutler -Date: Wed May 17 09:29:21 2017 +0200 - - Merge pull request #13450 from dreamhost/wip-18887-jewel - - jewel: msg: IPv6 Heartbeat packets are not marked with DSCP QoS - simple messenger - - Reviewed-by: Josh Durgin - -commit bb79663490468b4ac2832aa9e3cbac019c1f712d -Author: Jan Fajerski -Date: Thu Apr 20 18:38:43 2017 +0200 - - fs: normalize file open flags internally used by cephfs - - The file open flags (O_foo) are platform specific. Normalize these flags - before they are send to the MDS. For processing of client messages the - MDS should only compare to these normalized flags. - Otherwise this can lead to bogus flags getting transmitted on ppc64. - - Signed-off-by: Jan Fajerski - (cherry picked from commit 88d2da5e93198e69435e288ce00d216d5fe27f80) - - Conflicts: - src/client/Client.cc - Conflicts can be resolved by choosing changes from HEAD and - adding a call to ceph_flags_sys2wire where flags are logged. - src/mds/Server.cc - Conflicts can be resolved by choosing changes from HEAD and - while making sure that the MDS compares request O_ flags the - the CEPH_O_ flags, since all wire O_ flags are normalized. - -commit 4ceaa7cce9c9132d47564d79204b48b1d02e531c -Author: Robin H. Johnson -Date: Wed May 3 22:31:40 2017 -0700 - - msg/simple/Pipe: manual backport of fix in PR#14795 - - Manual backport of errno fixup from PR#14795 - (6f1037e22c2a304795895498cdc955e0ef80f8e3), as noted by - https://github.com/ceph/ceph/pull/13450#discussion_r114696885. - - Signed-off-by: Robin H. Johnson - -commit 6537fc741c3267472e607e581c002854d8eafd46 -Merge: af31b453f4 82ea0971b3 -Author: Nathan Cutler -Date: Wed May 3 13:04:27 2017 +0200 - - Merge pull request #14667 from smithfarm/wip-18699-jewel - - jewel: client: fix the cross-quota rename boundary check conditions - - Reviewed-by: Gregory Farnum - -commit af31b453f404cb73ee7bdb8b6b02d71ad8aab7e7 -Merge: 472034cbe0 335258f975 -Author: Nathan Cutler -Date: Wed May 3 10:33:24 2017 +0200 - - Merge pull request #14332 from shinobu-x/wip-19396-jewel - - jewel: Objecter::epoch_barrier isn't respected in _op_submit() - - Reviewed-by: Josh Durgin - -commit 472034cbe0735131f3419205e9f7e6530377bfba -Merge: afe98ba074 acf608a903 -Author: Nathan Cutler -Date: Wed May 3 10:32:37 2017 +0200 - - Merge pull request #14204 from dzafman/wip-18533-jewel - - jewel: core: two instances of omap_digest mismatch - - Reviewed-by: Josh Durgin - -commit afe98ba07436ea105f4c9c42a52795ce3757419e -Merge: 0353a91bfd 043d70461c -Author: Nathan Cutler -Date: Wed May 3 10:31:38 2017 +0200 - - Merge pull request #13884 from shinobu-x/wip-19119-jewel - - jewel: pre-jewel "osd rm" incrementals are misinterpreted - - Reviewed-by: Josh Durgin - -commit 0353a91bfd799077ddd87de901e8a07f5da6a13d -Merge: 460b12c259 608785a007 -Author: Nathan Cutler -Date: Wed May 3 10:30:32 2017 +0200 - - Merge pull request #13647 from mslovy/wip-19083-jewel - - jewel: osd: preserve allocation hint attribute during recovery - - Reviewed-by: Gregory Farnum - -commit 460b12c259f5563d9d1b2477149fe79486ba5bcd -Merge: 630cfca36c 905c4acb99 -Author: Nathan Cutler -Date: Fri Apr 28 10:24:35 2017 +0200 - - Merge pull request #14791 from smithfarm/wip-19774-jewel - - jewel: osd: promote throttle parameters are reversed - - Reviewed-by: Sage Weil - -commit 630cfca36c5a42a58882966c5598752aac9ff54e -Merge: 013e781b45 3ec1a9bf16 -Author: Nathan Cutler -Date: Fri Apr 28 10:23:45 2017 +0200 - - Merge pull request #14763 from smithfarm/wip-19562-jewel - - jewel: api_misc: [ FAILED ] LibRadosMiscConnectFailure.ConnectFailure - - Reviewed-by: Sage Weil - -commit 99c65bbc1899663d0b23e2cec3d6e516d0e95ad7 -Author: Yehuda Sadeh -Date: Mon Jan 9 13:04:43 2017 -0800 - - rgw: new rest api to retrieve object layout - - Signed-off-by: Yehuda Sadeh - (cherry picked from commit 2768583dc486109e49d209243675b99fdd39e92c) - -commit 33745a342433716ad8a07ef846dbea9b57de5017 -Author: Yehuda Sadeh -Date: Thu Jan 5 13:47:24 2017 -0800 - - rgw: rest api to read zone config params - - Signed-off-by: Yehuda Sadeh - (cherry picked from commit a220a2efbfd675e6abf14ae33c21005bcbf6dadf) - -commit 013e781b45e28af15da4ddc32025aad713f128e8 -Merge: d144d99f00 b698d1fa4c -Author: Nathan Cutler -Date: Thu Apr 27 21:46:41 2017 +0200 - - Merge pull request #14481 from dillaman/wip-19468-jewel - - jewel: librbd: is_exclusive_lock_owner API should ping OSD - - Reviewed-by: Jason Dillaman - -commit d144d99f008776dc02c5838feb5a9eb9b84a400b -Merge: 0f0cd82eda dbe90c79b8 -Author: Nathan Cutler -Date: Thu Apr 27 21:05:11 2017 +0200 - - Merge pull request #14666 from smithfarm/wip-19612-jewel - - jewel: librbd: Issues with C API image metadata retrieval functions - - Reviewed-by: Jason Dillaman - -commit 0f0cd82edabcb2feb0b29793e5b45eb424f1c42d -Merge: eba821ce9c 216156b5d4 -Author: Nathan Cutler -Date: Thu Apr 27 21:04:27 2017 +0200 - - Merge pull request #14664 from smithfarm/wip-19325-jewel - - jewel: rbd: [api] temporarily restrict (rbd_)mirror_peer_add from adding multiple peers - - Reviewed-by: Jason Dillaman - -commit eba821ce9c61bcb197895e6949ce30a5c07097fa -Merge: 25e29c9215 b8fd297eb6 -Author: Nathan Cutler -Date: Thu Apr 27 20:46:08 2017 +0200 - - Merge pull request #14694 from ceph/wip-bp-systemd - - [backport] qa/tasks: systemd test backport to jewel - - Reviewed-by: Nathan Cutler - -commit 25e29c921557883da4bf48ff2496bcec6448e6ec -Merge: d7b63e21ca 7c6c3c753c -Author: Nathan Cutler -Date: Thu Apr 27 08:44:26 2017 +0200 - - Merge pull request #13544 from shinobu-x/wip-18932-jewel - - jewel: tests: 'ceph auth import -i' overwrites caps, should alert user before overwrite - - Reviewed-by: Kefu Chai - -commit acf608a9034e915e38ccea6002ee808c46620433 -Author: David Zafman -Date: Mon Mar 20 17:28:45 2017 -0700 - - filestore, tools: Fix logging of DBObjectMap check() repairs - - Signed-off-by: David Zafman - (cherry picked from commit 1704f62c0831e6b07138f7dd14a89fef3c9ed2c1) - -commit fecc52338b2a58bf1730f7b7a3e4a293e45160d2 -Author: David Zafman -Date: Fri Mar 3 15:04:02 2017 -0800 - - osd: Simplify DBObjectMap by no longer creating complete tables - - Bump the version for new maps to 3 - Make clone less efficient but simpler - Add rename operation (use instead of clone/unlink) - For now keep code that understands version 2 maps - - Signed-off-by: David Zafman - (cherry picked from commit 738156a99ed1caf61e5a8230eb8048360056c08e) - - No ghobject_t::operator>() so use Kraken cmp_bitwise() instead - Need to use MIN_GHOBJ/MAX_GHOBJ instead of std::min/std::max - -commit 6902c3141eeaefaacd92f33877cf319872f626c7 -Author: David Zafman -Date: Wed Feb 15 16:17:32 2017 -0800 - - ceph-osdomap-tool: Fix seg fault with large amount of check error output - - Signed-off-by: David Zafman - (cherry picked from commit 1dda0411f4fbb14ce1e0062da9f14ec3af505d39) - -commit 4a3e4bcf40cd004fc53e7be467a29084dedc3e1c -Author: David Zafman -Date: Wed Feb 15 15:02:33 2017 -0800 - - osd: Add automatic repair for DBObjectMap bug - - Add repair command to ceph-osdomap-tool too - - Under some situations the previous rm_keys() code would - generated a corrupt complete table. There is no way to - figure out what the table should look like now. By removing - the entries we fix the corruption and aren't much worse off - because the corruption caused some deleted keys to re-appear. - - This doesn't breaking the parent/child relationship during - repair because some of the keys may still be contained - in the parent. - - Signed-off-by: David Zafman - (cherry picked from commit 4cd3c74c928a32e065ed9543d6c91d8718a6ae3d) - - Conflicts: - src/os/filestore/DBObjectMap.h (trivial) - -commit d4f0ac0a405266f638b25ec475c0110741e3c431 -Author: David Zafman -Date: Wed Feb 15 14:59:40 2017 -0800 - - ceph-osdomap-tool: Fix tool exit status - - Signed-off-by: David Zafman - (cherry picked from commit 666f14ed90655a2d1bedde8561949625db7a9e6c) - -commit 5f36c319cd7fa4be15efd27b8aabbebc99d8999c -Author: Samuel Just -Date: Fri Feb 10 15:51:42 2017 -0800 - - DBObjectMap: rewrite rm_keys and merge_new_complete - - Leverage the updated in_complete_region and needs_parent to simplify - these methods. - - Signed-off-by: Samuel Just - (cherry picked from commit c4dffb68eaafe724f7fdae93a4285a7f8003ea29) - -commit 1fe4b856a37b29c85a9317e514c7f15f8e4905d5 -Author: Samuel Just -Date: Fri Feb 10 15:50:57 2017 -0800 - - DBObjectMap: strengthen in_complete_region post condition - - Previously, in_complete_region didn't guarantee anything about - where it left complete_iter pointing. It will be handy for - complete_iter to be pointing at the lowest interval which ends - after to_test. Make it so. - - Signed-off-by: Samuel Just - (cherry picked from commit 97b35f4d7d4862da4b6f50ecaef0d292a671fd04) - -commit 85f2151fec991e5db13d8e6f44b27e092605fb35 -Author: Samuel Just -Date: Fri Feb 10 15:48:57 2017 -0800 - - DBObjectMap: fix next_parent() - - The previous implementation assumed that - lower_bound(parent_iter->key()) always leaves the iterator - on_parent(). There isn't any guarantee, however, that that - key isn't present on the child as well. - - Signed-off-by: Samuel Just - (cherry picked from commit 74a7631d0938d7b44894f022224eab10a90d5cec) - -commit 484ccda5c53a8ee151b9f97687edd160ad7ebbd7 -Author: Samuel Just -Date: Thu Feb 9 10:47:59 2017 -0800 - - test_object_map: add tests to trigger some bugs related to 18533 - - Signed-off-by: Samuel Just - (cherry picked from commit f131dbcf5bb17107c029f942a57e9bf4432a26ee) - -commit cdeb690869bf6dd52226476c21514cf03ff37d6e -Author: David Zafman -Date: Tue Feb 14 12:40:33 2017 -0800 - - test: Add ceph_test_object_map to make check tests - - Signed-off-by: David Zafman - (cherry picked from commit 0e97a01bd7291458881ee53cece2d887f6333669) - -commit cf5d588d39498c8d65bc64bd2935aecac8546e40 -Author: David Zafman -Date: Wed Feb 8 18:56:27 2017 -0800 - - ceph-osdomap-tool: Add --debug and only show internal logging if enabled - - Signed-off-by: David Zafman - (cherry picked from commit 5fb2b2d13953979e5da9f571ab8c4b0b510b8368) - -commit 4c4a06ff525d9fa2271099db73701c7994054d36 -Author: David Zafman -Date: Wed Feb 8 18:55:48 2017 -0800 - - osd: DBOjectMap::check: Dump complete mapping when inconsistency found - - Signed-off-by: David Zafman - (cherry picked from commit fcf1e17c645e8fad5216c3e59627c817e5c858c7) - -commit 6c128ff8c6944e57059008959f49bd03635c5417 -Author: David Zafman -Date: Wed Feb 8 15:38:51 2017 -0800 - - test_object_map: Use ASSERT_EQ() for check() so failure doesn't stop testing - - Signed-off-by: David Zafman - (cherry picked from commit 053a273cbc02d6902a4bb1f11db1ea946498df3a) - -commit aa769a9d08ac7490f5063bb44c3a44c77a5d7232 -Author: David Zafman -Date: Wed Feb 8 10:02:40 2017 -0800 - - tools: Check for overlaps in internal "complete" table for DBObjectMap - - Changed check to return an error count and fix tool error message - - Signed-off-by: David Zafman - (cherry picked from commit e5e8eb962db6187ea19b96ba29ac83469c90b4ea) - - Conflicts: - src/os/filestore/DBObjectMap.h (trivial) - -commit 761ee7c6af8802ab6b668a0b7ccaa819b2764456 -Author: David Zafman -Date: Wed Feb 8 09:40:49 2017 -0800 - - tools: Add dump-headers command to ceph-osdomap-tool - - Signed-off-by: David Zafman - (cherry picked from commit f4101591ad701a62fe027c4744ca8ea505f44bdc) - - Conflicts: - src/os/filestore/DBObjectMap.h (trivial) - -commit 117db1c6d6952d181614f4fe22f0b1866eed6f10 -Author: David Zafman -Date: Mon Feb 6 21:09:42 2017 -0800 - - tools: Add --oid option to ceph-osdomap-tool - - Signed-off-by: David Zafman - (cherry picked from commit 2d94889e9ee3359017b1efd560f3557ce03ccee6) - -commit 4d8120d3b3975ce101f6272240e5e43bcfe0e742 -Author: David Zafman -Date: Mon Feb 6 21:31:18 2017 -0800 - - osd: Remove unnecessary assert and assignment in DBObjectMap - - Fix and add comment(s) - - Signed-off-by: David Zafman - (cherry picked from commit 937e6a03ea4692cc44d53faa0615f8e808c9eb03) - -commit 86980a045b7176428062a19758d145e38e515b40 -Author: snakeAngel2015 -Date: Mon Jul 18 14:51:37 2016 +0800 - - rgw: add suport for creating S3 type subuser of admin rest api - - Fixes: http://tracker.ceph.com/issues/16682 - - The original code cannot support create s3 type subuser of admin rest api as when i execute the following command: - - ./s3curl.pl --id=personal --put -- http://radosgw.h3c.com:8000/admin/user?subuser\&uid=yrf2\&subuser=yrf2:yrf1\&key-type=s3 -v - - it would return msg as follows : - - < HTTP/1.1 403 Forbidden - < Date: Thu, 14 Jul 2016 07:04:40 GMT - * Server Apache/2.4.7 (Ubuntu) is not blacklisted - < Server: Apache/2.4.7 (Ubuntu) - < x-amz-request-id: tx00000000000000006608f-0057873988-8551-slave - < Accept-Ranges: bytes - < Content-Length: 114 - < Content-Type: application/json - < - * Connection #0 to host slave.com left intact - {"Code":"InvalidAccessKeyId","RequestId":"tx00000000000000006608f-0057873988-8551-slave","HostId":"8551-slave-us"} - - But i have modified the codes for support it ,and it will return actual msg as follows : - - "subusers": [ - { - "id": "yrf2:yrf1", - "permissions": "" - } - ], - "keys": [ - { - "user": "yrf2", - "access_key": "B46PXYFEWUX0IMHGKP8C", - "secret_key": "2JYxywXizqwiiMd74UXrJdSJMPNlBtYwF5z8rNvh" - }, - { - "user": "yrf2:yrf1", - "access_key": "INO55WXJ7JQ1ZZGSAB6B", - "secret_key": "GgCKEfF9hArV2hglunbO7KtvKZnbhmsDpqjSj5DL" - } - ], - - Please check it ,thanks . - - Signed-off-by: snakeAngel2015 - (cherry picked from commit 6535f6ad2137ee55bf5531e865c05aa10bd39bd0) - -commit d7b63e21ca9818e21afeb2945b882e24b7a9b10b -Merge: a3fae531cd f32b5c613a -Author: Nathan Cutler -Date: Wed Apr 26 21:32:31 2017 +0200 - - Merge pull request #14809 from tchaikov/wip-18193-jewel - - jewel: tests: test/librados/tmap_migrate: g_ceph_context->put() upon return - - Reviewed-by: Nathan Cutler - -commit a3fae531cdb8c2c64c66d38c947ca139f4ede83d -Merge: 89d6ddb1f5 6c4826606d -Author: Nathan Cutler -Date: Wed Apr 26 21:32:09 2017 +0200 - - Merge pull request #14701 from smithfarm/wip-18193-jewel - - jewel: core: transient jerasure unit test failures - - Reviewed-by: Kefu Chai - -commit 89d6ddb1f5ee08b0f019dfaadc5857a788c92a09 -Merge: c5f1fce21d be9e83281b -Author: Matt Benjamin -Date: Wed Apr 26 14:28:29 2017 -0400 - - Merge pull request #14776 from linuxbox2/jewel-pullup-civet-chunked - - [DNM] jewel: pullup civet chunked - -commit c5f1fce21d0996cce751b6bcca5e57da2cafc135 -Merge: 4d97e0ba8e 97cd21afc5 -Author: Nathan Cutler -Date: Wed Apr 26 19:32:07 2017 +0200 - - Merge pull request #14416 from smithfarm/wip-19557-jewel - - jewel: tests: upgrade/hammer-x failing with OSD has the store locked when Thrasher runs ceph-objectstore-tool on down PG - - Reviewed-by: Kefu Chai - -commit be9e83281b8c765cd111d5687a516fcd3ca521a5 -Author: Matt Benjamin -Date: Tue Apr 25 09:44:39 2017 -0400 - - civetweb: pullup chunked encoding by Marcus - - Fixes: http://tracker.ceph.com/issues/19736 - - Signed-off-by: Matt Benjamin - -commit 608785a0079f807ff860c56d96b1b67bf6a2ed74 -Author: yaoning -Date: Fri Jun 24 09:51:07 2016 +0800 - - os: make zero values noops for set_alloc_hint() in FileStore - - Signed-off-by: yaoning - (cherry picked from commit e2ec24f61b55457caccefecd56f9f08b98264802) - -commit f32b5c613a8acd50e32747c3581131d28d209efa -Author: Kefu Chai -Date: Wed Apr 26 22:58:30 2017 +0800 - - test/librados/tmap_migrate: g_ceph_context->put() upon return - - Signed-off-by: Kefu Chai - Conflict: test/librados/tmap_migrate.cc - this change is not cherry-picked from master, because tmap_migrate was - removed in master. so we are applying the same change in - cb1cda96713b2ec0f6418c4cbe3d964c2020729c to this test. - -commit 905c4acb99f9ea78ff615034dae969ab089bda06 -Author: Mark Nelson -Date: Tue May 3 09:56:47 2016 -0500 - - Fix reveresed promote throttle default parameters. - - Signed-off-by: Mark Nelson - (cherry picked from commit 793ceac2f3d5a2c404ac50569c44a21de6001b62) - -commit fb3ee2efcc13c37db90faa1bc8bad584bab22efa -Author: Marcus Watts -Date: Thu Apr 13 05:33:55 2017 -0400 - - rgw: swift: disable revocation thread if sleep == 0 || cache_size == 0 - - Keystone tokens can be revoked. This causes them to fail - validation. However, in ceph, we cache them. As long as - they're in the cache we trust them. To find revoked tokens - there's a call OSI-PKI/revoked but that's only useful for - pki tokens. Installations using fernet/uuid may not even - have the proper credentials to support the call, in which - case the call blows up in various ways filling up logs - with complaints. - - This code makes the revocation thread optional; by disabling it, - the complaints go away. A further fix is in the works - to use other more modern calls available in modern keystone - installations to properly deal with non-PKI/PKIZ tokens. - - (NB: jewel has this logic in src/rgw/rgw_swift.cc not in src/rgw/rgw_keystone.h) - - To disable the revocation thread, use at least one of these: - rgw_keystone_token_cache_size = 0 - using this will cause tokens to be validated on every call. - You may instead want to set - rgw_keystone_revocation_interval = 0 - using just this will disable the revocation thread, - but leaves the cache in use. That avoids the extra - validation overhead, but means token revocation won't - work very well. - - Fixes: http://tracker.ceph.com/issues/9493 - Fixes: http://tracker.ceph.com/issues/19499 - - Signed-off-by: Marcus Watts - (cherry picked from commit 003291a8cbca455c0e8731f66759395a0bb1f555) - -commit 4d97e0ba8ebc89b7797b0936a9e046ef59cc3899 -Merge: cc820a0d4c c2efeb4b62 -Author: Nathan Cutler -Date: Tue Apr 25 21:06:26 2017 +0200 - - Merge pull request #14686 from smithfarm/wip-19686-jewel - - jewel: osd: Give requested scrubs a higher priority - - Reviewed-by: David Zafman - -commit e552d91f73d996c44821f5fbfb28cfc3e5cddc9a -Author: Matt Benjamin -Date: Sat Nov 5 13:13:47 2016 -0400 - - rgw: add bucket size limit check to radosgw-admin - - The change adds a new list of all buckets x all users, with - fields for bucket name, tenant name, current num_objects, - current num_shards, current objects per shard, and the - corresponding fill_status--the latter consisting of 'OK', - 'WARN %', or 'OVER %.' - - The warning check is relative to two new tunables. The threshold - max objects per shard is set as rgw_bucket_safe_max_objects_per_shard, - which defaults to 100K. The value rgw_bucket_warning_threshold is - a percent of the current safe max at which to warn (defaults to - 90% of full). - - From review: - - * fix indentation (rgw_admin) - * if user a user_id is provided, check only buckets for that user - * update shard warn pct to be pct-of-fill (not 100 - pct-of-fill) - * print only buckets near or over per-shard limit, if --warnings-only - * s/bucket limitcheck/bucket limit check */ - * sanity shard limit should be 90, not 10 (because that changed) - * fixes for memleaks and other points found by cbodley - - Fixes: http://tracker.ceph.com/issues/17925 - - Signed-off-by: Matt Benjamin - (cherry picked from commit 7bc144ce36fedc16a3dedc54598b0d75fb8c68bc) - -commit cc820a0d4c2676799383aeaf49a9269e104853ce -Merge: 5ee54cfe8f d079b91479 -Author: Nathan Cutler -Date: Tue Apr 25 17:38:21 2017 +0200 - - Merge pull request #14605 from asheplyakov/19476-jewel - - jewel: rgw: don't return skew time in pre-signed url - - Reviewed-by: Casey Bodley - -commit 5ee54cfe8f0a955b11edb4819a652ee420f43e5d -Merge: 37254aee6c c05bd1cb1f -Author: Nathan Cutler -Date: Tue Apr 25 17:37:16 2017 +0200 - - Merge pull request #14660 from smithfarm/wip-19478-jewel - - jewel: rgw: zonegroupmap set does not work - - Reviewed-by: Casey Bodley - -commit 37254aee6c3f725cf60deb5c7a3fc28405c51216 -Merge: 09919f938f faeb8088ac -Author: Nathan Cutler -Date: Tue Apr 25 17:36:23 2017 +0200 - - Merge pull request #14607 from asheplyakov/19607-jewel - - jewel: rgw: multisite: fetch_remote_obj() gets wrong version when copying from remote - - Reviewed-by: Casey Bodley - -commit 09919f938f3c1e9e81963f676a9dd13114d54e73 -Merge: d5e1345a2a 527911fab7 -Author: Nathan Cutler -Date: Tue Apr 25 15:50:26 2017 +0200 - - Merge pull request #14587 from asheplyakov/19617-jewel - - jewel: mon/MonClient: make get_mon_log_message() atomic - - Reviewed-by: Sage Weil - -commit 6c4826606dc81fdd847959a49454c69a958bb1d8 -Author: Kefu Chai -Date: Mon Dec 5 20:23:21 2016 +0800 - - test/ceph_crypto: do not read ceph.conf in global_init() - - ForkDeathTest.MD5 expect an empty output while global_init() complains - if ceph.conf is missing if 0 is passed in as the `flags`. this test - passes if ceph.conf is in current working directory, but jenkins does - not prepare this file for this test. - - Fixes: http://tracker.ceph.com/issues/18128 - Signed-off-by: Kefu Chai - (cherry picked from commit c72a2271a8012a66d7bbccf5766a73da5bb878d6) - - Conflicts: - src/test/ceph_crypto.cc - jewel does not have - 5af29540675b674c1985ff98b28a783ed124acf6 - -commit aca2659952528aa1b40b02828293342128657194 -Author: Loic Dachary -Date: Thu Dec 8 12:40:42 2016 +0100 - - tests: fix erasure-code premature deallocation of cct - - The setup function returns before the run function, the cct variable - must be a data member, not a local variable that gets de-allocated - before run() starts. - - Signed-off-by: Loic Dachary - (cherry picked from commit efa1e54362423d4cfd1541fb8c68237b7b9ebbe3) - -commit 8bddd427d9a4bca75a352ca333847dbd6d2c369c -Author: Pan Liu -Date: Thu Feb 16 22:17:52 2017 +0800 - - rbd-nbd: no need create asok file for unmap and list-mapped commands. - - Fixes: http://tracker.ceph.com/issues/17951 - Signed-off-by: Pan Liu - (cherry picked from commit 72352653d585ef89043a4ece371b5c0cb3f6f32a) - -commit 328bfbd25c63b1b0e253865abedada7a9e5858e4 -Author: Jason Dillaman -Date: Tue Nov 29 12:36:00 2016 -0500 - - rbd-nbd: restart parent process logger after forking - - Fixes: http://tracker.ceph.com/issues/18070 - Signed-off-by: Jason Dillaman - (cherry picked from commit 29baf254d72cc593572b5a6215360ba51e3be198) - -commit 192e7bcdcd1bc02b40b29eff86a335cc8919f663 -Author: Kefu Chai -Date: Wed Nov 23 19:45:57 2016 +0800 - - crushtool: do not release g_ceph_context at exit - - it is but a work around of occasionally timeout. - - Signed-off-by: Kefu Chai - (cherry picked from commit d305cc51b18cbf4b2757bbacb5d43324461306a9) - -commit 8a2f27cc632c26d7c2b8e8528b4d459b1d78705b -Author: Kefu Chai -Date: Tue Nov 15 14:21:03 2016 +0800 - - common,test: g_ceph_context->put() upon return - - prior to this change, global_init() could create a new CephContext - and assign it to g_ceph_context. it's our responsibilty to release - the CephContext explicitly using cct->put() before the application - quits. but sometimes, we fail to do so. - - in this change, global_init() will return an intrusive_ptr, - which calls `g_ceph_context->put()` in its dtor. this ensures that - the CephContext is always destroyed before main() returns. so the - log is flushed before _log_exp_length is destroyed. - - there are two cases where global_pre_init() is called directly. - - ceph_conf.cc: g_ceph_context->put() will be called by an intrusive_ptr<> - deleter. - - rgw_main.cc: global_init() is called later on on the success code - path, so it will be taken care of. - - Fixes: http://tracker.ceph.com/issues/17762 - Signed-off-by: Kefu Chai - (cherry picked from commit cb1cda96713b2ec0f6418c4cbe3d964c2020729c) - - Conflicts: - src/ceph_fuse.cc - src/ceph_mgr.cc - src/global/global_init.cc - src/rgw/rgw_main.cc - src/test/compressor/test_compression.cc - src/test/compressor/test_compression_plugin.cc - src/test/compressor/test_compression_plugin_snappy.cc - src/test/compressor/test_compression_plugin_zlib.cc - src/test/compressor/test_compression_snappy.cc - src/test/compressor/test_compression_zlib.cc - src/test/erasure-code/TestErasureCode.cc - src/test/erasure-code/TestErasureCodeExample.cc - src/test/erasure-code/TestErasureCodeIsa.cc - src/test/erasure-code/TestErasureCodeJerasure.cc - src/test/erasure-code/TestErasureCodeLrc.cc - src/test/erasure-code/TestErasureCodePlugin.cc - src/test/erasure-code/TestErasureCodePluginIsa.cc - src/test/erasure-code/TestErasureCodePluginJerasure.cc - src/test/erasure-code/TestErasureCodePluginLrc.cc - src/test/erasure-code/TestErasureCodePluginShec.cc - src/test/erasure-code/TestErasureCodeShec.cc - src/test/erasure-code/TestErasureCodeShec_thread.cc - src/test/fio/fio_ceph_objectstore.cc - src/test/librados/misc.cc - src/test/mon/PGMap.cc - src/test/msgr/test_async_networkstack.cc - src/test/msgr/test_userspace_event.cc - src/test/objectstore/Allocator_test.cc - src/test/objectstore/BitAllocator_test.cc - src/test/objectstore/test_bluefs.cc - src/test/objectstore/test_bluestore_types.cc - src/test/objectstore/test_memstore_clone.cc - src/test/osd/TestPGLog.cc - src/test/rgw/test_http_manager.cc - src/test/rgw/test_rgw_compression.cc - src/test/test_mempool.cc - src/tools/rados/rados.cc - -commit d5e1345a2a6bd8456417db6bb60d61871165e6bd -Merge: 256f48f463 5096fc9c7c -Author: Nathan Cutler -Date: Tue Apr 25 15:42:43 2017 +0200 - - Merge pull request #14665 from smithfarm/wip-19610-jewel - - jewel: [librados_test_stub] cls_cxx_map_get_XYZ methods don't return correct value - - Reviewed-by: Jason Dillaman - -commit 256f48f46354f11a2f238f24fa9d890b55f4f4fc -Merge: 4f67da1b9d 21a83e1276 -Author: Nathan Cutler -Date: Tue Apr 25 09:20:19 2017 +0200 - - Merge pull request #14653 from smithfarm/wip-19662-jewel - - jewel: rgw_file: fix event expire check, don't expire directories being read - - Reviewed-by: Matt Benjamin - -commit 4f67da1b9d554a1a5371665573b1f56d3f903de3 -Merge: c2452c5364 33af18e592 -Author: Loic Dachary -Date: Tue Apr 25 09:17:33 2017 +0200 - - Merge pull request #14635 from smithfarm/wip-19690-jewel - - jewel: doc: Improvements to crushtool manpage - - Reviewed-by: Loic Dachary - -commit 040ff013d5c8aded8beedc59bbc4f5afccc64e46 -Author: Kefu Chai -Date: Wed Nov 16 11:56:09 2016 +0800 - - crushtool: s/exit(EXIT_FAILURE)/return EXIT_FAILURE/ - - so the destructor(s) can be called. - - Signed-off-by: Kefu Chai - (cherry picked from commit e01b89ed6be6b99fec5c725f4bc5769b42468cac) - - Conflicts: - src/tools/crushtool.cc - jewel does not have 17feefbcb3105553b763cb7ce123b20b77b95857 - -commit 8e993e6e43ecd6e2b444961d1e6c945081b64207 -Author: Kefu Chai -Date: Wed Nov 16 11:19:04 2016 +0800 - - global/signal_handler: reset injected segv after test - - ~CephContext() => ~TypedSingletonWrapper() => ~MempoolObs => - unregister_command() => ldout() << "unregister_command" => - Log::submit_entry() => *(volatile int *)(0) = 0xdead; - - Signed-off-by: Kefu Chai - (cherry picked from commit d932c8f2f23263924103a900714db82ee87f6eef) - - Conflicts: - src/log/Log.cc - jewel has "namespace log" instead of "namespace - logging" (trivial resolution) - -commit f1c0042b831826e5bd72ada79a4918a9c26bda24 -Author: Kefu Chai -Date: Tue Nov 15 14:42:35 2016 +0800 - - test_cors.cc: fix the mem leak - - Signed-off-by: Kefu Chai - (cherry picked from commit c8a3777203482cabf6739a8ba69b127df8697628) - -commit c49b114e8dde6c3f3c5a5b663bd8299b4f2feafb -Author: weiqiaomiao -Date: Tue Sep 6 16:34:52 2016 +0800 - - rgw: fix failed to create bucket if a non-master zonegroup has a single zone - - If a non-master zonegroup has a single zone, the metadata sync thread not running and - the non-master zonegroup can't sync user from master zonegroup, - so we can't create bucket(or other metadata update) in it - because the authenticated user not found in the zone of non-master zonegroup. - - Signed-off-by: weiqiaomiao - (cherry picked from commit 949af79b21098e6410bc29274cf36eae2d89faea) - - Conflicts: - src/rgw/rgw_rados.cc - retain d32654b7cd60ccc4e23d3f05b9e4385a697bacd6 - which was merged after this commit - -commit 43327f83efcbc5ce54be866a30da8dd1e58d6707 -Author: Loic Dachary -Date: Wed Feb 22 01:49:12 2017 +0100 - - ceph-disk: dmcrypt activate must use the same cluster as prepare - - When dmcrypt is used, the fsid cannot be retrieved from the data - partition because it is encrypted. Store the fsid in the lockbox to - enable dmcrypt activation using the same logic as regular activation. - - The fsid is used to retrive the cluster name that was used during - prepare, reason why activation does not and must not have a --cluster - argument. - - Fixes: http://tracker.ceph.com/issues/17821 - - Signed-off-by: Loic Dachary - (cherry picked from commit 7f66672b675abbc0262769d32a38112c781fefac) - - Conflicts: - src/ceph-disk/ceph_disk/main.py - in master, self.create_key() takes an - argument (self.args.cluster) but in jewel it takes no argument - -commit c2452c53641fca416268c2e31c774b8b7e609c88 -Merge: 82b8c89e47 01d04e28db -Author: Nathan Cutler -Date: Tue Apr 25 07:20:41 2017 +0200 - - Merge pull request #13608 from smithfarm/wip-19063-jewel - - jewel: tests: eliminate race condition in Thrasher constructor - - Reviewed-by: Kefu Chai - Reviewed-by: David Zafman - -commit 3ec1a9bf16e2c305096e11223aaa8db94dc4084d -Author: Sage Weil -Date: Fri Mar 31 10:06:42 2017 -0400 - - ceph_test_librados_api_misc: fix stupid LibRadosMiscConnectFailure.ConnectFailure test - - Sometimes the cond doesn't time out and it wakes up instead. Just repeat - the test many times to ensure that at least once it times out (usually - it doesn't; it's pretty infrequent that it doesn't). - - Fixes: http://tracker.ceph.com/issues/15368 - Signed-off-by: Sage Weil - (cherry picked from commit 8bc197400d94ee2716d3f2fa454247379a676cf9) - -commit 82b8c89e477610641a6e21123b7c929e9c560729 -Merge: 28c7ce595a 30c9527353 -Author: Nathan Cutler -Date: Mon Apr 24 22:27:33 2017 +0200 - - Merge pull request #14195 from cbodley/wip-19353 - - jewel: rgw: use separate http_manager for read_sync_status - - Reviewed-by: Yehuda Sadeh - -commit 28c7ce595ab8fef7c0076f091f249f6b3548ad54 -Merge: c05ecff3f0 2e50fe1684 -Author: Nathan Cutler -Date: Mon Apr 24 22:26:50 2017 +0200 - - Merge pull request #14066 from asheplyakov/19321-bp-jewel - - jewel: rgw: fix break inside of yield in RGWFetchAllMetaCR - - Reviewed-by: Yehuda Sadeh - -commit c05ecff3f0cb5af2f3aa52b68188742796335ddb -Merge: 3240cbf4ef dc4e7a1a86 -Author: Nathan Cutler -Date: Mon Apr 24 22:26:14 2017 +0200 - - Merge pull request #14064 from asheplyakov/19211-bp-jewel - - jewel: rgw: "cluster [WRN] bad locator @X on object @X...." in cluster log - - Reviewed-by: Yehuda Sadeh - -commit 3240cbf4ef60733f55800039051e85f2a7ace61d -Merge: d4672acbb8 85fbb00f6e -Author: Nathan Cutler -Date: Mon Apr 24 22:25:21 2017 +0200 - - Merge pull request #13842 from smithfarm/wip-19145-jewel - - jewel: rgw: a few cases where rgw_obj is incorrectly initialized - - Reviewed-by: Yehuda Sadeh - -commit d4672acbb82daf81eaf259e659dd627ec9f3bc79 -Merge: 61ed719d75 ec0668c201 -Author: Nathan Cutler -Date: Mon Apr 24 22:24:29 2017 +0200 - - Merge pull request #13837 from smithfarm/wip-19048-jewel - - jewel: rgw: multisite: some yields in RGWMetaSyncShardCR::full_sync() resume in incremental_sync() - - Reviewed-by: Yehuda Sadeh - -commit 61ed719d75bf3606fa9e3ae8c8ed776aa2ee4313 -Merge: f5e51db564 ced799f9c6 -Author: Nathan Cutler -Date: Mon Apr 24 22:23:24 2017 +0200 - - Merge pull request #13724 from asheplyakov/18626-bp-jewel - - jewel: rgw: Use decoded URI when verifying TempURL - - Reviewed-by: Yehuda Sadeh - -commit 66c3db7aee5b53f83e87ee8c8f081ab9b0336177 -Author: Casey Bodley -Date: Wed Mar 8 16:31:34 2017 -0500 - - rgw: data sync skips slo data when syncing the manifest object - - Fixes: http://tracker.ceph.com/issues/19027 - - Signed-off-by: Casey Bodley - (cherry picked from commit 8b69847d7b3e92c70090d1dddf7cea5c44fb6b20) - - Conflicts: bucket cleanup, overrides - -commit 303a62f7f15c69413165ed604869909587714a94 -Author: Casey Bodley -Date: Mon Mar 13 11:33:02 2017 -0400 - - rgw: RGWGetObj applies skip_manifest flag to SLO - - Signed-off-by: Casey Bodley - (cherry picked from commit 987377ae34382e107e1d54f0bfc1121fcedb4513) - -commit f3d99ae8aafaa65e91cd233b5ce8054678d8fa11 -Author: Casey Bodley -Date: Tue Feb 21 10:27:13 2017 -0500 - - rgw: allow system users to read SLO parts - - multisite data sync relies on fetching the object as the system user - - Fixes: http://tracker.ceph.com/issues/19027 - - Signed-off-by: Casey Bodley - (cherry picked from commit d50d18c500fd5dd89e7cada1162cf453b36df370) - - Conflicts: auth rework - -commit f5e51db5644d4eafdbe3b2d541582a17febed139 -Merge: c90cfb7a32 eac0e27193 -Author: Nathan Cutler -Date: Fri Apr 21 22:51:59 2017 +0200 - - Merge pull request #14643 from smithfarm/wip-revert-14427 - - Wip revert 14427 - - Reviewed-by: Josh Durgin - -commit db053da61807e26876d97786550a769295c7955a -Author: John Spray -Date: Mon Mar 27 12:56:31 2017 +0100 - - mds: validate prealloc_inos on sessions after load - - Mitigates http://tracker.ceph.com/issues/16842 - - Signed-off-by: John Spray - (cherry picked from commit c39aaf90ed1b23343eba2b341bb8ee6a50a4ea74) - - Conflicts: - src/mds/InoTable.cc - no 5259683e7819c22c14b21b1dd678a33e14574f21 in jewel - src/mds/InoTable.h - no 5259683e7819c22c14b21b1dd678a33e14574f21 in jewel - -commit 2b5eb8fa141fa8bd9173dee206ec3530d702fc3a -Author: John Spray -Date: Mon Mar 27 12:33:59 2017 +0100 - - mds: operator<< for Session - - Use this to get a nice human readable name - when available (also including the session id in - parentheses) - - Signed-off-by: John Spray - (cherry picked from commit 0f89787d8312f132ebb621f16c44e950b17a395a) - -commit 5b562145197833c6a6f48f006dfb13b6e38f9084 -Author: Henrik Korkuc -Date: Sun Feb 19 11:44:20 2017 +0200 - - client/Client.cc: add feature to reconnect client after MDS reset - - Client.cc marks session as stale instead of reconecting after received - reset from MDS. On MDS side session is closed so MDS is ignoring cap - renew. This adds option to reconnect stale client sessions instead of - just marking sessions stale. - - Fixes: http://tracker.ceph.com/issues/18757 - - Signed-off-by: Henrik Korkuc - (cherry picked from commit e0bbc704676ef4aed510daff075ef63c9e73b7b3) - -commit 8f21038d30097622e319ab986631b03f87f5d907 -Author: Kefu Chai -Date: Tue Mar 21 12:49:45 2017 +0800 - - doc: cephfs: fix the unexpected indent warning - - Signed-off-by: Kefu Chai - (cherry picked from commit e423f0b59711422b40c4b3de0bdc73b0947c04d3) - -commit f9a1954e113e807b5f9ee6fd56351adc2841730a -Author: Barbora Ančincová -Date: Thu Feb 16 10:45:36 2017 +0100 - - doc: additional edits in FUSE client config - - Signed-off-by: Bara Ancincova (bara@redhat.com) - (cherry picked from commit b6cad3364c020abd3acf906643fa0b6cbb862a0a) - -commit 018649f8a4628881c706a6df9b3eef48403fe91e -Author: Barbora Ančincová -Date: Thu Jan 26 12:23:34 2017 +0100 - - doc: Dirty data are not the same as corrupted data - - Signed-off-by: Bara Ancincova (bara@redhat.com) - (cherry picked from commit 80db40f8559128baadad42b925ae813e51a31409) - -commit 1d8a5b6d64a26a36fb3a227c52908631d95bef79 -Author: Barbora Ančincová -Date: Mon Jan 23 16:34:55 2017 +0100 - - doc: minor changes in fuse client config reference - - Signed-off-by: Bara Ancincova (bara@redhat.com) - (cherry picked from commit e57605681f10436f4b2c85e95179a2904b8c80da) - -commit 1ae46b2b94280b8b4a1db89b2cbed8f12bcbcc53 -Author: Patrick Donnelly -Date: Sun Jul 24 23:21:29 2016 -0400 - - doc: add client config ref - - Fixes: http://tracker.ceph.com/issues/16743 - - Signed-off-by: Patrick Donnelly - (cherry picked from commit 9ad2ccf29830d5309336fc7de877b6926e5dbacd) - -commit b8fd297eb64cd29c58596d0a5b0be8fbe9d94a72 -Author: Vasu Kulkarni -Date: Tue Apr 11 13:51:47 2017 -0700 - - use sudo to check check health - - Signed-off-by: Vasu Kulkarni - (cherry picked from commit 7af157ad4ce7f7e2b8de97ee10eeaf64b9099bc0) - -commit 1b91ffc0fbe76c5475f17d4e15ea295ee3680688 -Author: Vasu Kulkarni -Date: Wed Mar 29 09:27:20 2017 -0700 - - Add reboot case for systemd test - - test systemd units restart after reboot - - Signed-off-by: Vasu Kulkarni - (cherry picked from commit 7b587304a54d9b21041ffdfbc85fad8d87859c49) - -commit 3d8d1dad8a698a9fb45be7b31c92f8b71ee67720 -Author: Vasu Kulkarni -Date: Wed Mar 29 09:56:11 2017 -0700 - - Fix distro's, point to latest version - - Signed-off-by: Vasu Kulkarni - (cherry picked from commit 1947648669971c1bd1ca189870ed9b25bbd48d3a) - -commit c90cfb7a327fcbb9508e617c2353becb7e5cb45f -Merge: 327276cf3d 9b77b16b88 -Author: Nathan Cutler -Date: Thu Apr 20 22:05:57 2017 +0200 - - Merge pull request #14602 from asheplyakov/19646-jewel - - jewel: ceph-disk: enable directory backed OSD at boot time - - Reviewed-by: Loic Dachary - -commit 327276cf3d4fb7a2558620a8a7d9cc90e5d6e5c5 -Merge: 7008c64c51 25e43ac256 -Author: Nathan Cutler -Date: Thu Apr 20 22:04:41 2017 +0200 - - Merge pull request #14449 from smithfarm/wip-test-doc-oversight - - tests: fix oversight in yaml comment - - Reviewed-by: Loic Dachary - -commit f34489dd52f07aeb88e1dbf361ead63f09bb9c65 -Author: YunfeiGuan -Date: Mon Apr 10 05:48:47 2017 +0000 - - cephfs: fix mount point break off problem after mds switch occured - - The hot-standby become active as we expected but the mount piont broken strangely - when the active mds is down. The root reason is the new mds use last_cap_renews - decoded from ESesson::replay in find_idle_sessions and wrongly killed the session. - Maybe we should reset session->last_cap_renew to the current time when server send - OPEN to client in reconnect stage. - - Fixes: http://tracker.ceph.com/issues/19437 - Signed-off-by: Guan yunfei - (cherry picked from commit 4ef830c5d6f22bf0d4f82a8624c772ecbbda44a6) - - Conflicts: - src/mds/Server.cc (leave '\n' in because jewel does not have - 693132eb00b1803d5e97a79908521d5a6903e9f8; jewel ceph_clock_now takes a - CephContext object) - -commit 7008c64c5143a3ea19d6679a3e521897f74ff69a -Merge: d62644abcd f8aa6be06c -Author: Nathan Cutler -Date: Thu Apr 20 19:26:52 2017 +0200 - - Merge pull request #13606 from smithfarm/wip-19062-jewel - - jewel: build/ops: enable build of ceph-resource-agents package on rpm-based os - - Reviewed-by: Ken Dreyer - -commit d62644abcde0fd99da24922ce7638d173952ceb4 -Merge: 8df8960553 482bd1adab -Author: Nathan Cutler -Date: Thu Apr 20 18:18:05 2017 +0200 - - Merge pull request #14654 from smithfarm/wip-19461-jewel - - jewel: admin ops: fix the quota section - - Reviewed-by: Casey Bodley - -commit c2efeb4b62e6dc4e6975561eb8b20cfca66f7237 -Author: David Zafman -Date: Mon Apr 17 14:58:02 2017 -0700 - - osd: Give requested scrub work a higher priority - - Once started we now queue scrub work at higher priority than - scheduled scrubs. - - Fixes: http://tracker.ceph.com/issues/15789 - - Signed-off-by: David Zafman - (cherry picked from commit ebab8b1f4f67fbdec1e147c580329c1e2b5cf7cd) - - Conflicts: - src/osd/OSD.h - in jewel, the PGScrub() call is enclosed within - op_wq.queue(make_pair(...)) instead of enqueue_back() - -commit db86a24e7906ca7c70c60b1752f1230d56361bcd -Author: Yan, Zheng -Date: Wed Feb 22 17:33:05 2017 +0800 - - client: wait for lastest osdmap when handling set file/dir layout - - Fixes: http://tracker.ceph.com/issues/18914 - Signed-off-by: "Yan, Zheng" - (cherry picked from commit 76f5eb86cdd61dde4e6c7cfeb5cf34f0c0334f21) - - Conflicts: - src/client/Client.cc (jewel does not have 201c56039) - src/client/Client.h (jewel does not have 201c56039) - -commit 7b9283beec0c4bf8e2067558fc4e03e336849b77 -Author: Yang Honggang -Date: Thu Apr 13 20:09:07 2017 +0800 - - cephfs: fix write_buf's _len overflow problem - - After I have set about 400 64KB xattr kv pair to a file, - mds is crashed. Every time I try to start mds, it will crash again. - The root reason is write_buf._len overflowed when doing - Journaler::append_entry(). - - This patch try to fix this problem through the following changes: - - 1. limit file/dir's xattr size - 2. throttle journal entry append operations - - Fixes: http://tracker.ceph.com/issues/19033 - Signed-off-by: Yang Honggang joseph.yang@xtaotech.com - (cherry picked from commit eb915d0eeccbe523f8f70f6571880003ff459459) - -commit b52c5088618011a569f157616c5c667c2fc1e9fe -Author: John Spray -Date: Wed Mar 8 12:13:46 2017 +0000 - - mds: shut down finisher before objecter - - Some of the finisher contexts would try to call into Objecter. - We mostly are protected from this by mds_lock+the stopping - flag, but at the Filer level there's no mds_lock, so in the - case of file size probing we have a problem. - - Fixes: http://tracker.ceph.com/issues/19204 - Signed-off-by: John Spray - (cherry picked from commit 177a97d5c55ee6a2d5dcd3cf0893546190b10f7a) - - Conflicts: - src/mds/MDSRank.cc (no purge_queue.shutdown() in jewel because jewel - does not have 8ebf7d95a9071de24bb1e56a6423c505169cb4de) - -commit 96e801fb53941214c669ac01c1804a4d1f27086c -Author: John Spray -Date: Tue Mar 28 14:13:33 2017 -0400 - - mds: ignore ENOENT on writing backtrace - - We get ENOENT when a pool doesn't exist. This can - happen because we don't prevent people deleting - former cephfs data pools whose files may not have - had their metadata flushed yet. - - http://tracker.ceph.com/issues/19401 - Signed-off-by: John Spray - (cherry picked from commit 3fccc2372f2715d075b05e459140360cf6e7ca96) - -commit 8df896055366d777ca309dd4a7840f2a2a5a72cc -Merge: da888fa871 7468689314 -Author: Nathan Cutler -Date: Thu Apr 20 14:53:30 2017 +0200 - - Merge pull request #14680 from smithfarm/wip-19711-jewel - - jewel: [test] test_notify.py: rbd.InvalidArgument: error updating features for image test_notify_clone2 - - Reviewed-by: Jason Dillaman - -commit 74686893147f2d7e1fd1287de37587554fea96c6 -Author: Jason Dillaman -Date: Wed Apr 19 09:26:31 2017 -0400 - - test: rbd master/slave notify test should test active features - - Fixes: http://tracker.ceph.com/issues/19692 - Signed-off-by: Jason Dillaman - (cherry picked from commit 0dcba41cba96566d0b8da54cf0316d523b88ded2) - -commit 7347f1193963666ce07ff9163064aeb0cf343d38 -Author: Yan, Zheng -Date: Wed Apr 5 21:29:10 2017 +0800 - - mds: make C_MDSInternalNoop::complete() delete 'this' - - Fixes: http://tracker.ceph.com/issues/19501 - Signed-off-by: "Yan, Zheng" - (cherry picked from commit 424e0c6744e7f63459ca0ff7deab751726aa30cd) - - Conflicts: - src/mds/MDSContext.h (omit "override" because jewel does not have - 1a91aeab987870b3ccbcf2f1e476fac8b534d449) - -commit 824b19a9a66261a90c4e4de64e877a42438febd2 -Author: John Spray -Date: Wed Mar 29 19:38:37 2017 +0100 - - tools/cephfs: set dir_layout when injecting inodes - - When we left this as zero, the MDS would interpret it was HASH_LINUX - rather than the default HASH_RJENKINS. Potentially that - could cause problems if there perhaps were already dirfrags in - the metadata pool that were set up using rjenkins. Mainly - it just seems more appropriate to explicitly set this field - rather than hit the fallback behaviour. - - Related: http://tracker.ceph.com/issues/19406 - Signed-off-by: John Spray - (cherry picked from commit 7d6d542885bd29b71214f9ca52bd26e9183c5d01) - -commit eab56dae6714706cf9650dff172b20a764f5481c -Author: John Spray -Date: Thu Mar 9 13:15:46 2017 +0000 - - mon: fix hiding mdsmonitor informative strings - - Local `stringstream ss` declarations were hiding - the real variable used to feed back to the user. - - Fixes: http://tracker.ceph.com/issues/16709 - Signed-off-by: John Spray - (cherry picked from commit 00404ae9bd4cce0518a44d36d2d6a5612f4f9d04) - -commit da888fa8713178cc23c1069ce2651e02c98aad05 -Merge: 8068b546af 2271cd8128 -Author: Kefu Chai -Date: Thu Apr 20 19:10:33 2017 +0800 - - Merge pull request #14402 from shinobu-x/wip-17331-jewel - - jewel: ceph-disk: ceph-disk list reports mount error for OSD having mount options with SELinux context - - Reviewed-by: Loic Dachary - Reviewed-by: Brad Hubbard - Reviewed-by: Kefu Chai - -commit 8068b546afbe8e774c9d699205a32215e38c5d4f -Merge: 118ccad0ba 754b4a482c -Author: Nathan Cutler -Date: Thu Apr 20 13:06:33 2017 +0200 - - Merge pull request #13865 from smithfarm/wip-19158-jewel - - jewel: rgw: health check errors out incorrectly - - Reviewed-by: Radoslaw Zarzynski - -commit d57437e338984c5db84f9d16387e082ada5a0a33 -Author: John Spray -Date: Mon Mar 6 11:51:31 2017 +0000 - - mds: reset heartbeat in export_remaining_imported_caps - - This loop can be very long. - - Fixes: http://tracker.ceph.com/issues/19118 - Signed-off-by: John Spray - (cherry picked from commit 85071f1509beba4a390730e6a3a4332484646d63) - -commit 6adf1904ed7209b70328b20f248a701fbdd3127c -Author: John Spray -Date: Mon Mar 6 11:24:50 2017 +0000 - - mds: heartbeat_reset in dispatch - - Previously we only heartbeated in tick. However, our locking is - not guaranteed to be fair, so on a super-busy dispatch queue it may be - possible for the heartbeat to time out while the tick() function - is waiting for mds_lock. - - Fixes: http://tracker.ceph.com/issues/19118 - Signed-off-by: John Spray - (cherry picked from commit 819394549af10532419d88742fae3a69d2ea487d) - -commit 63f41d543f8a5f1f55a12612d39c6a2a1cf9c114 -Author: Yan, Zheng -Date: Wed Feb 15 11:45:26 2017 +0800 - - test/libcephfs: avoid buffer overflow when testing ceph_getdents() - - The buffer size should be at least "2 * sizeof(struct dirent)". - Otherwise, the code that checks dentry '..' overflow. - - Fixes: http://tracker.ceph.com/issues/18941 - Signed-off-by: "Yan, Zheng" - (cherry picked from commit fa6671345b8f3a82dcd232f99e55a982b0a641f1) - -commit 7146816065bea55b4e3fec59048a459bcff50f55 -Author: Zhi Zhang -Date: Fri Feb 10 10:56:46 2017 +0800 - - mds/StrayManager: aviod reusing deleted inode in StrayManager::_purge_stray_logged - - Signed-off-by: Zhi Zhang - (cherry picked from commit 4978e57419482384279d7e784a625f5e5c10961a) - -commit d8b139b5847cb46aa13486ee85e26f26421f36d6 -Author: John Spray -Date: Wed Feb 1 00:38:08 2017 +0000 - - tasks/cephfs: switch open vs. write in test_open_inode - - Do the write after opening the file, so that we get good - behaviour wrt the change in Mount.open_background that uses - file existence to confirm that the open happened. - - Signed-off-by: John Spray - (cherry picked from commit a027dba78fc8bc84ae39d7998b386ce21c01e1bf) - -commit e8ae80fc8721e500782c7e87a7cb89128c2a9687 -Author: John Spray -Date: Thu Jan 26 16:48:58 2017 +0000 - - qa: fix race in Mount.open_background - - Previously a later remote call could end up executing - before the remote python program in open_background - had actually got as far as opening the file. - - Fixes: http://tracker.ceph.com/issues/18661 - Signed-off-by: John Spray - (cherry picked from commit c6d91dd91252e703d08b8ac62ac6a47ee82c0bed) - -commit 36c86f71efc10f9a651aedf01dc6fb9d1bfed703 -Author: Yan, Zheng -Date: Wed Jan 25 15:28:23 2017 +0800 - - mds: don't purge strays when mds is in clientreplay state - - MDS does not trim log when it's in clientreplay state. If mds hang - at clientreplay state (due to bug), purging strays can submit lots - of log events and create very large mds log. - - Signed-off-by: "Yan, Zheng" - (cherry picked from commit 86bbc7fff02668077f27d0924ba3efe6544b77f6) - -commit 8b01cf33575783661a1e1151c8214d327a08b6f9 -Author: Yan, Zheng -Date: Wed Jan 25 11:03:45 2017 +0800 - - mds: skip fragment space check for replayed request - - when handling replayed request, stray directory can be different - from the stray directory used by the original request. The fragment - space check for stray directory can fail. - - Fixes: http://tracker.ceph.com/issues/18660 - Signed-off-by: "Yan, Zheng" - (cherry picked from commit afe889cbc5baab196567c2aad01f49fe90901fda) - -commit 82ea0971b3cb07c32ec837cb85de63d4068a70d7 -Author: Greg Farnum -Date: Wed Dec 14 12:09:44 2016 -0800 - - client: fix the cross-quota rename boundary check conditions - - We were previously rejecting a rename if either of the involved directories - was a quota root, even if the other directory was part of the same quota - "tree". What we really want to do is identify the correct quota root - (whether local or ancestral) for each directory and compare them. So - now we do. - - Signed-off-by: Greg Farnum - (cherry picked from commit 8e8892aa46accb519faa4bb9fecf66618f1b11b2) - - Conflicts: - src/client/Client.cc (do not pass perm to get_quota_root() because - jewel does not have 3caa4d233633fb7a67747f2c79c4a0ab89112294) - -commit dbe90c79b86743c7d143d33eb5389fcee0ac76ef -Author: Mykola Golub -Date: Tue Apr 11 22:31:43 2017 +0200 - - librbd: fix rbd_metadata_list and rbd_metadata_get - - - properly check for val_len in rbd_metadata_list - - don't expect output buffers are zero pre-filled - - Fixes: http://tracker.ceph.com/issues/19588 - Signed-off-by: Mykola Golub - (cherry picked from commit 75afc74ea681402e22b6dec8b83276d145fc786b) - -commit 5096fc9c7c62e3043a9a0638eb2516792526fdd7 -Author: Jason Dillaman -Date: Wed Apr 12 10:47:28 2017 -0400 - - test/librados_test_stub: fixed cls_cxx_map_get_keys/vals return value - - Fixes: http://tracker.ceph.com/issues/19597 - Signed-off-by: Jason Dillaman - (cherry picked from commit 9ffd464dac102f684d6dfa78e58d2cb45e165ed6) - -commit 216156b5d4e2666ca592eaab2211940028422bc0 -Author: Jason Dillaman -Date: Fri Mar 10 10:56:38 2017 -0500 - - rbd: prevent adding multiple mirror peers to a single pool - - The rbd-mirror daemon does not currently support replication - from multiple peers. Until that is supported, add a temporary - restriction to prevent confusion. - - Fixes: http://tracker.ceph.com/issues/19256 - Signed-off-by: Jason Dillaman - (cherry picked from commit c0c9d1014d57b3d5b95e7513fcc38d04b9ea5165) - -commit c05bd1cb1f570c42b1999fbc70c86b0a72a7fcb3 -Author: Orit Wasserman -Date: Wed Apr 5 13:31:08 2017 +0300 - - radosgw-admin: use zone id when creating a zone - - Fixes: http://tracker.ceph.com/issues/19498 - Signed-off-by: Orit Wasserman - (cherry picked from commit 3fea36d635fcba8ca584a1c0ec9f07840009402c) - -commit ba81cbbdee1ccf95ceff56eef0a1b2b06be4024a -Author: Casey Bodley -Date: Tue Mar 14 15:43:13 2017 -0400 - - qa: rgw task uses period instead of region-map - - Signed-off-by: Casey Bodley - (cherry picked from commit e3e3a71d1f1fb43bb4172ce2dfac9a28ca89df0f) - -commit a755c95262e10fac774716e460e2ba2eaee9df70 -Author: Casey Bodley -Date: Tue Mar 14 14:18:15 2017 -0400 - - rgw-admin: remove deprecated regionmap commands - - Fixes: http://tracker.ceph.com/issues/18725 - - Signed-off-by: Casey Bodley - (cherry picked from commit 5830c1849a0c0110d17c37784808e456e6dcb7b3) - - Conflicts: - src/rgw/rgw_admin.cc (trivial resolution) - -commit 482bd1adabf6f54c8f2bd73090d2e4c02d07f8a3 -Author: hrchu -Date: Wed Mar 29 02:17:04 2017 +0000 - - doc: rgw: correct the quota section - - Add the missing option and fix typo. - - Fixes: http://tracker.ceph.com/issues/19397 - - Signed-off-by: Chu, Hua-Rong - (cherry picked from commit 51a88267f0d7f51aeb62092949b66b9f6c062e15) - -commit 21a83e1276e415e98a3780d0374bb9d4feb191d9 -Author: Matt Benjamin -Date: Wed Mar 15 16:35:16 2017 -0400 - - rgw_file: remove unused rgw_key variable - - Signed-off-by: Matt Benjamin - (cherry picked from commit 1100a1c26e76485569cfebcf863b18cf908f6161) - -commit ebad040b3ab62e702afc52edd9d99d984b24cdc1 -Author: Matt Benjamin -Date: Fri Apr 14 15:56:37 2017 -0400 - - rgw_file: fix readdir after dirent-change - - Also, fixes link count computation off-by-one, update of state.nlink - after computation, link computation reset at start, and a time print - in debug log. - - Fixes: http://tracker.ceph.com/issues/19634 - - Signed-off-by: Matt Benjamin - - link count - - Signed-off-by: Matt Benjamin - (cherry picked from commit e0f80266ecd424bf9466579b3edc03911a7c5719) - -commit dd9833cacaec20e5bd1a70ec46a427a7352eb5d6 -Author: Matt Benjamin -Date: Tue Apr 11 06:42:07 2017 -0400 - - rgw_file: don't expire directories being read - - If a readdir expire event turns out to be older than last_readdir, - just reschedule it (but actually, we should just discard it, as - another expire event must be in queue. - - Fixes: http://tracker.ceph.com/issues/19625 - - Signed-off-by: Matt Benjamin - (cherry picked from commit 007b7451c26716c51207c161dc347e9a00da53f1) - -commit 30a5e857aade2474c5bd621c57938dfe333c3ae5 -Author: Matt Benjamin -Date: Wed Mar 15 16:40:35 2017 -0400 - - rgw_file: rgw_readdir: return dot-dirs only when *offset is 0 - - Signed-off-by: Matt Benjamin - (cherry picked from commit 61482c2b85a07519f2256b1a3f2b6d8aa99d5f06) - -commit fe836bfb7a286c87c57576d29d6862be2514ada5 -Author: Matt Benjamin -Date: Tue Apr 11 05:56:13 2017 -0400 - - rgw_file: chunked readdir - - Adjust readdir callback path for new nfs-ganesha chunked readdir, - including changes to respect the result of callback to not - continue. - - Pending introduction of offset name hint, our caller will just be - completely enumerating, so it is possible to remove the offset map - and just keep a last offset. - - Fixes: http://tracker.ceph.com/issues/19624 - - Signed-off-by: Matt Benjamin - (cherry picked from commit e0191d74e3aef06bf300df045a53a3952a71f651) - -commit 16eeb8c2acc8a1f843fa87967c2ee4c0ed2cd8bd -Author: Gui Hecheng -Date: Fri Mar 31 10:42:40 2017 +0800 - - rgw_file: fix missing unlock in unlink - - Fixes: http://tracker.ceph.com/issues/19435 - - Signed-off-by: Gui Hecheng - (cherry picked from commit cb6808a6366a70f54d0cc16437d16aa1b7819c84) - -commit 8c7cb8227c39b09060cdca4e11dec75cf7f2336c -Author: Matt Benjamin -Date: Mon Mar 13 21:52:08 2017 -0400 - - rgw_file: implement reliable has-children check (unlink dir) - - Bug report and discussion provided by - Gui Hecheng in nfs-ganesha upstream - github. Briefly, while a reliable check is potentially costly, - it is necessary. - - Fixes: http://tracker.ceph.com/issues/19270 - - Signed-off-by: Matt Benjamin - (cherry picked from commit b05f1c6d61aa4501a971e87de6dcaf3e58c3d9b4) - -commit 18f14dd86400b50f46930a9ef56666d82035507e -Author: Matt Benjamin -Date: Tue Apr 4 20:16:13 2017 -0400 - - rgw_file: introduce rgw_lookup type hints - - The new type hints optimize object type deduction, when the - rgw_lookup is called from an rgw_readdir callback. - - Fixes: http://tracker.ceph.com/issues/19623 - - Signed-off-by: Matt Benjamin - (cherry picked from commit 2e66c7a7cc763c5c0d6f5db04855f60f2b2ceed3) - -commit 118ccad0ba763b98c9ee3de225a947be5372dc0d -Merge: bf30ecd7cb 65465356b5 -Author: Nathan Cutler -Date: Thu Apr 20 11:08:45 2017 +0200 - - Merge pull request #14383 from smithfarm/wip-19547-jewel - - jewel: build/ops: rbdmap.service not included in debian packaging (jewel-only) - - Reviewed-by: Ken Dreyer - -commit bf30ecd7cb8dea43a9d55cb1b02c72bfda6bad09 -Merge: 376c5e4753 9e123e6d6c -Author: Nathan Cutler -Date: Thu Apr 20 11:02:39 2017 +0200 - - Merge pull request #14143 from smithfarm/wip-19355-jewel - - jewel: rgw: when converting region_map we need to use rgw_zone_root_pool - - Reviewed-by: Casey Bodley - -commit 376c5e47531a2803ac9b7ea236b989c528268a9f -Merge: da306df4de e2ee70a8ad -Author: Nathan Cutler -Date: Thu Apr 20 11:01:49 2017 +0200 - - Merge pull request #14136 from smithfarm/wip-19330-jewel - - jewel: rgw: upgrade to multisite v2 fails if there is a zone without zone info - - Reviewed-by: Casey Bodley - -commit da306df4deaaa782c0bad72539affd3ee1ec7efc -Merge: 57b210da41 5ee8feaba4 -Author: Nathan Cutler -Date: Thu Apr 20 11:00:35 2017 +0200 - - Merge pull request #13872 from smithfarm/wip-19163-jewel - - jewel: doc: radosgw-admin: add the 'object stat' command to usage - - Reviewed-by: Casey Bodley - -commit 57b210da4185d624524d05bcd9ad01c2df16ca76 -Merge: db92019e9a 9cd7dd8490 -Author: Nathan Cutler -Date: Thu Apr 20 10:59:33 2017 +0200 - - Merge pull request #13863 from smithfarm/wip-19155-jewel - - jewel: rgw: typo in rgw_admin.cc - - Reviewed-by: Casey Bodley - -commit db92019e9af43c2367fdf7865c0fab8cb38453e6 -Merge: 6dcd5fa847 6add2a457e -Author: Nathan Cutler -Date: Thu Apr 20 10:56:35 2017 +0200 - - Merge pull request #13779 from smithfarm/wip-18866-jewel - - jewel: rgw: 'radosgw-admin sync status' on master zone of non-master zonegroup - - Reviewed-by: Casey Bodley - -commit eac0e27193c67ffca60b5e7c61e7769ad8ace6aa -Author: Nathan Cutler -Date: Wed Apr 19 16:53:57 2017 +0200 - - Revert "osdc/Objecter: If osd full, it should pause read op which w/ rwordered flag." - - This reverts commit 2d68822c784eb4d62d3b0198ed4ec04404dbffb3. - - Signed-off-by: Nathan Cutler - -commit 0efe16d2566f0d6040f61fafd38c6661f08da1cd -Author: Nathan Cutler -Date: Wed Apr 19 16:53:49 2017 +0200 - - Revert "osdc/Objecter: resend RWORDERED ops on full" - - This reverts commit f2474042ecd6560323673170c13f2cb964406e70. - - Signed-off-by: Nathan Cutler - -commit 33af18e59298926d9636023faf871a191d582c1a -Author: Nathan Cutler -Date: Tue Apr 18 08:06:01 2017 +0200 - - doc: mention --show-mappings in crushtool manpage - - Fixes: http://tracker.ceph.com/issues/19649 - Signed-off-by: Loic Dachary - Signed-off-by: Nathan Cutler - (cherry picked from commit b48b6f4ed8b5f5b5852cbbfd5b3d5b650efb7f1b) - -commit 6dcd5fa8474c3a43ffca2394557b11f4906ee1ff -Merge: 7c006fc164 68fcb01211 -Author: Gregory Farnum -Date: Wed Apr 19 02:47:27 2017 -0400 - - Merge pull request #14596 from gregsfortytwo/wip-17916-divergent - - Wip 17916 divergent - - Reviewed-by: Josh Durgin - -commit 97cd21afc54efe3afb482b041f9c34ab6cdc682e -Author: Nathan Cutler -Date: Sun Apr 9 20:11:27 2017 +0200 - - tests: Thrasher: handle "OSD has the store locked" gracefully - - On slower machines (VPS, OVH) it takes time for the OSD to go down. - - Fixes: http://tracker.ceph.com/issues/19556 - Signed-off-by: Nathan Cutler - (cherry picked from commit a5b19d2d73540b730392f8001c8601f2cecc1b51) - -commit faeb8088ac3bd16bc04a1e5a55fec50285a7253f -Author: Casey Bodley -Date: Wed Apr 5 16:19:57 2017 -0400 - - rgw: fix for null version_id in fetch_remote_obj() - - commit 8b43c9781206c22d9aedb4beb8d669bf1e23169f fixed the wrong use of - the dest_obj's version, but removed the check for "null" version - - Signed-off-by: Casey Bodley - (cherry picked from commit 915370776df5b964c2ee8d9f9329562919eef8d5) - -commit 6180fcb4ec33bdade37f5693193712f8016b2560 -Author: Zhang Shaowen -Date: Fri Mar 17 16:26:56 2017 +0800 - - rgw: version id doesn't work in fetch_remote_obj - - Signed-off-by: Zhang Shaowen - (cherry picked from commit 8b43c9781206c22d9aedb4beb8d669bf1e23169f) - - Conflicts: - src/rgw/rgw_rados.cc: trivial: dest_obj.key.instance in master - versus dest_obj.get_instance() in Jewel - -commit d079b91479abfb474fad4fafe54c119f009ea572 -Author: liuchang0812 -Date: Fri Feb 10 18:02:03 2017 +0800 - - rgw: don't return skew time in pre-signed url - - Fixes: http://tracker.ceph.com/issues/18828 - - Signed-off-by: liuchang0812 - (cherry picked from commit dd8b348f4aad0124e8a4457117bf3f5f76af7bdb) - -commit 9b77b16b888b8efbf7d50d333e1880a6ec70d87a -Author: Loic Dachary -Date: Thu Apr 13 23:49:50 2017 +0200 - - ceph-disk: enable directory backed OSD at boot time - - https://github.com/ceph/ceph/commit/539385b143feee3905dceaf7a8faaced42f2d3c6 - introduced a regression preventing directory backed OSD from starting at - boot time. - - For device backed OSD the boot sequence starts with ceph-disk@.service - and proceeds to - - systemctl enable --runtime ceph-osd@.service - - where the --runtime ensure ceph-osd@12 is removed when the machine - reboots so that it does not compete with the ceph-disk@/dev/sdb1 unit at - boot time. - - However directory backed OSD solely rely on the ceph-osd@.service unit - to start at boot time and will therefore fail to boot. - - The --runtime flag is selectively set for device backed OSD only. - - Fixes: http://tracker.ceph.com/issues/19628 - - Signed-off-by: Loic Dachary - (cherry picked from commit f425a127b7487d2093c8c943f0bcdec3d673d601) - - Conflicts: - src/ceph-disk/ceph_disk/main.py: trivial: Jewel does not support - OpenRC and other inits, hence no corresponding 'elif' - -commit 68fcb01211e064f0d200cc9c9576254e9a6b949c -Author: Greg Farnum -Date: Mon Apr 17 18:09:55 2017 -0700 - - pglog: require users set a config option before ignoring divergent_priors - - Signed-off-by: Greg Farnum - -commit b9477303b010b3653934f77fa533df01aeff1c3c -Author: Greg Farnum -Date: Fri Apr 7 14:33:20 2017 -0700 - - osd: pglog: clean up divergent_priors off disk when running; don't assert on startup - - Fixes: http://tracker.ceph.com/issues/17916 - - Signed-off-by: Greg Farnum - -commit 7c006fc1640d2f6b9dc9002bfd994ecdf25510ee -Merge: e31a540dce 721b2083cd -Author: Gregory Farnum -Date: Mon Apr 17 17:41:19 2017 -0400 - - Merge pull request #14492 from gregsfortytwo/wip-jewel-snaptrim - - Backport snap trimming improvements to Jewel - - Reviewed-by: Josh Durgin - -commit 721b2083cd2ed8567d41aba6b17caf30b678e6b3 -Merge: a84dc8fe74 e31a540dce -Author: Gregory Farnum -Date: Mon Apr 17 17:39:59 2017 -0400 - - Merge branch 'jewel' into wip-jewel-snaptrim - -commit a84dc8fe747d1e17c7910c9857f7f5d3d56e2dff -Author: Greg Farnum -Date: Mon Apr 17 14:32:38 2017 -0700 - - PendingReleaseNotes: discuss snap trim improvements - - Signed-off-by: Greg Farnum - -commit 360a9d9af003c650cdf00534909d6488c702c413 -Author: Greg Farnum -Date: Wed Apr 12 16:30:55 2017 -0700 - - PrimaryLogPG: reimplement osd_snap_trim_sleep within the state machine - - Rather than blocking the main op queue, just pause for that amount of - time between state machine cycles. - - Also, add osd_snap_trim_sleep to a few of the thrasher yamls. - - Signed-off-by: Samuel Just - (cherry picked from commit 2ed7759cfeb03e71f0fbd98fe7c2db2bb741861c) - - Conflicts: - src/osd/PrimaryLogPG.cc - - Signed-off-by: Greg Farnum - (cherry picked from commit 67336454a4cee66522bc0ca01b2c58b8960f75ec) - - Conflicts: - qa/suites/rados/thrash/thrashers/default.yaml - qa/suites/rados/thrash/thrashers/pggrow.yaml - src/osd/OSD.h - src/osd/ReplicatedPG.cc - src/osd/ReplicatedPG.h - - Signed-off-by: Greg Farnum - -commit 18dbf6a0245e35dbbdb5ddb760182795b37983c0 -Author: Samuel Just -Date: Thu Jan 26 15:41:21 2017 -0800 - - rados: check that pool is done trimming before removing it - - Signed-off-by: Samuel Just - (cherry picked from commit 4aebf59d906fa3e03d21bdac182f89fe3cd4c802) - (cherry picked from commit 34398c29b3c57f00d932cf96570f882dce64a82b) - -commit 7f78450bd184335ba0098f3ded0fefa2c79a5dd5 -Author: Greg Farnum -Date: Tue Apr 11 14:04:19 2017 -0700 - - osd/ReplicatedPG: limit the number of concurrently trimming pgs - - This patch introduces an AsyncReserver for snap trimming to limit the - number of pgs on any single OSD which can be trimming, as with backfill. - Unlike backfill, we don't take remote reservations on the assumption - that the set of pgs with trimming work to do is already well - distributed, so it doesn't seem worth the implementation overhead to get - reservations from the peers as well. - - Signed-off-by: Samuel Just - (cherry picked from commit 21cc515adfb225ba70f1d80b1b76f0345c214c22) - - Conflicts: - src/osd/PrimaryLogPG.cc - src/osd/PrimaryLogPG.h - - Signed-off-by: Greg Farnum - - (cherry picked from commit 68ea24396ca6450d4d8042a7c5f51306b7d199fa) - (cherry picked from commit c7176b869898c870b56b1762958652d801af4c4c) - - Conflicts: Many. As evidenced by involving two distinct patches - in this one commit, it wasn't a clean backport. - - Signed-off-by: Greg Farnum - -commit 527911fab78b4752313a4a2a5d3ab0ae736bc50f -Author: Kefu Chai -Date: Mon Apr 10 14:53:46 2017 +0800 - - mon/MonClient: make get_mon_log_message() atomic - - * LogClient: move reset_session() into get_mon_log_message() and add a - "flush" param to the latter. so it can get_mon_log_message() - atomically. otherwise another call changing the log queue could sneak - in between reset_session() and get_mon_log_message(). - * MonClient: add a "flush" param to do_send() so we can reset the - LogClient session once we are connected to a monitor. - - Fixes: http://tracker.ceph.com/issues/19427 - Signed-off-by: Kefu Chai - (cherry picked from commit 5215e291da2b527d85e129eda86043490843178e) - - Conflicts: - src/mon/MonClient.cc: handle_auth: replaced 'log_client->reset_session(); - send_log();' sequence with newly introduced 'send_log(true);' like - the original patch does - -commit e31a540dcea96b3d5b4f7ecd20e2d54e81a68e2b -Merge: 7c36d1650f 06916a8798 -Author: Nathan Cutler -Date: Fri Apr 14 22:28:51 2017 +0200 - - Merge pull request #13834 from smithfarm/wip-18969-jewel - - jewel: rgw: Change loglevel to 20 for 'System already converted' message - - Reviewed-by: Casey Bodley - Reviewed-by: Nathan Cutler - -commit 7c36d1650fdb7a357ff724b37cf59b4da413f76b -Merge: 0e3aa2cb01 4c1f302f7d -Author: Nathan Cutler -Date: Fri Apr 14 22:26:51 2017 +0200 - - Merge pull request #13833 from smithfarm/wip-18908-jewel - - jewel: rgw: the swift container acl does not support field .ref - - Reviewed-by: Radoslaw Zarzynski - -commit b698d1fa4ce4aca5e392eeec600e3357c3cf71a9 -Author: Jason Dillaman -Date: Thu Mar 16 12:28:41 2017 -0400 - - librbd: is_exclusive_lock_owner API should ping OSD - - This is required to detect if a peer has been silently blacklisted - and is therefore no longer the lock owner. - - Fixes: http://tracker.ceph.com/issues/19287 - Signed-off-by: Jason Dillaman - (cherry picked from commit e15db05960a284bdf3701256722299d553cfd5aa) - - Conflicts: - src/librbd/ManagedLock.[h|cc]: logic moved to ExclusiveLock - - (cherry picked from commit 7e30b630e2806c73ea503871599f958b58df7934) - -commit 0e3aa2cb011be8a2af69040b94ee240ea6c1e663 -Merge: 8d5a5ddfec 419c9926d9 -Author: Nathan Cutler -Date: Thu Apr 13 11:09:27 2017 +0200 - - Merge pull request #13214 from ovh/bp-osd-updateable-throttles-jewel - - jewel: osd: allow client throttler to be adjusted on-fly, without restart - - Reviewed-by: Josh Durgin - -commit 8d5a5ddfecc7c020eeb8aeb927e007bc04885534 -Merge: 091aaa2ab7 bcd3c906e5 -Author: Nathan Cutler -Date: Thu Apr 13 10:44:23 2017 +0200 - - Merge pull request #14326 from shinobu-x/wip-15025-jewel - - jewel: osd: new added OSD always down when full flag is set - - Reviewed-by: Josh Durgin - -commit d311eea6bc5b2c88984cfc16340970ea1459b74a -Author: Robin H. Johnson -Date: Sat Feb 11 10:32:53 2017 -0800 - - msg/simple/Pipe: support IPv6 QoS. - - Extend DSCP marking for heartbeat packets to IPv6, as commit - 9b9a682fe035c985e416ee1c112fa58f9045a27c only implemented - support for IPv4. - - Conflicts: Cherry-picked 91a29bc490fdfbbef0875fa620c7ba1a1a6492ae from master to avoid conflict. - Backport: jewel, luminious - Fixes: http://tracker.ceph.com/issues/18887 - Signed-off-by: Robin H. Johnson - (cherry picked from commit 2d6021fbf7a728f73c2998be17e9224f14b83a30) - -commit 332b5174c769f395074255e075de8d2cc1ee4021 -Author: Jason Dillaman -Date: Thu Mar 16 12:27:08 2017 -0400 - - pybind: fix incorrect exception format strings - - Signed-off-by: Jason Dillaman - (cherry picked from commit 68617455f534a612ade1331f43b032ab524704ae) - -commit 091aaa2ab768858e840e2d05e0896c229ce69984 -Merge: 3f2e4cd2d5 d30c4d55ad -Author: Nathan Cutler -Date: Wed Apr 12 11:38:48 2017 +0200 - - Merge pull request #13874 from smithfarm/wip-19171-jewel - - jewel: doc: rgw S3 create bucket should not do response in json - - Reviewed-by: Abhishek Lekshmanan - -commit 3f2e4cd2d53f1cbdbdb1fc8687c5ffe0d46b346b -Merge: ea0bc6c553 0e11a938c5 -Author: Nathan Cutler -Date: Wed Apr 12 11:06:40 2017 +0200 - - Merge pull request #13492 from shinobu-x/wip-18516-jewel - - jewel: build/ops: "osd marked itself down" will not recognised if host runs mon + osd on shutdown/reboot - - Reviewed-by: Nathan Cutler - -commit ea0bc6c553454f7641a2594013412aee142fbd11 -Merge: 845972f4de d012c381e8 -Author: Nathan Cutler -Date: Wed Apr 12 11:01:23 2017 +0200 - - Merge pull request #13254 from shinobu-x/wip-14609-jewel - - jewel: common: radosstriper: protect aio_write API from calls with 0 bytes - - Reviewed-by: Kefu Chai - -commit 845972f4de1d803aa4dab0e1afaed693bbba088f -Merge: a3deef997f cfa37d6a16 -Author: Nathan Cutler -Date: Wed Apr 12 10:52:10 2017 +0200 - - Merge pull request #13489 from shinobu-x/wip-18955-jewel - - jewel: ceph-disk: bluestore --setgroup incorrectly set with user - - Reviewed-by: Loic Dachary - Reviewed-by: Nathan Cutler - -commit a3deef997ff0800b9e3e2d141cfc6fdc73cac837 -Merge: 702edb5519 39aab763a4 -Author: Nathan Cutler -Date: Wed Apr 12 10:51:42 2017 +0200 - - Merge pull request #14070 from smithfarm/wip-19339-jewel - - jewel: tests: dummy suite fails in OpenStack - - Reviewed-by: Loic Dachary - -commit 702edb5519e67bc5f8c5b65c6f63c9635cd758cf -Merge: f509ccc4b7 a20d2b89ee -Author: Nathan Cutler -Date: Wed Apr 12 10:51:12 2017 +0200 - - Merge pull request #14329 from smithfarm/wip-19493-jewel - - jewel: ceph-disk: Racing between partition creation & device node creation - - Reviewed-by: Loic Dachary - Reviewed-by: Sébastien Han - -commit f509ccc4b734697267cec35ff57dd7f1c5aaaddb -Merge: c8c4bff2af f2474042ec -Author: Nathan Cutler -Date: Wed Apr 12 10:48:29 2017 +0200 - - Merge pull request #14427 from smithfarm/wip-19140-jewel - - jewel: osdc/Objecter: If osd full, it should pause read op which w/ rwordered flag - - Reviewed-by: Kefu Chai - -commit c8c4bff2afa7fe0339a97f32e69bb9d0546f1318 -Merge: 349baea1a4 b5b441abaa -Author: Nathan Cutler -Date: Wed Apr 12 10:48:08 2017 +0200 - - Merge pull request #14324 from shinobu-x/wip-19371-jewel - - jewel: common: monitor creation with IPv6 public network segfaults - - Reviewed-by: Kefu Chai - -commit 349baea1a4486e475e0381a6b316d64a6ce0139c -Merge: dd466b7d9a 72e2476a13 -Author: Nathan Cutler -Date: Wed Apr 12 10:45:35 2017 +0200 - - Merge pull request #14112 from shinobu-x/wip-19192-jewel - - jewel: tools: brag fails to count "in" mds - - Reviewed-by: Kefu Chai - -commit dd466b7d9acb03c8830bdd83b3b73602c68083c2 -Merge: b8f8bd0a94 ee06517547 -Author: Nathan Cutler -Date: Wed Apr 12 10:45:11 2017 +0200 - - Merge pull request #14150 from smithfarm/wip-18823-jewel - - jewel: tests: run-rbd-unit-tests.sh assert in lockdep_will_lock, TestLibRBD.ObjectMapConsistentSnap - - Reviewed-by: Jason Dillaman - -commit b8f8bd0a949c917b119eca91aec95c6a971a1fb4 -Merge: 222916a375 1cc8d0d085 -Author: Nathan Cutler -Date: Wed Apr 12 10:44:47 2017 +0200 - - Merge pull request #14152 from smithfarm/wip-18893-jewel - - jewel: librbd: Incomplete declaration for ContextWQ in librbd/Journal.h - - Reviewed-by: Jason Dillaman - -commit 222916a3758b0253bcb851b5e955f6970f171306 -Merge: 49f84b1a14 b85677397e -Author: Nathan Cutler -Date: Wed Apr 12 10:44:25 2017 +0200 - - Merge pull request #14154 from smithfarm/wip-18948-jewel - - jewel: tests: additional rbd-mirror test stability improvements - - Reviewed-by: Jason Dillaman - -commit 49f84b1a14295e80ef35204ed691b3131c5f744c -Merge: 2a232d43a1 f6489d01ca -Author: Nathan Cutler -Date: Wed Apr 12 10:44:04 2017 +0200 - - Merge pull request #14148 from smithfarm/wip-18778-jewel - - jewel: rbd: rbd --pool=x rename y z does not work - - Reviewed-by: Jason Dillaman - -commit 2a232d43a18b47399f96abeac3ac800f204a9eaf -Merge: 413ac584d6 8bed107b84 -Author: Nathan Cutler -Date: Wed Apr 12 10:43:41 2017 +0200 - - Merge pull request #14083 from smithfarm/wip-19357-jewel - - jewel: rbd: systemctl stop rbdmap unmaps all rbds and not just the ones in /etc/ceph/rbdmap - - Reviewed-by: Jason Dillaman - -commit 413ac584d63fd469ea28defa68c39538444d01b1 -Merge: 23d595b1b0 547e867628 -Author: Nathan Cutler -Date: Wed Apr 12 10:43:21 2017 +0200 - - Merge pull request #13154 from smithfarm/wip-18496-jewel - - jewel: librbd: Possible deadlock performing a synchronous API action while refresh in-progress - - Reviewed-by: Jason Dillaman - -commit 23d595b1b0fb874adfd1507d86db261cf6a193a9 -Merge: 4add6f5580 915dbace5d -Author: Nathan Cutler -Date: Wed Apr 12 10:42:54 2017 +0200 - - Merge pull request #13244 from smithfarm/wip-18775-jewel - - jewel: rbd: qemu crash triggered by network issues - - Reviewed-by: Jason Dillaman - -commit 4add6f5580b7d65571bc426e95fac35be7150ff4 -Merge: 37ab19cc29 1a4e1e09b1 -Author: Nathan Cutler -Date: Wed Apr 12 10:42:23 2017 +0200 - - Merge pull request #13809 from asheplyakov/18321-bp-jewel - - jewel: librbd: remove image header lock assertions - - Reviewed-by: Jason Dillaman - -commit 37ab19cc29d61970e08af1b4627137cfcaa99474 -Merge: f7c04e3ca6 8d0140a9ed -Author: Nathan Cutler -Date: Wed Apr 12 10:41:56 2017 +0200 - - Merge pull request #13107 from smithfarm/wip-18669-jewel - - jewel: tests: [ FAILED ] TestLibRBD.ImagePollIO in upgrade:client-upgrade-kraken-distro-basic-smithi - - Reviewed-by: Jason Dillaman - -commit f7c04e3ca69c7fc134b267e825277eccea228c9b -Merge: d2909bdb4c a18a2dd108 -Author: Nathan Cutler -Date: Wed Apr 12 10:39:26 2017 +0200 - - Merge pull request #13585 from asheplyakov/jewel-bp-16585 - - jewel: msg: set close on exec flag - - Reviewed-by: Kefu Chai - -commit d2909bdb4cf135de850cc865f735cc61eb3d4ea6 -Merge: cd748603da 6d47615c11 -Author: Nathan Cutler -Date: Wed Apr 12 03:56:22 2017 +0200 - - Merge pull request #14371 from tchaikov/wip-19429-jewel - - jewel: tests: clone workunit using the branch specified by task - - Nathan Cutler - -commit cd748603da43c0b7377b5ef07aff618711b4879c -Merge: 1a20c12355 dd25a8f36b -Author: Nathan Cutler -Date: Wed Apr 12 03:54:11 2017 +0200 - - Merge pull request #14325 from shinobu-x/wip-18619-jewel - - jewel: osd: degraded and misplaced status output inaccurate - - Reviewed-by: Josh Durgin - -commit 1a20c123556e2f566af7e87ab3da6fc2ca328bb9 -Merge: 4838c4db4c 7fdf4d41c5 -Author: Nathan Cutler -Date: Wed Apr 12 03:53:25 2017 +0200 - - Merge pull request #14236 from smithfarm/wip-19392-jewel - - jewel: mon: remove bad rocksdb option - - Reviewed-by: Josh Durgin - -commit 4838c4db4c610926edc599677cb7e74f3bdc3077 -Merge: e26b7033eb 3860ccf16d -Author: Nathan Cutler -Date: Wed Apr 12 03:52:41 2017 +0200 - - Merge pull request #14181 from mslovy/wip-19394-jewel - - jewel: osd: bypass readonly ops when osd full - - Reviewed-by: Josh Durgin - -commit e26b7033ebd4a09210d482a54677806fe2c03505 -Merge: 389150bcc3 819af9e413 -Author: Nathan Cutler -Date: Wed Apr 12 03:49:49 2017 +0200 - - Merge pull request #14113 from shinobu-x/wip-19319-jewel - - jewel: cli: RadosImport::import should return an error if Rados::connect fails - - Reviewed-by: Brad Hubbard - Reviewed-by: David Zafman - -commit 389150bcc376702022cf15c0c79dbe856a97dffa -Merge: a8b10082f5 90de64bd81 -Author: Nathan Cutler -Date: Wed Apr 12 03:45:33 2017 +0200 - - Merge pull request #14047 from asheplyakov/reindex-on-pg-split - - jewel: osd: reindex properly on pg log split - - Reviewed-by: Sage Weil - Reviewed-by: Josh Durgin - -commit a8b10082f5761ea1132b85bd916a84f87e0276e2 -Merge: 32ed9b7897 ae498e84ff -Author: Nathan Cutler -Date: Wed Apr 12 03:45:15 2017 +0200 - - Merge pull request #14044 from mslovy/wip-19311-jewel - - jewel: core: os/filestore: fix clang static check warn use-after-free - - Reviewed-by: Sage Weil - Reviewed-by: Josh Durgin - -commit 32ed9b789733e21a2d597c7016eabf95763e50ae -Merge: 6705e911a2 335b5fa4a9 -Author: Nathan Cutler -Date: Wed Apr 12 03:44:25 2017 +0200 - - Merge pull request #13932 from asheplyakov/18911-bp-jewel - - jewel: rbd-nbd: check /sys/block/nbdX/size to ensure kernel mapped correctly - - Reviewed-by: Mykola Golub - Reviewed-by: Jason Dillaman - -commit 6705e911a229cee50d5ac992c7dd9af834f2400d -Merge: 3d21a0080d 714eb863c3 -Author: Nathan Cutler -Date: Wed Apr 12 03:42:13 2017 +0200 - - Merge pull request #13831 from jan--f/wip-19206-jewel - - jewel: fs: Invalid error code returned by MDS is causing a kernel client WARNING - - Reviewed-by: John Spray - -commit 3d21a0080dfd00c0b89ed23fd8049e46c3041af4 -Merge: 8a6d64331b 6b5322c5f6 -Author: Nathan Cutler -Date: Wed Apr 12 03:41:36 2017 +0200 - - Merge pull request #13827 from tchaikov/wip-19185-jewel - - jewel: osd: ReplicatedPG: try with pool's use-gmt setting if hitset archive not found - - Reviewed-by: Sage Weil - Reviewed-by: Josh Durgin - -commit 8a6d64331b73a4a24a59f4dcdb6771f1dc638879 -Merge: f96392a589 cebba011e5 -Author: Nathan Cutler -Date: Wed Apr 12 03:40:09 2017 +0200 - - Merge pull request #13788 from shinobu-x/wip-18235-jewel - - jewel: core: os/filestore/HashIndex: be loud about splits - - Reviewed-by: Josh Durgin - -commit f96392a58970ab1e81653d5657d3d680120a02b6 -Merge: 8fe6ffcfef 1d054c3856 -Author: Nathan Cutler -Date: Wed Apr 12 03:39:27 2017 +0200 - - Merge pull request #13786 from shinobu-x/wip-19129-jewel - - jewel: build/ops: ceph-base package missing dependency for psmisc - - Reviewed-by: Nathan Cutler - -commit 8fe6ffcfef8918fd8634e87255ad3fe7a544aa16 -Merge: 6f589a159e b249fd5bd8 -Author: Nathan Cutler -Date: Wed Apr 12 03:38:34 2017 +0200 - - Merge pull request #13732 from liewegas/wip-19119-jewel - - jewel: doc: PendingReleaseNotes: note about 19119 - - Reviewed-by: Nathan Cutler - -commit 6f589a159e4eb5f6c96634a742acaf6cd6e174c2 -Merge: b8f2d35884 8c7a1df251 -Author: Nathan Cutler -Date: Wed Apr 12 03:36:21 2017 +0200 - - Merge pull request #13541 from shinobu-x/wip-18929-jewel - - jewel: osd: restrict want_acting to up+acting on recovery completion - - Reviewed-by: Josh Durgin - Reviewed-by: Nathan Cutler - -commit b8f2d35884a52586a28f1ff4eaf99c8c3ba1c43f -Merge: 40d1443d68 2d17092fab -Author: Nathan Cutler -Date: Wed Apr 12 03:34:54 2017 +0200 - - Merge pull request #13477 from asheplyakov/jewel-bp-18951 - - jewel: osd: --flush-journal: sporadic segfaults on exit - - Reviewed-by: Sage Weil - Reviewed-by: Josh Durgin - -commit 40d1443d68b788737f83c10db3811bc37dbe8dda -Merge: 50e863e0f4 07501dec6f -Author: Nathan Cutler -Date: Wed Apr 12 03:33:25 2017 +0200 - - Merge pull request #13261 from shinobu-x/wip-18587-jewel - - jewel: mon: OSDMonitor: make 'osd crush move ...' work on osds - - Reviewed-by: Josh Durgin - Reviewed-by: Nathan Cutler - -commit 25e43ac25619d883d5a04e2df1cf6f57fea73fd0 -Author: Nathan Cutler -Date: Tue Apr 11 08:23:23 2017 +0200 - - tests: fix oversight in yaml comment - - When the file was copied from the hammer version, the word "hammer" - was not changed to "infernalis". - - Signed-off-by: Nathan Cutler - - This cannot be cherry-picked from master because the test has been dropped. - -commit f2474042ecd6560323673170c13f2cb964406e70 -Author: Sage Weil -Date: Thu Mar 2 21:20:08 2017 -0600 - - osdc/Objecter: resend RWORDERED ops on full - - Our condition for respecting the FULL flag is complex, and involves - the WRITE | RWORDERED flags vs the FULL_FORCE | FULL_TRY flags. Previously, - we could block a read bc of RWORDRED but not resend it later. - - Fix by capturing the complex condition in a respects_full() bool and using - it both for the blocking-on-send and resending-on-possibly-notfull-later - checks. - - Fixes: http://tracker.ceph.com/issues/19133 - Signed-off-by: Sage Weil - (cherry picked from commit c4b73f19a7be13ff412eef804efcd8c18ed4dae6) - -commit 2d68822c784eb4d62d3b0198ed4ec04404dbffb3 -Author: Jianpeng Ma -Date: Thu May 5 23:44:57 2016 +0800 - - osdc/Objecter: If osd full, it should pause read op which w/ rwordered flag. - - Signed-off-by: Jianpeng Ma - (cherry picked from commit 07b2a22210e26eac1b2825c30629788da05e5e12) - -commit 2271cd81282f3f026316134cbab630f3e4f47782 -Author: Brad Hubbard -Date: Mon Apr 3 13:37:17 2017 +1000 - - ceph-disk: Populate mount options when running "list" - - Also tidy up by moving duplicated code into a function - - Fixes: http://tracker.ceph.com/issues/17331 - Signed-off-by: Brad Hubbard - (cherry picked from commit 7943ab2e01e24f2dfc5b6f1d3ffdc8a49e01af45) - -commit 65465356b5e7cb6cba67f8cbb81259d21e888dfb -Author: Ken Dreyer -Date: Tue Feb 16 12:56:34 2016 -0700 - - debian: replace SysV rbdmap with systemd service - - Stop shipping /etc/init.d/rbdmap in the Debian packages. Ship the - rbdmap.service systemd unit file instead. - - The corresponding change has already been made for RPMs, in - 9224ac2ad25f7d017916f58b642c0ea25305c3e5. - - For Upstart-based systems (eg Ubuntu Trusty), the Debian packages - already contain rbdmap.conf. - - (This gets us a tiny bit closer to being able to remove the rbdmap SysV - script from our tree entirely.) - - Signed-off-by: Ken Dreyer - (cherry picked from commit 839807118dda2fb4d57ed9d50ec46e3ee0e2820a) - - Conflicts: - debian/ceph-common.install (retain /etc/init.d/rbdmap so jewel users can choose sysvinit or systemd) - debian/rules (retain /etc/init.d/rbdmap so jewel users can choose sysvinit or systemd) - -commit 6d47615c11a216733ba368f0dbd9a0a1b9c8fd35 -Author: Kefu Chai -Date: Sat Apr 1 23:04:22 2017 +0800 - - qa/tasks/workunit.py: use "overrides" as the default settings of workunit - - otherwise the settings in "workunit" tasks are always overridden by the - settings in template config. so we'd better follow the way of how - "install" task updates itself with the "overrides" settings: it uses the - "overrides" as the *defaults*. - - Fixes: http://tracker.ceph.com/issues/19429 - Signed-off-by: Kefu Chai - (cherry picked from commit 47080150a17d238f38d9da824d227393ad767aad) - -commit fdc71e75cd6361be49c4c20e77d3bdff017b38bf -Author: Kefu Chai -Date: Thu Mar 30 12:37:01 2017 +0800 - - tasks/workunit.py: specify the branch name when cloning a branch - - c1309fb failed to specify a branch when cloning using --depth=1, which - by default clones the HEAD. and we can not "git checkout" a specific - sha1 if it is not HEAD, after cloning using '--depth=1', so in this - change, we dispatch "tag", "branch", "HEAD" using three Refspec classes. - - Signed-off-by: Kefu Chai - Signed-off-by: Dan Mick - (cherry picked from commit 9ca7ccf5f1739f731da8bf31260594aea3a2932d) - - Conflicts: - qa/tasks/workunit.py (trivial resolution) - -commit 6c14a803894d71bae858705855551a44bdf9bf67 -Author: Dan Mick -Date: Tue Mar 28 20:08:13 2017 -0700 - - tasks/workunit.py: when cloning, use --depth=1 - - Help avoid killing git.ceph.com. A depth 1 clone takes about - 7 seconds, whereas a full one takes about 3:40 (much of it - waiting for the server to create a huge compressed pack) - - Signed-off-by: Dan Mick - (cherry picked from commit c1309fbef300a062138ac40eb5d3e5081b833072) - -commit f8aa6be06cc628b1cf64c9196c30045d020d657e -Author: Nathan Cutler -Date: Fri Apr 7 06:59:13 2017 +0200 - - build/ops: rpm: move $CEPH_EXTRA_CONFIGURE_ARGS to right place - - Signed-off-by: Nathan Cutler - -commit bb3eb4284fe74cbf78e4a406c7b5f67a8e3c84b3 -Author: Nathan Cutler -Date: Fri Apr 7 06:48:51 2017 +0200 - - build/ops: rpm: explicitly provide --with-ocf to configure - - Fixes: http://tracker.ceph.com/issues/19546 - Signed-off-by: Nathan Cutler - - (Note: This cannot be cherry-picked because master uses cmake, but - the fix does bring the jewel spec file into better alignment its master - counterpart, at least as far as this one little bit is concerned.) - -commit 30c952735327d05b6049e9364c1a053ebf651751 -Author: lu.shasha -Date: Mon Feb 27 15:52:43 2017 +0800 - - rgw: use separate http_manager for read_sync_status - - concurrent users of read_sync_status() use different cr managers, when get_resource must - use http_manager related to the cr manager. - - Fixes: http://tracker.ceph.com/issues/19236 - - Signed-off-by: Shasha Lu - (cherry picked from commit c412024889f8995d98096ac863bafee71624bd70) - -commit 87cb8474f523be1b281882c46a8a3597977a51c9 -Author: Casey Bodley -Date: Thu Jul 28 20:20:29 2016 -0400 - - rgw: pass cr registry to managers - - Signed-off-by: Casey Bodley - (cherry picked from commit ef4d7eab11fb5d2a41c9c28b9eb8b075aaff0d05) - - Conflicts: - src/rgw/rgw_rados.cc: removed ref to RGWSyncLogTrimThread (not backported) - -commit 1a6d7c0506d4bcda775dda05bb357d4d5695dabb -Author: Casey Bodley -Date: Fri Jul 22 11:00:16 2016 -0400 - - rgw: use separate cr manager for read_sync_status - - RGWCoroutinesManager::run() is not reentrant, so concurrent users of - read_sync_status() must use different managers - - Signed-off-by: Casey Bodley - (cherry picked from commit 6b1e40d7a21c549b55e6576ec56973c8d3c092d1) - -commit c466adee9c8f7a97ff7e99bee56da8ce51bf0f00 -Author: Casey Bodley -Date: Thu Jul 21 23:46:20 2016 -0400 - - rgw: change read_sync_status interface - - RGWDataSyncStatusManager::read_sync_status() now operates on the given - parameter, rather than its internal member variable. this allows - multiple concurrent readers, which is needed for the rest interface - - Signed-off-by: Casey Bodley - (cherry picked from commit b7cd4e0e8b879b5e528de75bea3307585b96cbf2) - -commit 36921a3f3d01547508dc6270f6b19f2576e067d9 -Author: Casey Bodley -Date: Wed Jun 8 11:24:11 2016 -0400 - - rgw: don't ignore ENOENT in RGWRemoteDataLog::read_sync_status() - - rest handlers for sync status need to return ENOENT errors. the only - other callers are in radosgw-admin, so the ENOENT errors are ignored at - those call sites instead - - Signed-off-by: Casey Bodley - (cherry picked from commit 2cc533b30c0f23c0750ea8d02c51b3b3d3b4821a) - -commit b249fd5bd816a63b445db12c8f846cfda199c8b8 -Author: Sage Weil -Date: Wed Mar 1 13:18:44 2017 -0600 - - PendingReleaseNotes: warning about 'osd rm ...' and #19119 - - See http://tracker.ceph.com/issues/19119 - - Signed-off-by: Sage Weil - (cherry picked from commit be96003c464481d8e84825178d600234a0d64d22) - -commit 335258f975a8e8539774e8cb22690d746ec90d9f -Author: Ilya Dryomov -Date: Tue Mar 28 11:49:08 2017 +0200 - - osdc/Objecter: respect epoch barrier in _op_submit() - - Epoch barrier instructs us to avoid sending (i.e. pause) any OSD ops - until we see a barrier epoch. The only thing epoch_barrier check in - target_should_be_paused() does is keep already paused ops paused. We - need to actually pause incoming OSD ops in _op_submit(). - - Fixes: http://tracker.ceph.com/issues/19396 - Signed-off-by: Ilya Dryomov - (cherry picked from commit f8e8efc0a53d7bd807cc0c2178aef7c4bed62ab7) - -commit a20d2b89ee13e311cf1038c54ecadae79b68abd5 -Author: Erwan Velu -Date: Fri Mar 31 14:54:33 2017 +0200 - - ceph-disk: Adding retry loop in get_partition_dev() - - There is very rare cases where get_partition_dev() is called before the actual partition is available in /sys/block/. - - It appear that waiting a very short is usually enough to get the partition beein populated. - - Analysis: - update_partition() is supposed to be enough to avoid any racing between events sent by parted/sgdisk/partprobe and - the actual creation on the /sys/block//* entrypoint. - On our CI that race occurs pretty often but trying to reproduce it locally never been possible. - - This patch is almost a workaround rather than a fix to the real problem. - It offer retrying after a very short to be make a chance the device to appear. - This approach have been succesful on the CI. - - Note his patch is not changing the timing when the device is perfectly created on time and just differ by a 1/5th up to 2 seconds when the bug occurs. - - A typical output from the build running on a CI with that code. - command_check_call: Running command: /usr/bin/udevadm settle --timeout=600 - get_dm_uuid: get_dm_uuid /dev/sda uuid path is /sys/dev/block/8:0/dm/uuid - get_partition_dev: Try 1/10 : partition 2 for /dev/sda does not in /sys/block/sda - get_partition_dev: Found partition 2 for /dev/sda after 1 tries - get_dm_uuid: get_dm_uuid /dev/sda uuid path is /sys/dev/block/8:0/dm/uuid - get_dm_uuid: get_dm_uuid /dev/sda2 uuid path is /sys/dev/block/8:2/dm/uuid - - fixes: #19428 - - Signed-off-by: Erwan Velu - (cherry picked from commit 93e7b95ed8b4c78daebf7866bb1f0826d7199075) - -commit 2d5d0aec60ec9689d44a53233268e9b9dd25df95 -Author: Erwan Velu -Date: Wed Mar 22 10:11:44 2017 +0100 - - ceph-disk: Reporting /sys directory in get_partition_dev() - - When get_partition_dev() fails, it reports the following message : - ceph_disk.main.Error: Error: partition 2 for /dev/sdb does not appear to exist - The code search for a directory inside the /sys/block/get_dev_name(os.path.realpath(dev)). - - The issue here is the error message doesn't report that path when failing while it might be involved in. - - This patch is about reporting where the code was looking at when trying to estimate if the partition was available. - - Signed-off-by: Erwan Velu - (cherry picked from commit 413c9fcfbe8e6ab33d73b8428090ccacc33c5d15) - -commit bcd3c906e5b57e7f44df1963c6e11b78ff89482c -Author: Mingxin Liu -Date: Mon Mar 13 23:41:58 2017 +0800 - - osd: don't share osdmap with objecter when preboot - - Signed-off-by: Mingxin Liu - (cherry picked from commit a5a3644eecc49b4eea890c6999fe87536495dcbe) - -commit dd25a8f36bef1901f3ce6193cfcbdaf7ab2424a1 -Author: David Zafman -Date: Wed Jan 18 08:33:40 2017 -0800 - - osd: Calculate degraded and misplaced more accurately - - Calculate num_object_copies based on the larger of pool size, - up set size and acting set size. - - Calculate num_objects_degraded as the difference between num_object_copies - and all copies found on acting set and backfilling up set OSDs. - - Calculate num_objects_misplaced as all copies on acting set OSDs not in up set - less copies that have been backfilled to up set OSDs. - - Fixes: http://tracker.ceph.com/issues/18619 - - Signed-off-by: David Zafman - (cherry picked from commit 8423bc40759cca137f61e7b755411719a84369d4) - -commit b5b441abaa852e85ddefd8b22835c9b85898cc06 -Author: Fabian Grünbichler -Date: Wed Mar 22 16:13:50 2017 +0100 - - common: fix segfault in public IPv6 addr picking - - sockaddr is only 16 bytes big, so declaring net as sockaddr - and then casting to sockaddr_in6 in case of IPv6 cannot - work. - - using sockaddr_storage works for both IPv4 and IPv6, and is - used in other code parts as well. - - note that the tests did not find this issue as they declared - the bigger structs and casted the references to (sockaddr *) - - Fixes: http://tracker.ceph.com/issues/19371 - Signed-off-by: Fabian Grünbichler - (cherry picked from commit ae2ee3d3835fe25b35eeb1a841ee5234cd69eb65) - -commit 7fdf4d41c5bef14269cb302301a08d5a3a57a768 -Author: Sage Weil -Date: Mon May 2 15:29:12 2016 -0400 - - mon: remove bad rocksdb option - - Signed-off-by: Sage Weil - (cherry picked from commit 0ac671ece258e509f71a05253e62a9878e279840) - -commit 3860ccf16d7dfb137face9886c3d7f29cd527835 -Author: Jianpeng Ma -Date: Thu May 5 23:07:06 2016 +0800 - - osd: bypass readonly ops when osd full. - - Signed-off-by: Jianpeng Ma - (cherry picked from commit e2a0ae8e88e6b7354b14adb503fd8ba8525bee39) - See: http://tracker.ceph.com/issues/19394 - - Signed-off-by: yaoning - -commit b85677397ef9a3fe16c087e67d3f752851bbe070 -Author: Jason Dillaman -Date: Mon Nov 21 15:31:43 2016 -0500 - - qa/workunits/rbd: resolve potential rbd-mirror race conditions - - Fixes: http://tracker.ceph.com/issues/18935 - Signed-off-by: Jason Dillaman - (cherry picked from commit 63eae97afc1a92412525468263fb8696a243ebac) - -commit 1cc8d0d08560af4e8785d7c4c6a925b1fc9f988e -Author: Boris Ranto -Date: Wed Feb 8 23:47:57 2017 +0100 - - librbd: Include WorkQueue.h since we use it - - We use m_work_queue of type ContextWQ in handle_update function but we - do not include common/WorkQueue.h that defines ContextWQ. This results - in dereference of an incomplete type and causes build error in latest - Fedora rawhide (future 26). - - Fixes: http://tracker.ceph.com/issues/18862 - - Signed-off-by: Boris Ranto - (cherry picked from commit 480f82847ad1fc7959f1fe5a90761a5a24550993) - -commit ee06517547ae174472d739f966c0a27d3a97d742 -Author: Jason Dillaman -Date: Wed Jan 18 20:54:22 2017 -0500 - - librbd: avoid possible recursive lock when racing acquire lock - - Fixes: http://tracker.ceph.com/issues/17447 - Signed-off-by: Jason Dillaman - (cherry picked from commit 5e46e8eb664f573bd70ae7c96a6d9a98b0deb09e) - -commit f6489d01ca41d6979b5de28e3cde6b43fcaa8edb -Author: Gaurav Kumar Garg -Date: Mon Jan 30 13:03:20 2017 +0100 - - rbd: destination pool should be source pool if it is not specified - - Currently if user perform image rename operation and user give pool - name as a optional parameter (--pool=) then currently - its taking this optional pool name for source pool and making - destination pool name default pool name. - With this fix if user provide pool name as a optional pool name - parameter then it will consider both soruce and destination pool - name as optional parameter pool name. - - Fixes: http://tracker.ceph.com/issues/18326 - - Reported-by: МАРК КОРЕНБЕРГ - Signed-off-by: Gaurav Kumar Garg - (cherry picked from commit 01f23aa99fb694da326ab408e75b33c640ce660b) - -commit 9e123e6d6c8bbd54514b498df5a22d961f0cefbb -Author: Orit Wasserman -Date: Sun Mar 12 12:11:28 2017 +0200 - - rgw: use rgw_zone_root_pool for region_map like is done in hammer - - Fixes: http://tracker.ceph.com/issues/19195 - Signed-off-by: Orit Wasserman - (cherry picked from commit c91dd6d9efd148e0fe0f027dde537e977de9aa26) - -commit e2ee70a8ad51992bbd763d2465f6d8a01dad6a31 -Author: Orit Wasserman -Date: Thu Mar 9 13:03:24 2017 +0200 - - rgw: skip conversion of zones without any zoneparams - - Fixes: http://tracker.ceph.com/issues/19231 - Signed-off-by: Orit Wasserman - (cherry picked from commit 36cf5a5c8179c6313346b2e29286c537c6fefce8) - -commit c7d292bf6714d7aaf10412e5109badb90f5dc208 -Author: Orit Wasserman -Date: Thu Mar 9 11:16:26 2017 +0200 - - rgw: better debug information for upgrade - - Signed-off-by: Orit Wasserman - (cherry picked from commit e9f3bf8eab1dd46a92f54b0f7afe1f4c0e4204db) - -commit 11f5c841c7698e6239017478fa05f742b7c0ab1c -Author: Danny Al-Gaaf -Date: Tue Jan 31 18:01:32 2017 +0100 - - rgw/rgw_rados.cc: prefer ++operator for non-primitive iterators - - Signed-off-by: Danny Al-Gaaf - (cherry picked from commit 7086cf9a73f2ec1eb96c0e752beb1b74fca18570) - -commit 819af9e4139997cd845dc24a137d43218d8a40a8 -Author: Brad Hubbard -Date: Tue Mar 21 12:22:20 2017 +1000 - - tools/rados: Check return value of connect - - Fail gracefully if Rados::connect returns an error. - - Fixes: http://tracker.ceph.com/issues/19319 - Signed-off-by: Brad Hubbard - (cherry picked from commit c119091ef0844e4a1ddd790a8bfef8f06bb57d58) - -commit 72e2476a130e14abcd541ff61328454cb69ad9c3 -Author: Kefu Chai -Date: Mon Mar 6 11:33:27 2017 +0800 - - brag: count the number of mds in fsmap not in mdsmap - - this change was introduced in 4e9b953 - - Fixes: http://tracker.ceph.com/issues/19192 - Signed-off-by: Peng Zhang - (cherry picked from commit 2d25a9c0c760664d3de33ecca0e0272c1031cd46) - -commit 3cb192779a1c3662d27bba7715eb31c5f7b6a5b7 -Author: Oleh Prypin -Date: Thu Jun 30 00:51:50 2016 +0300 - - brag: Assume there are 0 MDS instead of crashing when data is missing - - Signed-off-by: Oleh Prypin - -commit 8bed107b84efdc8c735245cdfb51bfd8d07da13b -Author: David Disseldorp -Date: Fri Feb 10 19:19:46 2017 +0100 - - doc: update description of rbdmap unmap[-all] behaviour - - Fixes: http://tracker.ceph.com/issues/18884 - - Signed-off-by: David Disseldorp - (cherry picked from commit f987396e126d5e61240a6645ffed439f79b072b4) - -commit da4e0b56c60f4bc2c67daa5dfe4d5255ab8bfc03 -Author: Nathan Cutler -Date: Thu Dec 15 18:23:41 2016 +0100 - - doc: add verbiage to rbdmap manpage - - Fixes: http://tracker.ceph.com/issues/18262 - Signed-off-by: Nathan Cutler - (cherry picked from commit fbac4a081547d83bb2436cd60b0b7ee7250f8a6c) - -commit 167d4fd7ccf0cdac536f95250bbfa3e9879ab769 -Author: David Disseldorp -Date: Fri Feb 10 17:50:12 2017 +0100 - - rbdmap: unmap RBDMAPFILE images unless called with unmap-all - - When called with a "map" parameter, the rbdmap script iterates the list - of images present in RBDMAPFILE (/etc/ceph/rbdmap), and maps each entry. - When called with "unmap", rbdmap currently iterates *all* mapped RBD - images and unmaps each one, regardless of whether it's listed in the - RBDMAPFILE or not. - - This commit adds functionality such that only RBD images listed in the - configuration file are unmapped. This behaviour is the new default for - "rbdmap unmap". A new "unmap-all" parameter is added to offer the old - unmap-all-rbd-images behaviour, which is used by the systemd service. - - Fixes: http://tracker.ceph.com/issues/18884 - - Signed-off-by: David Disseldorp - (cherry picked from commit e58413abf408cbe254232e563f3e30d2dc0d707c) - -commit 39aab763a44e45e025c311cdfff95116df11a4c4 -Author: Nathan Cutler -Date: Thu Dec 15 13:01:02 2016 +0100 - - Revert "dummy: reduce run time, run user.yaml playbook" - - This reverts commit d4e3cec1851ae35889127b90912e133178085bc6. - - Fixes: http://tracker.ceph.com/issues/18259 - Signed-off-by: Nathan Cutler - (cherry picked from commit a8a2a8c3e68b910dbaeb3186576898bf9f89f9fd) - -commit 2e50fe1684f73ebe96969c341242b6f20c8470a0 -Author: Casey Bodley -Date: Thu Oct 20 15:01:01 2016 -0400 - - rgw: fix break inside of yield in RGWFetchAllMetaCR - - the yield macro is implemented with for/switch, so the breaks in - RGWFetchAllMetaCR weren't being applied to the for loop as expected - - so any of these breaks send RGWFetchAllMetaCR into an infinite loop - - removed the yield {} block, so that breaks will apply to the for loop as - intended, then added a single yield; statement to allow the - entries_index consumer to run one per iteration - - Fixes: http://tracker.ceph.com/issues/17655 - - Signed-off-by: Casey Bodley - (cherry picked from commit 190bd385a7be52867d65740c410884f5c8cbc21f) - -commit dc4e7a1a865ea0ae7362c1b6a7a542aa5f72107d -Author: Casey Bodley -Date: Fri Mar 3 12:10:40 2017 -0500 - - rgw: delete_system_obj() fails on empty object name - - Signed-off-by: Casey Bodley - (cherry picked from commit 67401193f871db95a6045915fa59dce8c5dd1012) - -commit e9a577c8c535702d1eb285429978bdbb395e2d5c -Author: Casey Bodley -Date: Fri Mar 3 11:42:45 2017 -0500 - - rgw: if user.email is empty, dont try to delete - - Fixes: http://tracker.ceph.com/issues/18980 - - Signed-off-by: Casey Bodley - (cherry picked from commit 022ecf0fcc8e44912c8758ee1d9a452dc23cbbce) - -commit 90de64bd81fedcb9540e40d50420e169a4a81248 -Author: Alexey Sheplyakov -Date: Mon Mar 20 14:05:17 2017 +0400 - - jewel: osd/PGLog: reindex properly on pg log split - - When PGLog::IndexedLog::split_into runs it builds the list, which means - the old indices are wrong (point to bad memory), but index() will not - rebuild them because ever since b858e86 we won't rebuild them if they - are already built. Fix that by calling unindex() before the split. - - Based on 643ae42cf27f16dd6ed4e1402acc0483bb9fca74. Notice that both - the child and the parent log are re-indexed in Jewel, so the only - problem is missing unindex(). - - Signed-off-by: Alexey Sheplyakov - -commit ae498e84ffcff7424721f0d2704ec739d1cc092d -Author: liuchang0812 -Date: Tue Dec 20 13:21:40 2016 +0800 - - os/filestore: fix clang static check warn use-after-free - - Signed-off-by: liuchang0812 - (cherry picked from commit 1d359455b3dd6abb383542ba596a03f14ac54dbd) - See: http://tracker.ceph.com/issues/19311 - - Signed-off-by: yaoning - -commit 335b5fa4a9694620546422f9a02bdcc16549d7cc -Author: Mykola Golub -Date: Thu Feb 2 11:11:35 2017 +0100 - - rbd-nbd: check /sys/block/nbdX/size to ensure kernel mapped correctly - - Fixes: http://tracker.ceph.com/issues/18335 - Signed-off-by: Mykola Golub - (cherry picked from commit 596e5ea8a5df72002672eef0a6d20572ca6f60f0) - - Conflicts: - qa/workunits/rbd/rbd-nbd.sh: the original commit removes - TOO_LARGE_IMAGE test, do the same thing - src/tools/rbd_nbd/rbd-nbd.cc: help git to add - "include/stringify.h" - - Other changes: - src/tools/rbd_nbd/rbd-nbd.cc: #include so - the code compiles - - Signed-off-by: Alexey Sheplyakov - -commit ced799f9c6558482d538f8dec854c62162685ad0 -Author: Michal Koutný -Date: Thu Jan 26 16:08:09 2017 -0500 - - rgw: Use decoded URI when verifying TempURL - - Instead of calliing url_decode directly, we reuse s->decoded_uri that is - initialized in RGWREST::preprocess(). - - Fixes: http://tracker.ceph.com/issues/18590 - - Adapted from 4e1318f4dcbfd64c3ec94f4addf6e38ddd6c013a. Cherry-picking - that patch requires a quite a number of unrelated changes, hence this - patch does s/s->info.request_uri/s->decoded_uri/ to keep the fix as - minimal as possible. - - Signed-off-by: Alexey Sheplyakov - -commit 043d70461c1eb874d9185f9bd671930fad05ff65 -Author: Ilya Dryomov -Date: Wed Mar 1 17:19:04 2017 +0100 - - osd/OSDMap: don't set weight to IN when OSD is destroyed - - Since commit 4e28f9e63644 ("osd/OSDMap: clear osd_info, osd_xinfo on - osd deletion"), weight is set to IN when OSD is deleted. This changes - the result of applying an incremental for clients, not just OSDs. - Because CRUSH computations are obviously affected, pre-4e28f9e63644 - servers disagree with post-4e28f9e63644 clients on object placement, - resulting in misdirected requests. - - Fixes: http://tracker.ceph.com/issues/19119 - Signed-off-by: Ilya Dryomov - (cherry picked from commit a6009d1039a55e2c77f431662b3d6cc5a8e8e63f) - -commit 6b5322c5f62f4b90c4206c6ddcc70d090fa7eeb9 -Author: Kefu Chai -Date: Tue Mar 7 18:49:46 2017 +0800 - - osd/ReplicatedPG: try with pool's use-gmt setting if hitset archive not found - - due to http://tracker.ceph.com/issues/19185, a hammer OSD could store - pg_hit_set_info_t with "use-gmt = true" even the pool setting is false. - so we use the pool setting as a fallback if the hitset archive is not - found locally and the pool.use_gmt does not match with hitset.use_gmt. - - Fixes: http://tracker.ceph.com/issues/19185 - Signed-off-by: Kefu Chai - - Conflicts: - osd/ReplicatedPG.cc: This cannot be cherry-picked from master - because: hammer should upgrade to jewel first before moving to a - higher version. so there is no necessary to include this workaround - in releases later than jewel. - -commit d30c4d55ad52e2b63fdbd06ce256d92fc3fd36c9 -Author: Abhishek Lekshmanan -Date: Thu Feb 16 17:40:50 2017 +0100 - - doc: rgw: make a note abt system users vs normal users - - Mention that system users don't behave like normal users in context of - normal rest operations - - Fixes: http://tracker.ceph.com/issues/18889 - Signed-off-by: Abhishek Lekshmanan - (cherry picked from commit a47bcf70c9f51a6601b809cba219f5615b204d34) - - Conflicts: - doc/radosgw/multisite.rst (trivial whitespace difference) - -commit 5ee8feaba469886f9e3bd3909475ffef62ba261d -Author: root -Date: Tue Feb 7 14:37:36 2017 +0530 - - rgw: Let the object stat command be shown in the usage - - Fixes: http://tracker.ceph.com/issues/19013 - Signed-off-by: Pavan Rallabhandi - (cherry picked from commit 0fe76f83d19be098ef54fb0492a376fef3aa9e23) - -commit 754b4a482cb0369215beed58103a1e241231cf77 -Author: root -Date: Tue Feb 21 16:33:29 2017 +0530 - - rgw: Correct the return codes for the health check feature - Fixes: http://tracker.ceph.com/issues/19025 - Signed-off-by: Pavan Rallabhandi - - (cherry picked from commit 4da2bf310f6d43423554c32e43ebf90ad2c3f3a9) - -commit 9cd7dd84909abdb9e603ff3aeb9958cdab8c70ad -Author: Ronak Jain -Date: Wed Feb 22 12:03:46 2017 +0530 - - rgw: Fixes typo in rgw_admin.cc - - Issue: http://tracker.ceph.com/issues/19026 - Signed-off-by: Ronak Jain - (cherry picked from commit 58837ef6ce8cbcfc2cac29d5f833b2cf62d8737a) - -commit 85fbb00f6ef5f11bc5d615ccd8e2202ce3896fd1 -Author: Yehuda Sadeh -Date: Mon Feb 27 10:35:01 2017 -0800 - - rgw: don't init rgw_obj from rgw_obj_key when it's incorrect to do so - - Fixes: http://tracker.ceph.com/issues/19096 - - rgw_obj_key currently deals with the bucket index key, and not - representing a (name, instance, ns) tupple. Need to initialize - it in two steps. - - Signed-off-by: Yehuda Sadeh - (cherry picked from commit 392c5d9dae6ba699014ffe6e1e67818fa62d7e41) - -commit ec0668c201a71b4a17ef0ab3c5908f57229aa6ef -Author: Casey Bodley -Date: Tue Nov 29 11:29:41 2016 -0500 - - rgw: fix for broken yields in RGWMetaSyncShardCR - - Fixes: http://tracker.ceph.com/issues/18076 - - Signed-off-by: Casey Bodley - (cherry picked from commit e62d48a9bf2e309eab1a863f167af5267ebcc371) - -commit 6afe3efa4b636ede1cd77086cb2e70ed09fa2e95 -Author: Abhishek Lekshmanan -Date: Mon Jul 25 11:21:11 2016 +0200 - - rgw: kill a compile warning for rgw_sync - - killing the compile warning for - - ``` - /ceph/src/rgw/rgw_sync.cc:1462:12: - warning: suggest explicit braces to avoid ambiguous ‘else’ [-Wparentheses] - if (can_adjust_marker) yield { - ^ - ``` - - Signed-off-by: Abhishek Lekshmanan - Signed-off-by: Casey Bodley - (cherry picked from commit 831640bb46621a6f003ad562cef7928ffa9a7ad3) - -commit 06916a8798439ec033294d791749ce7381d92f51 -Author: Vikhyat Umrao -Date: Mon Feb 13 23:07:25 2017 +0530 - - rgw: change log level to 20 for 'System already converted' message - - Fixes: http://tracker.ceph.com/issues/18919 - - Signed-off-by: Vikhyat Umrao - (cherry picked from commit 55b567c767830170d04de4cdc8f10aba30a3f379) - -commit 4c1f302f7d71bedb0dead220f17eeb84e7e3f737 -Author: Jing Wenjun -Date: Wed Jan 11 05:28:43 2017 +0800 - - rgw: the swift container acl should support field .ref - - On the openstack-swift. The container acl supports .ref, which is ignored on ceph swift. - - Fixes: http://tracker.ceph.com/issues/18484 - Signed-off-by: Jing Wenjun - (cherry picked from commit b06f9cd9f0900db7b0d0fbcaea69cdd0d4b10132) - - Conflicts: - src/rgw/rgw_acl_swift.cc - no extract_referer_urlspec() in jewel - see https://github.com/ceph/ceph/pull/8657 - -commit 714eb863c30df4e653068e6ea16630504e58b704 -Author: xie xingguo -Date: Tue Jun 14 19:32:01 2016 +0800 - - server: negative error code when responding to client - - As the comment suggests. Also a zero or positive return code - shall indicates a success, which does not match our intention here. - - Signed-off-by: xie xingguo - (cherry picked from commit 26931f888ce4661765cca106b3a3dc66702266df) - Signed-off-by: Jan Fajerski - -commit 1a4e1e09b1e562bf97cfe96f5cb9f937b6987165 -Author: Jason Dillaman -Date: Tue Dec 13 14:10:58 2016 -0500 - - librbd: remove image header lock assertions - - This assertions can sporadically fail if the watch is lost and - recovered in the background. Upon a true loss of the lock, the - client would either be blacklisted or it would have completed - all in-flight ops before releasing. - - Fixes: http://tracker.ceph.com/issues/18244 - Signed-off-by: Jason Dillaman - (cherry picked from commit ce4f2a52ec0a794d89e7576b59c9b9aefe3db288) - - Conflicts: - src/librbd/operation/SnapshotCreateRequest.cc: rbd class - does not support the snapshot namespaces in Jewel, skip - the corresponding argument - -commit cebba011e502f7009208bbddc295eb17f88f1bb9 -Author: Dan van der Ster -Date: Fri Dec 9 22:06:26 2016 +0100 - - os/filestore/HashIndex: be loud about splits - - Filestore splits are a rare yet important enough event that an - OSD should visibly report when they happen. - - Without this reporting an operator could spend hours trying to - understand the cause of any split-induced slow requests. - - Fixes: http://tracker.ceph.com/issues/18235 - Signed-off-by: Dan van der Ster - (cherry picked from commit 61c47acd3a1f3e01f0106d4a541bb7f28a1301d8) - -commit 1d054c3856a63ceebe44f66ff83fda691c374f71 -Author: Nathan Cutler -Date: Thu Mar 2 12:41:07 2017 +0100 - - build/ops: add psmisc dependency to ceph-base (deb and rpm) - - Fixes: http://tracker.ceph.com/issues/19129 - Signed-off-by: Nathan Cutler - (cherry picked from commit 769b695465162bc8424abf8e2f259e6765b5bbff) - - Conflicts: - debian/control (jewel does not have f11acf2b 7e71cd2c) - -commit 6add2a457e2826b71c0e9e82c6f6686cecbc4584 -Author: Jing Wenjun -Date: Fri Nov 25 21:31:22 2016 +0800 - - rgw: metadata sync info should be shown at master zone of slave zonegroup - - When executing 'radosgw-admin sync status', the metadata sync info should be shown on the srceen at master zone of slave zonegroup. - - Using the function store->is_meta_master() instead of 'zonegroup.is_master && zone.id == zonegroup.master_zone' - - Fixes: http://tracker.ceph.com/issues/18091 - Signed-off-by: Jing Wenjun - (cherry picked from commit c12d0af2f98b246a73cc3ee027449a22192795b3) - -commit 0e11a938c5c9acd8a50efa9a154ea3bf21bcafc5 -Author: Boris Ranto -Date: Wed Jan 25 12:39:40 2017 +0100 - - systemd: Start OSDs after MONs - - Currently, we start/stop OSDs and MONs simultaneously. This may cause - problems especially when we are shutting down the system. Once the mon - goes down it causes a re-election and the MONs can miss the message - from the OSD that is going down. - - Resolves: http://tracker.ceph.com/issues/18516 - - Signed-off-by: Boris Ranto - (cherry picked from commit 7f4acf45dd0d86e7d9992a8c30e5876fb57b1914) - - Conflicts: - systemd/ceph-osd@.service (jewel does not have 4179aa8d) - -commit 3bdd4398f1dcad0b7e22f1750ca524b97feca15a -Author: yaoning -Date: Mon Jun 6 13:31:52 2016 +0800 - - osd: preserve allocation hint attribute during recovery - - Signed-off-by: yaoning - (cherry picked from commit e15be792960da6bac2bd469acf7d30007be61781) - - Conflicts: - src/osd/ReplicatedBackend.cc (in master, it contains alloc_hint_flags for set_alloc_hint) - src/osd/ReplicatedPG.cc (in master, it contains alloc_hint_flags in object_info_t struct) - src/osd/osd_types.cc (in master, it contains alloc_hint_flags in message serialization) - alloc_hint_flags is used in master bluestore, filestore does not use alloc_hint_flags. - therefore, remove alloc_hint_flags here in jewel - - Signed-off-by: yaoning - -commit 8d0140a9eda814beadf1f59c9b4205f30a1d2e35 -Author: Jason Dillaman -Date: Tue Jan 24 09:24:52 2017 -0500 - - librbd: improve debug logging for lock / watch state machines - - Signed-off-by: Jason Dillaman - (cherry picked from commit cc046597983bd491cc66081cc33d9046264fe24b) - - Conflicts: - NOTE: cherry-picked from kraken commit to avoid conflicts - -commit 62ce3461c3b205eaa9062113526cf572184d0a27 -Author: Jason Dillaman -Date: Mon Jan 23 21:24:41 2017 -0500 - - test: use librados API to retrieve config params - - The CephContext object is not ABI-stable, so it is necessary to - use the "conf_get" librados methods to safely retrieve a setting. - - Fixes: http://tracker.ceph.com/issues/18617 - Signed-off-by: Jason Dillaman - (cherry picked from commit 2ed02f3cd56bf89984c3538ac3f21ec2321cd3b7) - - Conflicts: - src/test/librbd/test_librbd.cc (jewel does not have - 006138e2d80b779d8c15b141002bb4b3852f6c4a or - cb3712e08cdc2c37a983b479f4692bbdfe83b220) - -commit 01d04e28db7c2969b86df5b38a20b9eb156cf393 -Author: Nathan Cutler -Date: Thu Feb 2 23:23:54 2017 +0100 - - tests: Thrasher: eliminate a race between kill_osd and __init__ - - If Thrasher.__init__() spawns the do_thrash thread before initializing the - ceph_objectstore_tool property, do_thrash races with the rest - of Thrasher.__init__() and in some cases do_thrash can call kill_osd() before - Trasher.__init__() progresses much further. This can lead to an exception - ("AttributeError: Thrasher instance has no attribute 'ceph_objectstore_tool'") - being thrown in kill_osd(). - - This commit eliminates the race by making sure the ceph_objectstore_tool - attribute is initialized before the do_thrash thread is spawned. - - Fixes: http://tracker.ceph.com/issues/18799 - Signed-off-by: Nathan Cutler - (cherry picked from commit b519d38fb1967628ad8a1c46fcfb3f984de58790) - - Conflicts: - qa/tasks/ceph_manager.py (jewel has only one if statement after - "self.thread = gevent.spawn(self.do_thrash)" while master has four; - jewel lacks 66836c957ffd974dec136997e23261ec7de2f0aa which disables - ceph-objectstore-tool testing in master) - -commit 08a667883b68ccc72e3a4bc3013856deef1df93d -Author: Nathan Cutler -Date: Sat Feb 18 14:33:25 2017 +0100 - - rpm: build ceph-resource-agents by default - - To align with debian build - - Fixes: http://tracker.ceph.com/issues/17613 - Signed-off-by: Nathan Cutler - (cherry picked from commit 3e157bf16c3020ac11cb26df5df3ed331faf3c25) - -commit d22becab0f2c541584ce891d392760a5c4f1d153 -Author: Yan Jun -Date: Thu Jul 14 19:10:29 2016 +0800 - - msg/simple: cleanups - - should save the `errno` which may be changed by `ldout` and/or `<<` operator - - Signed-off-by: Yan Jun - (cherry picked from commit 91a29bc490fdfbbef0875fa620c7ba1a1a6492ae) - Signed-off-by: Robin H. Johnson - -commit a18a2dd108678d2e4b57e08b559c1f9a262d6923 -Author: Kefu Chai -Date: Fri Jun 17 13:58:55 2016 +0800 - - msg/simple: set close on exec on server sockets - - mds execv() when handling the "respawn" command, to avoid fd leakage, - and enormous CLOSE_WAIT connections after respawning, we need to set - FD_CLOEXEC flag for the socket fds. - - Fixes: http://tracker.ceph.com/issues/16390 - Signed-off-by: Kefu Chai - (cherry picked from commit f019ad563ce90f5aea0d8dd8b7b98688441596e0) - -commit 91a968b8fc7b363cae351b8648259211a1e71d18 -Author: Kefu Chai -Date: Fri Jun 17 01:17:05 2016 +0800 - - msg/async: set close on exec on server sockets - - mds execv() when handling the "respawn" command, to avoid fd leakage, - and enormous CLOSE_WAIT connections after respawning, we need to set - FD_CLOEXEC flag for the socket fds. - - Fixes: http://tracker.ceph.com/issues/16390 - Signed-off-by: Kefu Chai - (cherry picked from commit eaf68c724144d07f9506037a14d9192cb9f16d70) - - Conflicts: - src/msg/async/AsyncMessenger.cc: Processor::accept(): applied - the hunk manually (invoke set_close_on_exec on a socket - returned by accept) - -commit 547e867628975c7144590e9332aa62b0ef82a433 -Author: Jason Dillaman -Date: Thu Jan 5 12:12:57 2017 -0500 - - librbd: possible deadlock with flush if refresh in-progress - - Fixes: http://tracker.ceph.com/issues/18419 - Signed-off-by: Jason Dillaman - (cherry picked from commit b95f92a5572d3035c20eba07e76d2c825a9853f7) - - Conflicts: - src/librbd/ImageState.h (master commit just adds a function - declaration, so just add it to jewel as well) - -commit 07501dec6f1c70afd4e4c2a50d7f874c39f2220b -Author: Sage Weil -Date: Wed Jan 18 17:02:54 2017 -0600 - - mon/OSDMonitor: make 'osd crush move ...' work on osds - - Currently it only allows you to move buckets, which is annoying and much - less useful. To move an OSD you need to use create-or-move, which is - harder to use. - - Fixes: http://tracker.ceph.com/issues/18587 - Signed-off-by: Sage Weil - (cherry picked from commit 47956475dea8bb8e07331dd76344a60b776b5158) - - Conflicts: - qa/workunits/mon/crush_ops.sh: adapted "ceph osd find" to jewel syntax - -commit 7c6c3c753ccdd3baea834338e1a761f05b4e0a12 -Author: Vikhyat Umrao -Date: Thu Feb 16 23:51:11 2017 +0530 - - auth: 'ceph auth import -i' overwrites caps, if caps are not specified - in given keyring file, should alert user and should not allow this import. - Because in 'ceph auth list' we keep all the keyrings with caps and importing - 'client.admin' user keyring without caps locks the cluster with error[1] - because admin keyring caps are missing in 'ceph auth'. - - [1] Error connecting to cluster: PermissionDeniedError - - Fixes: http://tracker.ceph.com/issues/18932 - - Signed-off-by: Vikhyat Umrao - (cherry picked from commit 90144aa64c11a685b6a7cb3aafea75d427f569be) - -commit 8c7a1df251e8289e7cf2df5b3096b91d8640695d -Author: Sage Weil -Date: Tue Feb 14 15:00:09 2017 -0500 - - osd/PG: restrict want_acting to up+acting on recovery completion - - On recovery completion we recalculate want_acting to see if we - should add recently backfilled osds into acting. However, at - this point we may have gotten infos from others OSDs outside - of up/acting that could be used for want_acting. We currently - assert that only up/acting osds are used in - PG::RecoveryState::Active::react(const AdvMap&), so we must - restrict want_acting to up/acting here. - - We could remove this restriction, but it would mean - - 1) checking on every map change that want_acting hasn't been - invalidated, and if so, recalculating want_acting and requesting - a new pg_temp. Also, presumably - - 2) on each new info, checking whether we can construct a better - want_acting, and if so, doing it. - - That would be a good thing, but is a more complicated change. In - reality this case comes up very rarely, so simply make our - post-recovery want_acting calculation limit itself to up+acting. - - See 1db67c443d84dc5d1ff53cc820fdfd4a2128b680 for the assertion. - - Signed-off-by: Sage Weil - (cherry picked from commit 0f2dee9aa48a00a7f2f809cd4d20e98df771da81) - -commit cfa37d6a1674e3f6f8eef4d8519823a7af70df01 -Author: craigchi -Date: Thu Feb 16 19:21:48 2017 +0800 - - ceph-disk: Fix getting wrong group name when --setgroup in bluestore - - ceph-disk prepare --setgroup will be wrong when using with - bluestore - - Signed-off-by: craigchi - (cherry picked from commit a8c0870e7370a0948e8e7fd53d3376b85bf9c649) - -commit 2d17092fab8080f819369d74d4c76d8ae58d899b -Author: Alexey Sheplyakov -Date: Tue Feb 7 16:47:45 2017 +0400 - - ceph-osd: --flush-journal: sporadic segfaults on exit - - FileStore holds a number of recources like op thread pool and work - queue, key/value DB threads, etc. These should be properly stopped - (released) before exiting to avoid segfaults on exit. - - Note: more code paths (mkfs, dump_journal, etc) need similar fixes, - these will be submitted as separate patches. - - Fixes: http://tracker.ceph.com/issues/18820 - Signed-off-by: Alexey Sheplyakov - (cherry picked from commit 00184814c156f6194a6ba4b696073ca1c18a3f8f) - - Adjustments: - - release g_ceph_context in the same way as the main code path does - -commit d012c381e8c59994ea9a40dc006d23f1bdd6a026 -Author: Sebastien Ponce -Date: Tue May 10 11:27:59 2016 +0200 - - radosstriper : protect aio_write API from calls with 0 bytes - - an assertion was failing so far, while we only have to return without doing anything - - Signed-off-by: Sebastien Ponce - (cherry picked from commit 7cce1e8c51640f466d8bb37a21c0d5f1b00db8ab) - -commit 915dbace5ddea69fff29f7965f213229b6fbc0ac -Author: Jason Dillaman -Date: Tue Jan 17 11:55:00 2017 -0500 - - osdc: cache should ignore error bhs during trim - - A read error (such as injecting a timeout into an OSD op) might result - in a bh in an error state. These should be trimable by the cache. - - Fixes: http://tracker.ceph.com/issues/18436 - Signed-off-by: Jason Dillaman - (cherry picked from commit 5910ed9de9856b5821488a1836487bbbd3d6460e) - -commit 419c9926d9ed57cb60228bc95956a9a1471b92cb -Author: Piotr Dałek -Date: Tue Jan 31 16:07:18 2017 +0100 - - OSD: allow client throttler to be adjusted on-fly, without restart - - This patch allows the osd_client_message_cap and - osd_client_message_size_cap to be adjusted on-fly, using admin socket - functionality. - - Fixes: http://tracker.ceph.com/issues/18791 - Signed-off-by: Piotr Dałek - (cherry picked from commit 64c309d7e18a975931b526e6f5d6f610c3a0d632) - - Conflicts: - src/osd/OSD.cc (suppressed post-jewel option) - -commit 957c19b844fb44cde78ad59f872815f82bbf23b8 -Author: Ali Maredia -Date: Thu Nov 10 13:58:35 2016 -0500 - - swift: added "--cluster" to rgw-admin command for multisite support - - Signed-off-by: Ali Maredia - -commit 8423bc1eefa45366bdd215a17c61701c9b05dfdd -Merge: 6dc30c4140 c078534376 -Author: Andrew Schoen -Date: Thu Apr 23 14:49:30 2015 -0500 - - Merge pull request #470 from ceph/wip-remote - - Add timeouts to Remote connection functions - -commit 6dc30c4140a833fd6cd126f8b5c1eceebad90510 -Merge: d55484f9e5 c078534376 -Author: Andrew Schoen -Date: Mon Apr 20 12:38:29 2015 -0500 - - Merge pull request #466 from ceph/wip-11426 - - Log stderr in get_latest_image_version_deb() - -commit d55484f9e562779e7d47a8f63ff029337dc01eef -Merge: c078534376 cd72cf2b31 -Author: Dan Mick -Date: Wed Apr 15 10:30:46 2015 -0700 - - Merge pull request #462 from ceph/wip-ssh-keys - - When modifying authorized_keys, store a backup - -commit c078534376d594aa3bf70d1d2e5dfc09ca8ae248 -Merge: cd72cf2b31 fce2ed683f -Author: Zack Cerza -Date: Tue Apr 14 11:38:15 2015 -0600 - - Merge pull request #460 from zhouyuan/mkdir_p - - Make parent directories as needed - -commit cd72cf2b3132e7c6371aa4dbfe7564ad3ad0509e -Author: Yehuda Sadeh -Date: Mon Nov 24 09:59:30 2014 -0800 - - swift: set full access to subusers creation - - Default subuser permissions are 'none'. - - Signed-off-by: Yehuda Sadeh - -commit fce2ed683fcd3798db968a40200f8e8f215595fa -Author: Zack Cerza -Date: Wed Aug 6 10:06:34 2014 -0600 - - Remove most ceph-specific tasks. They are in ceph-qa-suite now. - - Signed-off-by: Zack Cerza - -commit fac452ae55594aea482db5c13a0bd0207b6ecff6 -Author: Zack Cerza -Date: Thu Mar 27 11:35:28 2014 -0500 - - Revert "Lines formerly of the form '(remote,) = ctx.cluster.only(role).remotes.keys()'" - - This reverts commit d693b3f8950ffd1f2492a4db0f8234fee31f00f0. - -commit e98b107302e06fa5c3c628a7ab0e7455de9ab568 -Author: Warren Usui -Date: Fri Feb 28 19:13:40 2014 -0800 - - Lines formerly of the form '(remote,) = ctx.cluster.only(role).remotes.keys()' - and '(remote,) = ctx.cluster.only(role).remotes.iterkeys()' would fail with - ValueError and no message if there were less than 0 or more than 1 key. - Now a new function, get_single_remote_value() is called which prints out - more understandable messages. - - Fixes: 7510 - Reviewed-by: Josh Durgin - Signed-off-by: Warren Usui - -commit e5fe884edfec66cc7e520938bd520aa8f1344f85 -Merge: 8e2cdbf5ed f8bf53c4fe -Author: Alfredo Deza -Date: Wed Feb 19 16:40:17 2014 -0500 - - Merge pull request #186 from ceph/wip-7369 - - Fix #7369: "sed expression must be raw string" - -commit 8e2cdbf5ede871ebde260e6bdaec13daae03cfc5 -Merge: 132f3e8ae6 f8bf53c4fe -Author: Zack Cerza -Date: Fri Feb 14 14:59:04 2014 -0600 - - Merge pull request #188 from ceph/wip-calamari-onefile - - Add Calamari test tasks, test script - -commit 132f3e8ae664a35cad89896b7300be13b5c604ec -Merge: 24363351c3 7b63876676 -Author: Zack Cerza -Date: Fri Feb 14 11:50:55 2014 -0600 - - Merge pull request #192 from ceph/wip-6537-wusui - - Readjust the indentation of mon_clock_skew_check.py and mon_thrash.py. - -commit 24363351c3b1d23a8d6ef4e3f952855415f97ba8 -Merge: 4eb147291f 7b63876676 -Author: Zack Cerza -Date: Fri Feb 14 11:45:51 2014 -0600 - - Merge pull request #194 from ceph/wip-6534-wusui - - Add docstrings to internal.py - -commit 4eb147291fccdb2164a46e977ee43dbf50894e5a -Merge: ad9aaf8fa3 f4284b520a -Author: Zack Cerza -Date: Fri Feb 14 11:45:19 2014 -0600 - - Merge pull request #193 from ceph/wip-6538-wusui - - Add doc strings to Swift tests - -commit ad9aaf8fa35d49c33373fa69df7b38d3aca6abc5 -Merge: 7b63876676 f8bf53c4fe -Author: Zack Cerza -Date: Fri Feb 14 11:43:15 2014 -0600 - - Merge pull request #187 from ceph/wip-better-debug - - Debug output improvements - -commit f4284b520a554b1cbe130731741e53a7fcf4e35f -Author: Warren Usui -Date: Thu Feb 13 21:11:34 2014 -0800 - - Add doc strings to Swift tests - - Fixes: 6538 - Signed-off-by: Warren Usui - -commit 7b63876676f1a6845ba3b9147cf7bb2348ef2468 -Author: Warren Usui -Date: Tue Feb 11 20:21:06 2014 -0800 - - Add docstrings to s3 related tasks. - - Fixes: 6536 - Signed-off-by: Warren Usui - -commit f8bf53c4fe52009abdf730b05e2cb2ddbb412dea -Author: Zack Cerza -Date: Tue Sep 24 14:19:24 2013 -0500 - - Fix namespace collision - -commit db6efe3e0ba7446ca42baf2a50eef18a10cf4a10 -Merge: 611733c8b5 66555a4039 -Author: Alfredo Deza -Date: Tue Sep 24 08:17:22 2013 -0700 - - Merge pull request #106 from ceph/wip-mirror - - Remove lots of ceph.com hardcoding; default to upstream sources - -commit 66555a4039c61db9b96d6eecf8d2f298c98b6bad -Author: Zack Cerza -Date: Fri Sep 20 15:53:58 2013 -0500 - - Don't hardcode the git://ceph.com/git/ mirror - - Default to https://github.com/ceph/ but add a ~/teuthology.yaml option - -commit 611733c8b5ea55206df50c122efac612cb146c2f -Merge: 2346f1d735 6e8a3807c7 -Author: Sage Weil -Date: Fri Sep 6 13:24:34 2013 -0700 - - Merge pull request #78 from ceph/wip-6247 - - Move helper scripts to /usr/local/bin to clean up logs. - -commit 6e8a3807c766f728027c3099eebfa24cdc645bd1 -Author: Zack Cerza -Date: Fri Sep 6 15:08:01 2013 -0500 - - Helper scripts live in /usr/local/bin now! - -commit 2346f1d735ccb40d00b8ff61d4acb446f684b3b2 -Author: Joe Buck -Date: Fri Aug 23 19:54:53 2013 -0700 - - s3tests: extend for multi-region tests - - Added code to the s3tests task to extract - multi-region info so that that data - can be added to the S3TEST_CONF file - used to run S3 tests. - - Signed-off-by: Joe Buck - Reviewed-by: Josh Durgin - -commit d3b6d633e161bc422f4781c7be5011f8a7ed8a32 -Merge: 442a36c57c 09b01b27a3 -Author: Alfredo Deza -Date: Mon Aug 19 14:10:18 2013 -0700 - - Merge pull request #41 from ceph/wip-3791 - - Various usability and documentation fixes - -commit 442a36c57c3173b36ce17e5f85a49abf7e80ea93 -Merge: 09b01b27a3 9b2c4fa4ad -Author: wusui -Date: Fri Aug 16 14:47:59 2013 -0700 - - Merge pull request #40 from ceph/wip-teutharm-wusui - - Wip teutharm wusui - -commit 09b01b27a3a1310d4257133def60896ad37fb575 -Author: Zack Cerza -Date: Thu Aug 15 08:49:35 2013 -0500 - - Fix some instances where print is being used instead of log - -commit 9b2c4fa4ad4c4258b26526afb0c16c71ce47f593 -Author: Josh Durgin -Date: Wed Jul 31 13:32:58 2013 -0700 - - s3/swift tests: call radosgw-admin as the right client - - This allows the right region and zone info to be read from ceph.conf - - Signed-off-by: Josh Durgin - -commit 3b3816df58a3ba1f2f850faf8969ad070aa0046e -Author: Yehuda Sadeh -Date: Thu Jul 25 16:47:34 2013 -0700 - - s3tests: clone correct branch - - Signed-off-by: Yehuda Sadeh - -commit f5170fb460a2183aed33122b4e6d0117a4220fec -Merge: 7207a31e58 bd56af707a -Author: Sandon Van Ness -Date: Thu Jul 25 19:50:39 2013 -0700 - - Merge branch 'master' of github.com:ceph/teuthology - -commit 7207a31e5812dd0b29f2d6378360015622ddf4aa -Merge: bd56af707a 3da945512e -Author: Sandon Van Ness -Date: Thu Jul 25 19:50:02 2013 -0700 - - Merge remote-tracking branch 'origin/wip-sandon-vm' - - Conflicts: - teuthology/lock.py - teuthology/misc.py - teuthology/task/install.py - -commit bd56af707ae50c98ec46344cf57eb333061847b1 -Merge: 3da945512e 343a42c0d8 -Author: Josh Durgin -Date: Fri Jul 19 14:44:51 2013 -0700 - - Merge branch 'wip-centos-rgw' - -commit 343a42c0d86af5b8630a30716c03fc84ba22f944 -Author: Josh Durgin -Date: Tue Jul 9 18:50:52 2013 -0700 - - s3tests: fix client configurations that aren't dictionaries - - They're always used as dictionaries later on. - - Signed-off-by: Josh Durgin - -commit 3da945512ed78081be5dbe9ba59c836a311e1973 -Merge: 2c34d1971f 253cc98d98 -Author: Sage Weil -Date: Mon Jun 24 16:18:36 2013 -0700 - - Merge pull request #15 from ceph/wip-ulimits - - Reviewed-by: Warren Usui - -commit 253cc98d98855d65be7ebcdd46a39aa1004f8e67 -Author: Sage Weil -Date: Sun Jun 23 09:15:28 2013 -0700 - - enable-coredump -> adjust-ulimits - - and set max_files to be big, too! - -commit 2c34d1971f1e82311b364bf8efe60b223158d676 -Merge: 61dba20d1f b366ad334a -Author: Warren Usui -Date: Tue May 7 19:27:51 2013 -0700 - - Merge branch 'wip-teuth4768a-wusui' - - Conflicts: - teuthology/task/install.py - -commit 61dba20d1fce774eac5b56f0d61d4229460875c0 -Merge: a9f3eb6310 5a7267f85c -Author: Sage Weil -Date: Mon May 6 21:31:36 2013 -0700 - - Merge branch 'next' - -commit a9f3eb631064931cbdde7ef218c16e154bdb9991 -Author: Sage Weil -Date: Thu May 2 13:47:46 2013 -0700 - - s3tests: add force-branch with higher precdence than 'branch' - - This way we can force a branch despite something in overrides. - - Signed-off-by: Sage Weil - -commit b366ad334af55867ba781e22c8f87b6ac7775bf2 -Merge: 2a51e32891 5a7267f85c -Author: Josh Durgin -Date: Wed May 1 09:52:02 2013 -0700 - - Merge remote branch 'origin/next' - -commit 5a7267f85c80f88aca1b0081b07de1de3909f2e7 -Author: Josh Durgin -Date: Tue Apr 30 17:07:53 2013 -0700 - - fix some errors found by pyflakes - - Signed-off-by: Josh Durgin - -commit f866037f045887ccc5da15404935ce3361a74a08 -Author: Josh Durgin -Date: Tue Apr 30 13:23:22 2013 -0700 - - s3tests: revert useless portion of 1c50db6a4630d07e72144dafd985c397f8a42dc5 - - Perhaps it was attempting to debug something, but it shouldn't have been committed. - - Signed-off-by: Josh Durgin - -commit 2dcce57c00264b85cfb906223dfb89db9cc61ba5 -Author: Josh Durgin -Date: Tue Apr 30 16:49:04 2013 -0700 - - rgw tests: remove users after each test - - These should all be cleanup up at some point. They're - almost all the same code. - - Signed-off-by: Josh Durgin - -commit 3c604251d9014dba45f3adbb2008bac16ff6346d -Author: Josh Durgin -Date: Tue Apr 30 16:47:34 2013 -0700 - - rgw tests: clean up immediately after the test - - There's no need for an explicit cleanup function, so move it back - to where it came from (except in s3roundtrip, which did not have it). - - Instead, since these use a nested contextmanager, pass through - and yield to the top-level run_tasks after the nested - contextmanager has finished (and thus run all the cleanup steps - in the subtasks for this test). - - Signed-off-by: Josh Durgin - -commit 022bd4aa42312b80917169282cbfba655bbad6f1 -Author: Yehuda Sadeh -Date: Tue Apr 30 07:06:03 2013 -0700 - - swift, s3readwrite: add missing yield - - Signed-off-by: Yehuda Sadeh - -commit 820c72b8d0177b01f69887ec64d98702db37077c -Author: Yehuda Sadeh -Date: Mon Apr 29 11:24:04 2013 -0700 - - s3tests, s3readwrite, swift: cleanup explicitly - - Cleaning up test dir explicitly after run, so that - consecutive runs don't fail. - - Signed-off-by: Yehuda Sadeh - -commit 2a51e328913f16917a881129b475a4aeeab24ed0 -Merge: cccadb9b03 617534e769 -Author: Yehuda Sadeh -Date: Wed Feb 20 14:10:50 2013 -0800 - - Merge remote-tracking branch 'origin/wip-3634' - -commit cccadb9b03ca4421a5fd841a61bf252c329e3649 -Merge: 3eb19c8107 fa1f89478a -Author: Sage Weil -Date: Tue Feb 19 21:04:24 2013 -0800 - - Merge branch 'unstable' - - Conflicts: - teuthology/task/workunit.py - -commit fa1f89478a76373cb33cf2524e2ebf68b3cd622c -Author: Sander Pool -Date: Wed Feb 6 19:16:52 2013 +0000 - - Install ceph debs and use installed debs - - The ceph task installs ceph using the debian - packages now, and all invocations of binaries installed - in {tmpdir}/binary/usr/local/bin/ are replace with - the use of the binaries installed in standard locations - by the debs. - - Author: Sander Pool - Signed-off-by: Sam Lang - -commit 3eb19c810725b011baacdb8a6f5b172f4720a39a -Author: Sam Lang -Date: Wed Jan 23 14:37:39 2013 -0600 - - Replace /tmp/cephtest/ with configurable path - - Teuthology uses /tmp/cephtest/ as the scratch test directory for - a run. This patch replaces /tmp/cephtest/ everywhere with a - per-run directory: {basedir}/{rundir} where {basedir} is a directory - configured in .teuthology.yaml (/tmp/cephtest if not specified), - and {rundir} is the name of the run, as given in --name. If no name - is specified, {user}-{timestamp} is used. - - To get the old behavior (/tmp/cephtest), set test_path: /tmp/cephtest - in .teuthology.yaml. - - This change was modivated by #3782, which requires a test dir that - survives across reboots, but also resolves #3767. - - Signed-off-by: Sam Lang - Reviewed-by: Josh Durgin - -commit 2f829870e140c87b30e5b7aa3ad237a90dcb2179 -Author: Yehuda Sadeh -Date: Fri Dec 21 10:20:02 2012 -0800 - - task/swift: change upstream repository url - - Signed-off-by: Yehuda Sadeh - -commit 334d6386753510c312898552c6f92313942786ef -Merge: b8e6ce4db9 26df886d82 -Author: Joao Eduardo Luis -Date: Thu Nov 29 00:53:59 2012 +0000 - - Merge branch 'wip-mon-thrasher' - -commit b8e6ce4db9a603ce3523b1759c65eeadee55daa7 -Author: Sage Weil -Date: Thu Nov 22 13:59:58 2012 -0800 - - s3tests: fix typo - -commit 26df886d825e28c25b630887b8dcc1c8c6d687d8 -Author: Yehuda Sadeh -Date: Mon Nov 19 16:19:06 2012 -0800 - - rgw-logsocket: a task to verify opslog socket works - - Signed-off-by: Yehuda Sadeh - -commit 617534e76978acb09a9f925f18bba475a65a7dd2 -Author: Sage Weil -Date: Mon Sep 10 11:08:57 2012 -0700 - - s3tests: run against arbitrary branch/sha1 of s3-tests.git - -commit 7d5c7ee8c6f2bfedd193a8d3b7102b4cfe0bf74f -Author: Sage Weil -Date: Wed Jun 6 16:00:55 2012 -0700 - - pull s3-tests.git using git, not http - -commit ce951cf4caffd5d6883cc8dcd24372bcdf03690a -Author: Sage Weil -Date: Sat May 5 09:30:41 2012 -0700 - - ceph.newdream.net -> ceph.com - -commit 2b879905fcfd660e242ed1a804d1c8301d17ab84 -Merge: 1ac4bb10fc 1970713a2f -Author: Mark Nelson -Date: Wed Mar 14 15:32:23 2012 -0500 - - Merge branch 'master' of github.com:ceph/teuthology - -commit 1970713a2fc43e3afae376712356ca93a65d9e1f -Author: Sage Weil -Date: Fri Mar 2 10:55:19 2012 -0800 - - github.com/NewDreamNetwork -> github.com/ceph - -commit 1ac4bb10fc4b7d8d07c44b0e92b1627c721ab925 -Author: Josh Durgin -Date: Tue Feb 21 14:54:33 2012 -0800 - - Add necessary imports for s3 tasks, and keep them alphabetical. - -commit 92110e5a4460281139233dcea3f629d01182d398 -Author: Yehuda Sadeh -Date: Tue Feb 21 12:12:03 2012 -0800 - - rgw: access key uses url safe chars - - Signed-off-by: Yehuda Sadeh - -commit 709d9441127fec93da74c7702cafa54a47e10e8f -Author: Sage Weil -Date: Sun Jan 15 22:48:33 2012 -0800 - - use local mirrors for (most) github urls - - A cronjob on ceph.newdream.net updates these every 15 minutes. Sigh. - -commit 9598e47949ba65030c722947dc433e38875b1bd6 -Author: Tommi Virtanen -Date: Mon Dec 5 10:07:25 2011 -0800 - - Rename "testrados" and "testswift" tasks to not begin with "test". - - Anything "test*" looks like a unit test, and shouldn't be used for - actual code. - -commit 6236e7db22edac7b51fc6329188b6afa74f3fc78 -Author: Yehuda Sadeh -Date: Thu Nov 17 16:53:21 2011 -0800 - - testswift: fix config - -commit 1dd607cabb07126769b4beb1ba6677e21c448719 -Author: Yehuda Sadeh -Date: Wed Nov 16 16:00:01 2011 -0800 - - rgw: add swift task - - still not completely working (for some reason it skips all the tests) - -commit cb425c158085568cd92c239a071e282c74eddf1a -Author: Greg Farnum -Date: Fri Sep 30 09:26:42 2011 -0700 - - s3-tests: use radosgw-admin instead of radosgw_admin - - Signed-off-by: Greg Farnum - -commit 37d7d515345ab04c333d6fada722e432e5816eb3 -Author: Tommi Virtanen -Date: Fri Sep 16 11:09:45 2011 -0700 - - s3tests: Clone repository from github. - - Signed-off-by: Tommi Virtanen - -commit 29a242d97dd4a5a9110710027cdddb244b8b0e29 -Author: Tommi Virtanen -Date: Tue Sep 13 14:53:02 2011 -0700 - - Move orchestra to teuthology.orchestra so there's just one top-level package. - -commit ec49a5f263f71aa473257e3fd49d86e475fe9456 -Author: Tommi Virtanen -Date: Fri Sep 9 13:22:03 2011 -0700 - - Callers of task s3tests.create_users don't need to provide dummy "fixtures" dict. - -commit d7d995e82b45e6077040b467c8ef9a82a573faf7 -Author: Stephon Striplin -Date: Tue Aug 9 13:43:46 2011 -0700 - - allow s3tests.create_users defaults be overridden - -commit 0086109767d5bfbbc370ca13d3fe91895b207821 -Author: Josh Durgin -Date: Thu Jul 14 16:47:29 2011 -0700 - - Make targets a dictionary mapping hosts to ssh host keys. - -commit 1b2c96416f554c0890b2690291b9a2dc8a6dc17a -Author: Tommi Virtanen -Date: Wed Jul 6 14:17:24 2011 -0700 - - Skip s3-tests marked fails_on_rgw, they will fail anyway. - -commit 06fb9b95e39985630d89e1635dcd12510686d9cd -Author: Tommi Virtanen -Date: Tue Jul 5 09:27:28 2011 -0700 - - The shell exits after the command, hence there is no need for pushd/popd. - -commit cd524a6904bf8254edc73a9148308f642638e33d -Author: Josh Durgin -Date: Fri Jun 24 17:09:47 2011 -0700 - - Add s3tests task.