X-Git-Url: https://gerrit.opnfv.org/gerrit/gitweb?a=blobdiff_plain;f=src%2Fceph%2Fdoc%2Fchangelog%2Fv0.94.3.txt;fp=src%2Fceph%2Fdoc%2Fchangelog%2Fv0.94.3.txt;h=7c0a60d7570b6c0c3556c0a6fb56e7eea5a2a213;hb=812ff6ca9fcd3e629e49d4328905f33eee8ca3f5;hp=0000000000000000000000000000000000000000;hpb=15280273faafb77777eab341909a3f495cf248d9;p=stor4nfv.git diff --git a/src/ceph/doc/changelog/v0.94.3.txt b/src/ceph/doc/changelog/v0.94.3.txt new file mode 100644 index 0000000..7c0a60d --- /dev/null +++ b/src/ceph/doc/changelog/v0.94.3.txt @@ -0,0 +1,2660 @@ +commit 95cefea9fd9ab740263bf8bb4796fd864d9afe2b (tag: refs/tags/v0.94.3, refs/remotes/gh/hammer) +Author: Jenkins +Date: Wed Aug 26 10:39:37 2015 -0700 + + 0.94.3 + +commit 697101e4dfd9822050ce401b5f6212bfd81fea89 +Merge: 88e7ee7 81a311a +Author: Gregory Farnum +Date: Tue Aug 18 12:43:57 2015 +0100 + + Merge pull request #5589 from ceph/hammer-12709 + + Workunits : fs/misc/chmod.sh : Include ACL characters in permission check + + Reviewed-by: Greg Farnum + +commit 81a311a744987564b70852fdacfd915523c73b5d +Author: Yazen Ghannam +Date: Mon May 4 12:33:16 2015 -0400 + + Workunits : fs/misc/chmod.sh : Include ACL characters in permission check. + + Signed-off-by: Yazen Ghannam + (cherry picked from commit d3dbfffefb0ae53583350f53258dc902670da659) + +commit 88e7ee716fdd7bcf81845087021a677de5a50da8 +Merge: bb12f92 1a32379 +Author: Loic Dachary +Date: Tue Aug 4 13:02:17 2015 +0200 + + Merge pull request #5160 from theanalyst/wip-11910-hammer + + mon: pg ls is broken + + Reviewed-by: Kefu Chai + +commit bb12f925cf0d78f97eefc2e271e73596050b9919 +Merge: e801d4c e19f928 +Author: Loic Dachary +Date: Tue Aug 4 12:34:18 2015 +0200 + + Merge pull request #5384 from dachary/wip-12502-hammer + + rgw: need conversion tool to handle fixes following #11974 + + Reviewed-by: Yehuda Sadeh + +commit e801d4c943b8004ef613345505df91057913cd39 +Merge: 78a4024 154f18c +Author: Loic Dachary +Date: Tue Aug 4 11:22:08 2015 +0200 + + Merge pull request #5405 from ceph/wip-12465-hammer + + Log::reopen_log_file: take m_flush_mutex + + Reviewed-by: Loic Dachary + +commit 78a4024c14253503b770070aa36a090c6b8f1eaf +Merge: a451e88 7034720 +Author: Sage Weil +Date: Fri Jul 31 15:18:52 2015 -0400 + + Merge pull request #5121 from theanalyst/wip-11983-hammer + + FAILED assert(!old_value.deleted()) in upgrade:giant-x-hammer-distro-basic-multi run + + Reviewed-by: Sage Weil + +commit a451e882ab9a929d240747b4b09786ca4b4ce377 +Merge: 218f537 1063f52 +Author: Sage Weil +Date: Fri Jul 31 15:17:00 2015 -0400 + + Merge pull request #5269 from dachary/wip-12362-hammer + + stuck incomplete + + Reviewed-by: Sage Weil + +commit 218f537491a46a0251bef7690a7f5a86b988ee63 +Merge: 07fa83a 8abc46a +Author: Loic Dachary +Date: Fri Jul 31 20:55:48 2015 +0200 + + Merge pull request #5117 from theanalyst/wip-12099-hammer + + rgw: rados objects wronly deleted + + Reviewed-by: Yehuda Sadeh + +commit 07fa83aefc9bd4ea1495fa1e117a438b2c460e46 +Merge: a69d431 56c2688 +Author: Loic Dachary +Date: Fri Jul 31 20:55:29 2015 +0200 + + Merge pull request #5118 from theanalyst/wip-12042-hammer + + DragonDisk fails to create directories via S3: MissingContentLength + + Reviewed-by: Yehuda Sadeh + +commit a69d431ea52b5de7fc5cfe142ff3b69ff64e8048 +Merge: 5353480 c78cc00 +Author: Loic Dachary +Date: Fri Jul 31 20:55:14 2015 +0200 + + Merge pull request #5214 from SUSE/wip-12299-hammer + + RGW Swift API: support for 202 Accepted response code on container creation + + Reviewed-by: Yehuda Sadeh + +commit 5353480f1df297ad9dd1bd3154887ed2564f0280 +Merge: fb9156f a5dbcbb +Author: Loic Dachary +Date: Fri Jul 31 20:54:57 2015 +0200 + + Merge pull request #5226 from SUSE/wip-12322-hammer + + rgw: keystone does not support chunked input + + Reviewed-by: Yehuda Sadeh + +commit fb9156f412f83c0ce1fe4c5c9ccd57fd79f0c992 +Merge: 7193c16 b1618a9 +Author: Loic Dachary +Date: Fri Jul 31 20:54:41 2015 +0200 + + Merge pull request #5227 from SUSE/wip-12323-hammer + + RGW Swift API: XML document generated in response for GET on account does not contain account name + + Reviewed-by: Yehuda Sadeh + +commit 7193c16b65fdc1694b968899d23eae0638d89f11 +Merge: ac86490 e39dce7 +Author: Loic Dachary +Date: Fri Jul 31 20:54:26 2015 +0200 + + Merge pull request #5228 from theanalyst/wip-11872-hammer + + RGW does not send Date HTTP header when civetweb frontend is used + + Reviewed-by: Yehuda Sadeh + +commit ac86490821336ce024940d48d82f7a5ff7a302b1 +Merge: 33dbfc6 557865c +Author: Loic Dachary +Date: Fri Jul 31 20:54:02 2015 +0200 + + Merge pull request #5229 from theanalyst/wip-12242-hammer + + Fix tool for #11442 does not correctly fix objects created via multipart uploads + + Reviewed-by: Yehuda Sadeh + +commit 33dbfc6919840882c6cbc10dad2fc24cf0720bf9 +Merge: 99ca62f e50caab +Author: Loic Dachary +Date: Fri Jul 31 20:53:41 2015 +0200 + + Merge pull request #5237 from theanalyst/wip-12245-hammer + + rgw: empty json response when getting user quota + + Reviewed-by: Yehuda Sadeh + +commit 99ca62f2bf1e21a41cb7b6ecdb8a8731a18de195 +Merge: 1f5f319 2357b6c +Author: Loic Dachary +Date: Fri Jul 31 20:53:18 2015 +0200 + + Merge pull request #5284 from SUSE/wip-12398-hammer + + rgw: Properly respond to the Connection header with Civetweb + + Reviewed-by: Yehuda Sadeh + +commit 1f5f31905bb5f499a2db4a02993dbc6efa1c4251 +Merge: 5cbb6cf 9458b84 +Author: Loic Dachary +Date: Fri Jul 31 20:53:04 2015 +0200 + + Merge pull request #5285 from SUSE/wip-12399-hammer + + rgw: multipart list part response returns incorrect field + + Reviewed-by: Yehuda Sadeh + +commit 5cbb6cfb69aad0db470f99e39e33f4b4b1abfb95 +Merge: 1df93e1 e4b55b3 +Author: Loic Dachary +Date: Fri Jul 31 20:52:43 2015 +0200 + + Merge pull request #5286 from SUSE/wip-12400-hammer + + rgw: radosgw-admin dumps user info twice + + Reviewed-by: Yehuda Sadeh + +commit 1df93e19a0275ed218c8f83bc674f16d1856f241 +Merge: a48cbc0 2ecb3b7 +Author: Yan, Zheng +Date: Fri Jul 31 09:50:22 2015 +0800 + + Merge pull request #5427 from dachary/wip-12088-hammer-part-2 + + Fh ref count will leak if readahead does not need to do read from osd + +commit 2ecb3b7f4a49c574bc178a106c6bf0d8247f2a5e +Author: Zhi Zhang +Date: Wed Jul 22 10:54:53 2015 +0800 + + Fh ref count will leak if readahead does not need to do read from osd + + The 3c8cdeacf46ae4031189d2ef6948aa3b6ab4ae43 backport introduced a leak. + + http://tracker.ceph.com/issues/12319 Fixes: #12319 + + Signed-off-by: Zhi Zhang + +commit a48cbc0a847f19ea613b76a479acc831e9316c62 +Merge: 06c27cd 5ef0846 +Author: Loic Dachary +Date: Thu Jul 30 21:43:48 2015 +0200 + + Merge pull request #5120 from theanalyst/wip-11999-hammer + + cephfs Dumper tries to load whole journal into memory at once + + Reviewed-by: Greg Farnum + +commit 06c27cdd420598c497766ee5879335942a0acc09 +Merge: 19abe5e 408880b +Author: Loic Dachary +Date: Thu Jul 30 21:43:21 2015 +0200 + + Merge pull request #5119 from theanalyst/wip-12098-hammer + + kernel_untar_build fails on EL7 + + Reviewed-by: Greg Farnum + +commit 19abe5ee35c099c67b56ac268710fcd20bec60d3 +Merge: e3d17e4 4c199bf +Author: Loic Dachary +Date: Thu Jul 30 17:00:14 2015 +0200 + + Merge pull request #5417 from dachary/wip-11998-hammer + + debian/control: ceph-common (>> 0.94.2) must be >= 0.94.2-2 + + Reviewed-by: Sage Weil + +commit 4c199bf57dc54dc5e5f45cd9b34878a8459d434e +Author: Loic Dachary +Date: Thu Jul 30 09:43:20 2015 +0200 + + debian/control: ceph-common (>> 0.94.2) must be >= 0.94.2-2 + + The d8733be2ef8874b9a858a7ffddfb81b9b656e9a6 backport introduced a + regression by adding an incorrect Depends / Break combo supposed to + reflect the fact that ceph_argparse moved from ceph to ceph-common after + v0.94.2. It assumed the package is released under the 0.94.2 version + where in reality it is released under the 0.94.2-1xxx version (where xxx + is trusty, jessie etc.). + + The Depends / Break combo is changed to use 0.94.2-2 instead. + + See also http://tracker.ceph.com/issues/12529 for a larger discussion. + + http://tracker.ceph.com/issues/11998 Fixes: #11998 + + Signed-off-by: Loic Dachary + +commit e3d17e49731569ea92917f574d42d93258c77189 +Merge: cbba706 89aa8ff +Author: Loic Dachary +Date: Wed Jul 29 16:22:36 2015 +0200 + + Merge pull request #5248 from ceph/wip-11833-hammer + + mon: add an "osd crush tree" command + + Reviewed-by: Kefu Chai + +commit cbba7064c6cc4cde3e8a49c25ce671e91d31b9c7 +Merge: 8355bda 3c8cdea +Author: Loic Dachary +Date: Wed Jul 29 10:49:08 2015 +0200 + + Merge pull request #5222 from ceph/hammer-12088 + + client: reference counting 'struct Fh' + + Reviewed-by: John Spray + +commit 8355bdab56bc4e5ce4d20ba3486c082f06d8dcd1 +Merge: 52d0e5d ec70533 +Author: Orit Wasserman +Date: Tue Jul 28 23:33:18 2015 +0200 + + Merge pull request #5231 from theanalyst/wip-12243-hammer + + Civetweb RGW appears to report full size of object as downloaded when only partially downloaded + +commit 52d0e5da5ebad7fe42c2e469cea9773c7714c2b5 +Merge: 7fd31b1 03c07d7 +Author: Loic Dachary +Date: Tue Jul 28 22:40:23 2015 +0200 + + Merge pull request #5243 from theanalyst/wip-12239-hammer + + librbd/internal.cc: 1967: FAILED assert(watchers.size() == 1) + + Reviewed-by: Josh Durgin + +commit 7fd31b1b3c2c8e9dd3d9e5464775422215f7a4bc +Merge: 7230de3 5c812c1 +Author: Loic Dachary +Date: Tue Jul 28 22:40:03 2015 +0200 + + Merge pull request #5241 from theanalyst/wip-12238-hammer + + [ FAILED ] TestLibRBD.ExclusiveLockTransition + + Reviewed-by: Josh Durgin + +commit 7230de317736a71a5764cf224bd1309da1c7b3c6 +Merge: 6b6228f 7132277 +Author: Loic Dachary +Date: Tue Jul 28 22:30:23 2015 +0200 + + Merge pull request #5265 from SUSE/wip-12368-hammer + + linking ceph to tcmalloc causes segfault on SUSE SLE11-SP3 + + Reviewed-by: Loic Dachary + +commit 6b6228f8949e975cac763513898ea9704cb8baf1 +Merge: d62c3ea f99f312 +Author: Loic Dachary +Date: Tue Jul 28 22:27:40 2015 +0200 + + Merge pull request #5280 from ceph/wip-12384-hammer + + librbd: add valgrind memory checks for unit tests + + Reviewed-by: Loic Dachary + +commit d62c3ea344d9e49e9586867e872e8d5b3f019948 +Merge: 7b57ff8 b872882 +Author: Loic Dachary +Date: Tue Jul 28 22:26:25 2015 +0200 + + Merge pull request #5279 from ceph/wip-12237-hammer + + A client opening an image mid-resize can result in the object map being invalidated + + Reviewed-by: Loic Dachary + +commit 7b57ff8a9ced6c2f22456ed034cc83d07f82fbb3 +Merge: 481728a f819332 +Author: Loic Dachary +Date: Tue Jul 28 22:10:03 2015 +0200 + + Merge pull request #5283 from SUSE/wip-12397-hammer + + ceph.spec.in: 95-ceph-osd.rules, mount.ceph, and mount.fuse.ceph not installed properly on SUSE + + Reviewed-by: Loic Dachary + +commit 481728a04dd2c85096c3bc01cc37da9642b038ca +Merge: 54bb924 d8733be +Author: Loic Dachary +Date: Tue Jul 28 21:54:33 2015 +0200 + + Merge pull request #5206 from SUSE/wip-11998-hammer + + /usr/bin/ceph from ceph-common is broken without installing ceph + + Reviewed-by: Loic Dachary + +commit 54bb924e68ae2b4df65576a5d788d593b9d9e722 +Merge: e099058 c5c627f +Author: Loic Dachary +Date: Tue Jul 28 21:47:29 2015 +0200 + + Merge pull request #5055 from SUSE/wip-12044-hammer + + rgw/logrotate.conf calls service with wrong init script name + + Reviewed-by: Loic Dachary + +commit e0990583298277f1c631f7c2d2260d6c3fa64c9f +Merge: 8b93978 e149916 +Author: Loic Dachary +Date: Tue Jul 28 21:46:11 2015 +0200 + + Merge pull request #5040 from SUSE/wip-11964-hammer + + systemd: Increase max files open limit for OSD daemon + + Reviewed-by: Loic Dachary + +commit 8b93978881375d063fe2df8f40406ea650dda766 +Merge: 5a7cab2 22f58ce +Author: Loic Dachary +Date: Tue Jul 28 21:45:44 2015 +0200 + + Merge pull request #5038 from SUSE/wip-11876-hammer + + ceph-post-file fails on rhel7 + + Reviewed-by: Loic Dachary + +commit 5a7cab205bb1b3fdbf49a852cb978fc28eba8212 +Merge: 5218eff 38d36b1 +Author: Loic Dachary +Date: Tue Jul 28 21:45:25 2015 +0200 + + Merge pull request #5030 from SUSE/wip-12092-hammer + + packaging: add SuSEfirewall2 service files + + Reviewed-by: Loic Dachary + +commit 5218eff07c303fb2762ea9f38b9a9c23c24efcae +Merge: 0b54d50 8acfb99 +Author: Loic Dachary +Date: Tue Jul 28 21:38:14 2015 +0200 + + Merge pull request #5028 from SUSE/wip-12090-hammer + + rcceph script is buggy + + Reviewed-by: Loic Dachary + +commit 0b54d50ecd1445dfc1a46552adb83b9dae9210d9 +Merge: 45beb86 37d77d3 +Author: Loic Dachary +Date: Tue Jul 28 21:37:44 2015 +0200 + + Merge pull request #5026 from SUSE/wip-12087-hammer + + max files open limit for OSD daemon is too low + + Reviewed-by: Loic Dachary + +commit e19f928bd770a37f2f631c4cd796e2e30a494234 +Author: Yehuda Sadeh +Date: Fri Jun 26 16:56:28 2015 -0700 + + rgw: conversion tool to fix broken multipart objects + + Fixes: #12079 + + Broken multipart objects: multipart objects that created on 0.94.2 + and that start with underscore have a bad locator on their tail objects. + This extends the tool that was needed for older issue we've had with + hammer multipart objects (that start with underscore). The same usage + applies: + + $ radosgw-admin bucket check --check-head-obj-locator \ + --bucket= [--fix] + + Signed-off-by: Yehuda Sadeh + (cherry picked from commit f02ca6107172cecd80a490df9f0d66204e62326c) + +commit 28d32f6090724d62b6168d64031454f44eb4cc88 +Author: Yehuda Sadeh +Date: Fri Jun 26 13:49:55 2015 -0700 + + rgw: only scan for objects not in namespace + + Fixes: #11984 + The tool should only work on the head objects, and these are not inside + any namespace. + + Signed-off-by: Yehuda Sadeh + (cherry picked from commit 8103908548bf7d6c9fa47fb181cd450670bae8d6) + +commit e22e2b43b4039a44f5f8fbbe59edc21fbe118bdc +Author: Yehuda Sadeh +Date: Wed Apr 22 16:04:35 2015 -0700 + + rgw_admin: add --remove-bad flag to bucket check + + Add this flag so that the bad object will be removed (should be called + only after user has verified that objects content is correct). + + Signed-off-by: Yehuda Sadeh + (cherry picked from commit 06d67d9139a95b704b80de527381fd1bbf7981ce) + +commit 154f18ce3e52094fe84b058565a865ed97b079d6 (refs/remotes/gh/wip-12465-hammer) +Author: Samuel Just +Date: Fri Jul 24 15:38:18 2015 -0700 + + Log::reopen_log_file: take m_flush_mutex + + Otherwise, _flush() might continue to write to m_fd after it's closed. + This might cause log data to go to a data object if the filestore then + reuses the fd during that time. + + Fixes: #12465 + Backport: firefly, hammer + Signed-off-by: Samuel Just + (cherry picked from commit 8778ab3a1ced7fab07662248af0c773df759653d) + +commit b8728823493b9dfde0333fb41725002fc50e4d9b (refs/remotes/gh/wip-12237-hammer) +Author: Jason Dillaman +Date: Sun Jul 5 10:47:38 2015 -0400 + + librados_test_stub: read op should return number of bytes read + + Signed-off-by: Jason Dillaman + (cherry picked from commit f8a7b507983e31399831e802e99429b95386ed41) + +commit 7d9fce3aa3832a1b8bd7f18abd4745dbc0033582 +Author: Jason Dillaman +Date: Sun Jul 5 10:35:28 2015 -0400 + + tests: fixed TestObjectMap.InvalidateFlagInMemoryOnly + + librados and librados_test_stub return different result codes + for a read full object operation. + + Signed-off-by: Jason Dillaman + (cherry picked from commit 2ace2b77f8ed83e753fe4a48bcc997f5d1dd465f) + +commit 4a77be0a65c8b4ec3dc437721f8c321737b260de +Author: Jason Dillaman +Date: Sun Jul 5 11:09:09 2015 -0400 + + librbd: don't attempt to invalidate an object map in R/O mode + + The ImageWatcher is not initialized when in R/O mode, which + resulted in a NULL pointer dereference. + + Signed-off-by: Jason Dillaman + (cherry picked from commit 64d740f8fa10ba872e324ec2580a4d8c3f99a9ce) + +commit 0aea70f68b299441e692efdce6d5e7ff18b78c39 +Author: Jason Dillaman +Date: Tue Jun 23 11:17:12 2015 -0400 + + tests: add new unit tests for object map invalidation + + Signed-off-by: Jason Dillaman + (cherry picked from commit 0215e9753c09460f6fc84ded9397e36a209f2e32) + +commit c732cb889b4a61254d06703bf032082e56b196de +Author: Jason Dillaman +Date: Wed Mar 25 09:41:13 2015 -0400 + + librbd: move object map codes to common location + + These codes will need to be accessible from cls_rbd and librbd. + + Signed-off-by: Jason Dillaman + (cherry picked from commit 4ac584c34d576b489ed4c4862703b8fb427b3bc2) + +commit 27c99ea972a7b218ea591b208d0d1dd51eef6f95 +Author: Jason Dillaman +Date: Tue Jun 23 11:14:51 2015 -0400 + + librbd: only update image flags when holding exclusive lock + + It was possible for a client to open an image while another client + was shrinking an image. This would result in the former invalidating + the object map on-disk if it openned the image between updating the + image header and resizing the object map. + + Fixes: #11791 + Backport: hammer + Signed-off-by: Jason Dillaman + (cherry picked from commit eb81a6a7e391327ac993fd406443b206a7f7bffc) + +commit ef453630200ab72373f08357ca6b5ac5c5bbb397 +Author: Jason Dillaman +Date: Fri Jul 17 12:43:46 2015 -0400 + + librbd: new ImageWatcher::is_lock_supported method + + The new version does not attempt to acquire the snap_lock, to avoid + cases where a recursive lock would result. + + Signed-off-by: Jason Dillaman + +commit e4b55b398e68e870a7cf21276e63da2c4c6e3faa +Author: guce +Date: Sat Jul 11 14:08:33 2015 +0800 + + Fixes: #12286 radosgw-admin: after subuser modify print only once user info. + + remove rgw_admin.cc OPT_SUBUSER_MODIFY, show_user_info code block. + + switch (opt_cmd) { + ... + case OPT_SUBUSER_MODIFY: + show_user_info(info, formatter); //show first time (remove this) + break; + ... + } + + // output the result of a user operation + if (output_user_info) { + ... + show_user_info(info, formatter); //show second time + } + + test fix: + before: after subuser modify print twice user info. + after changes, do the same procedure, print only once user info. + + Signed-off-by: guce guce@h3c.com + (cherry picked from commit c604dd97fc179e5c2f640818c0f6e7cf99701947) + +commit 9458b845bf863ccf878873c4f0b089ddf84c7203 +Author: Henry Chang +Date: Wed Apr 22 18:26:45 2015 +0800 + + rgw: fix ListParts response + + The response XML element name should be 'ListPartsResult'. + + Fixes: #11494 + + Signed-off-by: Henry Chang + (cherry picked from commit caa9f0e461f1eed526fc43ee74699a7243aef9b8) + +commit 2357b6c808f4f7c5997af48149585a6051c04b8f +Author: Wido den Hollander +Date: Sat Jul 11 00:01:52 2015 +0200 + + rgw: If the client sends a Connection: close header respond accordingly. + + HTTP/1.1 assumes Keep-Alive by default, but if a Connection: close header is send + the server should respond with it as well. + + This makes the client close the connection after the request. + + Fixes: #12298 + (cherry picked from commit 79197d3711edc4b04a7ea4335b6e1b65754996d5) + +commit f819332e2826eae14849c5e68a380d1d87039d22 +Author: Nathan Cutler +Date: Thu Jul 9 21:38:46 2015 +0200 + + ceph.spec.in: install 95-ceph-osd.rules, mount.ceph, and mount.fuse.ceph properly on SUSE + + http://tracker.ceph.com/issues/12261 Fixes: #12261 + + Signed-off-by: Nathan Cutler + (cherry picked from commit 5ce38b9536efabf99a236c7a9d15c149fa4c16a6) + +commit d8733be2ef8874b9a858a7ffddfb81b9b656e9a6 +Author: Ken Dreyer +Date: Tue Apr 14 07:58:17 2015 -0600 + + debian: move ceph_argparse into ceph-common + + Prior to this commit, if a user installed the "ceph-common" Debian + package without installing "ceph", then /usr/bin/ceph would crash + because it was missing the ceph_argparse library. + + Ship the ceph_argparse library in "ceph-common" instead of "ceph". (This + was the intention of the original commit that moved argparse to "ceph", + 2a23eac54957e596d99985bb9e187a668251a9ec) + + http://tracker.ceph.com/issues/11388 Refs: #11388 + + Reported-by: Jens Rosenboom + Signed-off-by: Ken Dreyer + (cherry picked from commit 110608e5bdd9e2f03020ad41f0c2d756684d4417) + + Conflicts: + debian/ceph.install + There is no ceph_daemon.py in hammer + debian/control + Depends/Replaces/Breaks version adapted (from 9.0.0 to 0.94.2) + also adapted ceph-dbg Replaces/Breaks + +commit f99f3125ff76628e2525dca00bb7b983f941a08b (refs/remotes/gh/wip-12384-hammer) +Author: Zhiqiang Wang +Date: Fri Mar 20 16:15:42 2015 +0800 + + test: potential memory leak in FlushAioPP + + Should call the release function instead of deleting it to free + librbd::RBD::AioCompletion and librbd::AioCompletion. Otherwise there is + a potential memory leak. + + Signed-off-by: Zhiqiang Wang + (cherry picked from commit ada7ec860cb7901c560c12a5af36dc7c23051b76) + +commit a4fc63af630e77586e3ba2f17df3b6be4a1e2055 +Author: Jason Dillaman +Date: Tue Apr 28 15:25:49 2015 -0400 + + pybind: fix valgrind warning on rbd_get_parent_info call + + Signed-off-by: Jason Dillaman + (cherry picked from commit 2586e3ba1e20603a87c833513e09dae9281beb4d) + +commit aa3eb28f6be62991bc790de5c19cb7b6e30fa189 +Author: Jason Dillaman +Date: Tue Apr 28 11:12:00 2015 -0400 + + osdc: invalid read of freed memory + + The bytes not in cache stat was potentially reading the bh length + from a deleted bufferhead. + + Signed-off-by: Jason Dillaman + (cherry picked from commit 5ccc4422d6172376bd6f1be8d3a99c0a54eab807) + +commit 18ede754388372cf210d7db87fa46f3536cf0e44 +Author: Jason Dillaman +Date: Tue Apr 28 10:56:15 2015 -0400 + + krbd: fix incorrect types in the krbd API + + The C API functions were referencing the C++ CephContext + instead of the C rados_config_t. Additionally, the ceph + namespace was missing on the Formatter class. + + Signed-off-by: Jason Dillaman + (cherry picked from commit 740fd275a60630e60b3bcf41637a2ca486885d9c) + +commit 488578c1d557ebec7e50d53e45ed46f42984f4f8 +Author: Jason Dillaman +Date: Tue Apr 28 10:54:47 2015 -0400 + + fsx: cleanup crypto library at exit + + Also made small tweaks so that it can be compiled under + a C++ compiler. + + Signed-off-by: Jason Dillaman + (cherry picked from commit c44f8e7fbc19924a9453d8c032c624ebb6c0296f) + +commit 97ff6cb2f8fdd4d946eeab338ec225450e3ad8f3 +Author: Jason Dillaman +Date: Fri Apr 24 14:29:59 2015 -0400 + + tests: add run-rbd-valgrind-unit-tests.sh + + Signed-off-by: Jason Dillaman + (cherry picked from commit 5534faaa469b8a6a4c9687aad1a6723f3e859353) + +commit e690907cbb3b229f84f1e996d58636d00f823e8f +Author: Jason Dillaman +Date: Fri Apr 24 00:23:03 2015 -0400 + + valgrind: update valgrind suppressions for lttng-ust + + Signed-off-by: Jason Dillaman + (cherry picked from commit 8d87bdf597aad3d6be47aedd216a673bd9093a24) + +commit fe013e0a813c5697e917da642143388de60e8528 +Author: Jason Dillaman +Date: Fri Apr 24 00:21:15 2015 -0400 + + librbd: TaskFinisher should finish all queued tasks + + The destructor wasn't waiting for all Finisher tasks + to complete before stopping the thread. + + Signed-off-by: Jason Dillaman + (cherry picked from commit 8e20240e4155e2f0398e79f4c0095d2d6ba1d4cb) + +commit 43cd3ac923c9accfb81acf41f5bd12b8a05322c7 +Author: Jason Dillaman +Date: Thu Apr 23 23:10:23 2015 -0400 + + tests: fix valgrind errors with librbd unit test + + Signed-off-by: Jason Dillaman + (cherry picked from commit ed5472a10eb515e2a177a640c3f6ed929db9ee4f) + +commit 5d8d6a1a776f833847edc80d2a9b31ecb440ade5 +Author: Jason Dillaman +Date: Thu Apr 23 23:09:45 2015 -0400 + + tests: librbd should release global data before exit + + Signed-off-by: Jason Dillaman + (cherry picked from commit 6ab1bb5614a5d257a82cf8ea280eef5c90cf765b) + +commit 13f926e4e96d0b7178a9762bbbf589961dba47b7 +Author: Jason Dillaman +Date: Thu Apr 23 23:08:51 2015 -0400 + + librados_test_stub: cleanup singleton memory allocation + + Signed-off-by: Jason Dillaman + (cherry picked from commit 54c88255b74741d882b88f791497862635357634) + +commit 45beb86423c3bd74dbafd36c6822e71ad9680e17 +Merge: 5e399b0 582cf73 +Author: Loic Dachary +Date: Fri Jul 17 19:48:05 2015 +0200 + + Merge pull request #5046 from ceph/wip-12109-hammer + + librbd: new QA client upgrade tests + + Reviewed-by: Loic Dachary + +commit 1063f5275d1031812d564a1bd8ada64bed561026 +Author: Samuel Just +Date: Wed May 20 12:08:15 2015 -0700 + + PG::find_best_info: ignore info.les for incomplete peer + + See included update to doc/dev/osd_internals/last_epoch_started.rst + + Fixes: 11687 + Signed-off-by: Samuel Just + (cherry picked from commit 371d9baa120dc0302e9e61d3bc0e309dfaa773a0) + +commit 5e399b035d7cf861cf66a8ead00b388c4857cbb6 +Merge: 706b1c7 ad5745b +Author: Samuel Just +Date: Thu Jul 16 14:58:49 2015 -0700 + + Merge pull request #5159 from theanalyst/wip-11701-hammer + + make the all osd/filestore thread pool suicide timeouts separately configurable + + Reviewed-by: Samuel Just + +commit 713227791ab28c5e09073acb7b2c3c83ca0f0d6a +Author: Thorsten Behrens +Date: Mon Mar 16 00:13:38 2015 +0100 + + Conditional-compile against minimal tcmalloc. + + Certain older systems (SLE11 in this case) do not provide the full + tcmalloc functionality, due to e.g. incomplete libunwind + pieces. Use --with-tcmalloc-minimal to enable the cut-down + version. + + Here's how the various mem allocator switches interact now: + + --with-jemalloc: overrides --with-tcmalloc & --with-tcmalloc-minimal + --with-tcmalloc-minimal: overrides --with-tcmalloc + --with-tcmalloc: the default. use --without-tcmalloc to disable + + Signed-off-by: Thorsten Behrens + (cherry picked from commit c6f1c07113ca19547fdac10cd9b817a60142aee2) + +commit 706b1c7c5bcaaff96aa6950302b7aef097918d30 +Merge: daf5450 5e72479 +Author: Sage Weil +Date: Thu Jul 16 11:04:52 2015 -0400 + + Merge pull request #5252 from ceph/wip-12021-hammer + + OSDMonitor: allow addition of cache pool with non-empty snaps with co… + + Reviewed-by: Sage Weil + +commit daf5450765684b0b2ed049320d7463b637321e5a +Merge: d20f513 bd91fb0 +Author: Loic Dachary +Date: Thu Jul 16 01:15:51 2015 +0200 + + Merge pull request #4891 from theanalyst/wip-11740-hammer + + crush: take crashes due to invalid arg + + Reviewed-by: Loic Dachary + +commit ad5745bfd768b52ae6a766391232becad8587641 +Author: Samuel Just +Date: Wed May 6 10:49:00 2015 -0700 + + OSD: add command_wq suicide timeout + + Signed-off-by: Samuel Just + (cherry picked from commit df4e5de819c30003cfbe50a071c49039cf534419) + + Conflicts: + src/common/config_opts.h + Trivial merge conflict + +commit 059a579c02b312bbd32fa41485c361ae3847a3ba +Author: Samuel Just +Date: Wed May 6 10:54:31 2015 -0700 + + OSD: add remove_wq suicide timeout + + Signed-off-by: Samuel Just + (cherry picked from commit f2fbfa32a16666be46359f0eab7b04ca80a753f5) + +commit b8826bc4e3da6fcb9338ad6c01af1a88e6585a4d +Author: Samuel Just +Date: Wed May 6 10:52:40 2015 -0700 + + OSD: add scrub_wq suicide timeout + + Signed-off-by: Samuel Just + (cherry picked from commit 547a7041edc833f3cc8e04d388574809e30a8af6) + +commit 878dd403930a2058656a99c14b465358e134843c +Author: Samuel Just +Date: Wed May 6 10:51:28 2015 -0700 + + OSD: add snap_trim_wq suicide timeout + + Signed-off-by: Samuel Just + (cherry picked from commit e1073a4a577211672148a4112bd633831552d66f) + +commit 11575832a37ea247a8febe912b3058f51a464ab6 +Author: Samuel Just +Date: Wed May 6 10:50:19 2015 -0700 + + OSD: add recovery_wq suicide timeout + + Signed-off-by: Samuel Just + (cherry picked from commit 85311b656852af75bfbbc6699f92fc6aa233c316) + + Conflicts: src/common/config_opts.h + + There was a merge conflict due to introduction of `osd_recovery_sleep` + which was introduced in #3829 + +commit a82b4505848c09ad0094768c886f2015bdaa1148 +Author: Samuel Just +Date: Wed May 6 11:02:19 2015 -0700 + + OSD: add op_wq suicide timeout + + Signed-off-by: Samuel Just + +commit 89aa8ff9855ae868d59bd10fe3a3aab8517e90fc +Author: Kefu Chai +Date: Tue Jun 2 23:52:22 2015 +0800 + + mon: add an "osd crush tree" command + + * to print crush buckets/items in a tree + + Fixes: #11833 + Signed-off-by: Kefu Chai + (cherry picked from commit 5436c290f3622feb8d4b279ed6552b2510e0cee9) + + Conflicts: + src/test/mon/osd-crush.sh: + do not start mon as run() takes care of it already + +commit d20f513d9b185eff82bee2ca719b5453358e740b +Merge: 8753b2b 3d74164 +Author: Loic Dachary +Date: Tue Jul 14 20:43:02 2015 +0200 + + Merge pull request #4899 from theanalyst/wip-11911-hammer + + start_flush: filter out removed snaps before determining snapc's + + Reviewed-by: Samuel Just + +commit 8753b2b14536c34a7b6dec927c7a5b8100de7f68 +Merge: 3d72652 ecac1a4 +Author: Loic Dachary +Date: Tue Jul 14 20:42:45 2015 +0200 + + Merge pull request #4868 from SUSE/wip-11879-hammer + + Clock skew causes missing summary and confuses Calamari + + Reviewed-by: Samuel Just + +commit 3d72652d7ba6b2fff3d39ea7965c3c61d5fa0a04 +Merge: 9a79e8e fdb43eb +Author: Loic Dachary +Date: Tue Jul 14 16:42:12 2015 +0200 + + Merge pull request #4883 from SUSE/wip-11638-hammer + + ceph.spec.in: ceph-common subpackage def needs tweaking for SUSE/openSUSE + + Reviewed-by: Ken Dreyer + +commit 03c07d76ac8361ddd302f5bc0575aee7fb5edc99 +Author: Jason Dillaman +Date: Fri Jun 26 09:59:36 2015 -0400 + + librbd: assertion failure race condition if watch disconnected + + It's possible for librbd's watch of the header object to be reset by + connection issues just prior to the image being removed. This will + causes an assertion failure which assumes at least one watcher on the + image. + + Fixes: #12176 + Backport: hammer, firefly + Signed-off-by: Jason Dillaman + (cherry picked from commit af276de4f299960e43761904c043924cec5fef11) + +commit 5c812c1552d954f2c91c000332ddc74c9e91825a +Author: Jason Dillaman +Date: Thu Jun 25 16:51:31 2015 -0400 + + librbd: prevent object map updates from being interrupted + + Object map updates were being canceled in-flight when the exclusive lock + is released. This resulted in an ERESTART error code bubbling up to + AioRequest. + + Fixes: 12165 + Backport: hammer + Signed-off-by: Jason Dillaman + (cherry picked from commit 590cdc90edaf4f4ff06c97eb2f43b92ab9b60084) + + Conflicts: + src/librbd/ObjectMap.h + conflict due to a variable `m_snap_id' which was introduced in PR #4140 + which is dropped as we are not backporting that feature + +commit e50caab2251bb68fea1adbd17acc43aa98ab1206 +Author: wuxingyi +Date: Tue Jun 23 01:46:48 2015 +0000 + + rgw: fix empty json response when getting user quota + + Fixes: #12117 + Signed-off-by: wuxingyi + (cherry picked from commit 64fceed2202c94edf28b8315fe14c9affa8c0116) + +commit ec705336551436517c16bffdc6bf5467899ae4bb +Author: Yehuda Sadeh +Date: Thu Jun 25 14:31:03 2015 -0700 + + rgw: error out if frontend did not send all data + + Fixes: #11851 + The civetweb mg_write() doesn't return error when it can't flush all data + to the user, it just sends the total number of bytes written. Modified the + client io to return total number of bytes and return an error if didn't + send anything. + + Signed-off-by: Yehuda Sadeh + (cherry picked from commit daa679c3dd3770a6d5421e2cc9a36924f4550439) + +commit 557865c85bb907fe69248c4f1acb88320a7c1bb5 +Author: Yehuda Sadeh +Date: Wed Jun 17 15:11:28 2015 -0700 + + rgw: fix reset_loc() + + Fixes: #11974 + + Only need to set locator for underscore if namespace is empty + + Signed-off-by: Yehuda Sadeh + (cherry picked from commit d3bd27f4855df6bb207b656527138026af1a36a2) + +commit b1618a97fef644dc3dced502d600de6a5d55d085 +Author: Radoslaw Zarzynski +Date: Mon Apr 20 14:55:00 2015 +0200 + + rgw: fix lack of account name in XML listing of Swift account. + + Fixes: #11431 + Backport: hammer + Signed-off-by: Radoslaw Zarzynski + (cherry picked from commit 837388bbc39a1bf9019302c3a4d3a3fe22caeeb4) + +commit e39dce7935dd513b77ce34bc79d70a2c23437cbb +Author: Radoslaw Zarzynski +Date: Wed Feb 18 15:48:43 2015 +0100 + + rgw: generate the "Date" HTTP header for civetweb. + + Fixes: #10873 + Backport: hammer + Signed-off-by: Radoslaw Zarzynski + (cherry picked from commit ea384f83b601f60e135c3d3f960fdb75a919dd84) + +commit a5dbcbbdddce6cdeccb1e6f5641601d673cd1896 +Author: Hervé Rousseau +Date: Mon Apr 27 17:54:30 2015 +0200 + + Swift: Set Content-Length when requesting/checking Keystone tokens + + Running Keystone with WSGIChunkedRequest=On is not supported. + + We have to make sure that we set the Content-Length header when getting + an admin token and checking revoked tokens, otherwise Keystone returns + a HTTP 411 error. + + Same applies when checking revoked tickets. + + Fixes: #11473 + Backport: Hammer, Firefly + Signed-off-by: Hervé Rousseau + (cherry picked from commit 24f477417fdac9d68902fa211c8edf92a2e8729f) + +commit 3c8cdeacf46ae4031189d2ef6948aa3b6ab4ae43 +Author: Yan, Zheng +Date: Wed Jul 8 10:11:43 2015 +0800 + + client: reference counting 'struct Fh' + + The async readahead finisher needs to reference 'struct Fh'. But + it's possible user closes FD and free the corresponding 'struct Fh' + before async readahead finishes. + + Fixes: #12088 + Signed-off-by: Yan, Zheng + (cherry picked from commit 34b939a81d38173b882c429b28dedce778504ba8) + +commit c78cc00afb6deb8022db60dbe8649335f61bd345 +Author: Radoslaw Zarzynski +Date: Thu Mar 19 14:52:18 2015 +0100 + + rgw: rectify 202 Accepted in response for PUT on existing bucket. + + Fixes: #11148 + Backport: hammer + Signed-off-by: Radoslaw Zarzynski + (cherry picked from commit 3998fe7e02a6c25a3302c80a9c9907357fd3a23e) + +commit 9a79e8e7da5f34f1adaf6137e01bcd42766ae677 +Merge: 5527720 7f1c0cc +Author: Kefu Chai +Date: Sun Jul 12 02:19:57 2015 +0800 + + Merge pull request #5208 from tchaikov/wip-11975-hammer + + tests: TEST_crush_reject_empty must not run a mon + + Reviewed-by: Loic Dachary + +commit 7f1c0cc9cd3deab925440b56d82c3e61a8ba5ab1 +Author: Kefu Chai +Date: Sat Jul 11 23:04:33 2015 +0800 + + crush/CrushTester: return EINVAL if crushtool returns non-zero + + this backports a tiny part of ec02441, otherwise + CrushTester will return 1, and "ceph" cli will take it + as EPERM, which is miss leading, and fails + osd-crush.sh:TEST_crush_reject_empty. + + Signed-off-by: Kefu Chai + +commit 2aaeea145b24b972a0b98549c3527ccf98f4c96f +Author: Loic Dachary +Date: Fri Jul 10 16:23:47 2015 +0200 + + tests: TEST_crush_reject_empty must not run a mon + + * Back in Hammer, the osd-crush.sh individual tests did not run the + monitor, it was taken care of by the run() function. An attempt to run + another mon fails with: + + error: IO lock testdir/osd-crush/a/store.db/LOCK: Resource temporarily + unavailable + + This problem was introduced by cc1cc033930e8690a57674e842a003f6bbc7a242 + from https://github.com/ceph/ceph/pull/4936 + * replace test/mon/mon-test-helpers.sh with test/ceph-helpers.sh as + we need run_osd() in this newly added test + * update the run-dir of commands: ceph-helpers.sh use the different + convention for the run-dir of daemons. + + http://tracker.ceph.com/issues/11975 Refs: #11975 + + Signed-off-by: Loic Dachary + +commit 80afb81124a0d2ef25a23a12c86617ab1da3a4bd +Author: Loic Dachary +Date: Tue Apr 21 16:11:33 2015 +0200 + + ceph-helpers: implement test_expect_failure + + To display the output in case the command did not fail with the expected + output. + + Signed-off-by: Loic Dachary + (cherry picked from commit 5871781b10ff0b26c731b70d1898c474006cbee3) + +commit 6b5e9a1df7dfb3a971e40aec35119ec019515b69 +Author: Loic Dachary +Date: Wed Jun 10 23:16:01 2015 +0200 + + tests: display the output of failed make check runs + + After a make check fails, it shows a summary but not the output of the + failed tests although they contain information to diagnose the problem. + + Set the VERBOSE=true automake variable which is documented to collect + and display the failed script output at the end of a run (the content of + the test-suite.log file (valid from automake-1.11 up). + + http://www.gnu.org/software/automake/manual/automake.html#index-VERBOSE + + Also remove the run-make-check.sh that did the same in a way that is not + compatible with automake-1.11. + + Signed-off-by: Loic Dachary + (cherry picked from commit 3a55cb029bb7db9542d2b14f2deda90feb0ae0f6) + +commit 552772025cb8d5f51ffb3a069d1bd93bc73f1123 +Merge: f4d77c2 1440122 +Author: Loic Dachary +Date: Fri Jul 10 10:17:24 2015 +0200 + + Merge pull request #4889 from theanalyst/wip-11484-hammer + + OPT_INT option interprets 3221225472 as -1073741824, and crashes in Throttle::Throttle() + + Reviewed-by: Kefu Chai + +commit f4d77c22aa51edb45211e080f3fdf28a7a0cfdd4 +Merge: 5088105 a62c3aa +Author: Loic Dachary +Date: Fri Jul 10 10:16:25 2015 +0200 + + Merge pull request #4776 from tchaikov/wip-11279-hammer + + ceph: cli interactive mode does not understand quotes + + Reviewed-by: Kefu Chai + +commit 5088105300c013b1b804c938a30ac63ba710556d +Merge: e3b1f7b 0b6d442 +Author: Loic Dachary +Date: Fri Jul 10 10:14:42 2015 +0200 + + Merge pull request #4657 from ceph/wip-hammer-11535-admin-socket + + common/admin_socket: close socket descriptor in destructor + + Reviewed-by: Loic Dachary + +commit e3b1f7be9e8474fbec98076790ff683bccd44ce9 +Merge: dd29a86 558d639 +Author: Kefu Chai +Date: Fri Jul 10 16:07:48 2015 +0800 + + Merge pull request #4687 from SUSE/wip-7387-hammer + + utf8 and old gcc breakage on RHEL6.5 + + Reviewed-by: Kefu Chai + +commit dd29a869db5503fc9e2c6d1d44ee4311d95af20c +Merge: 7f1fb57 0e5e7e1 +Author: Kefu Chai +Date: Fri Jul 10 16:00:00 2015 +0800 + + Merge pull request #5122 from theanalyst/wip-11982-hammer + + ceph fails to compile with boost 1.58 + + Reviewed-by: Kefu Chai + +commit 7f1fb574608800c3e6aa12df6c7888acbf397a52 +Merge: adc7016 5141301 +Author: Kefu Chai +Date: Fri Jul 10 15:59:35 2015 +0800 + + Merge pull request #4936 from ceph/wip-11975-hammer + + mon crashes when "ceph osd tree 85 --format json" + + Reviewed-by: Kefu Chai + +commit adc70161d14fc2b51e6c6f38580f76ff0067717a +Merge: 2d68db8 82988d6 +Author: Loic Dachary +Date: Fri Jul 10 09:48:44 2015 +0200 + + Merge pull request #4892 from theanalyst/wip-11760-hammer + + ceph-disk: get_partition_type fails on /dev/cciss... + + Reviewed-by: Loic Dachary + +commit 2d68db8371263645642cf28473deea4456ca7021 +Merge: 1cffe8c ba1a016 +Author: Loic Dachary +Date: Fri Jul 10 09:43:57 2015 +0200 + + Merge pull request #4877 from SUSE/wip-11902-hammer + + admin/build-doc: script fails silently under certain circumstances + + Reviewed-by: Loic Dachary + +commit 51413011417b76f5ad2830d9f93fbfe78c77e467 +Author: Kefu Chai +Date: Tue May 26 18:11:59 2015 +0800 + + mon: add "--check" to CrushTester::test_with_crushtool() + + so we don't need to call CrushTester::check_name_maps() in OSDMonitor.cc + anymore. + + Fixes: #11680 + Signed-off-by: Kefu Chai + (cherry picked from commit c6e634875316cf17368d497e6dc4f6f4b5dd65d2) + +commit 5ec27cf589b4535f07e28a86bd304f7a46427ac4 +Author: Kefu Chai +Date: Tue May 26 17:51:50 2015 +0800 + + crushtool: rename "--check-names" to "--check" + + * because "--check" also checks for the max_id + + Note: edited since we do not have the fix introduced in 46103b2 in + hammer. + + Signed-off-by: Kefu Chai + (cherry picked from commit 9381d53acdce85fcbff828926b911e050ba36e51) + +commit 2a8fe8862a15342cc5716c146487d0b42af0fbf6 +Author: Kefu Chai +Date: Tue May 26 16:58:23 2015 +0800 + + mon: check the new crush map against osdmap.max_osd + + Fixes: #11680 + Signed-off-by: Kefu Chai + (cherry picked from commit 22e6bd6e01d5df3f3e897562597e22ca1737f8c8) + +commit c0b0f52ddbd4e22998a36addfb32f27614183e19 +Author: Kefu Chai +Date: Tue May 26 15:35:10 2015 +0800 + + crushtool: enable check against max_id + + add an argument "max_id" for "--check-names" to check if any item + has an id greater or equal to given "max_id" in crush map. + + Note: edited since we do not have the fix introduced in 46103b2 in + hammer. + + Signed-off-by: Kefu Chai + (cherry picked from commit d0658dd3cdf072b2a7c2a1986f8785a697c591ee) + +commit f041bbebf98a2aff2ad542e8d0c12c46af427573 +Author: Kefu Chai +Date: Tue May 26 15:34:33 2015 +0800 + + crush/CrushTester: check if any item id is too large + + Signed-off-by: Kefu Chai + (cherry picked from commit e640d89240017956b8c7411babb86be0f1e2b172) + +commit cc1cc033930e8690a57674e842a003f6bbc7a242 +Author: Kefu Chai +Date: Mon May 25 20:14:32 2015 +0800 + + mon: validate new crush for unknown names + + * the "osd tree dump" command enumerates all buckets/osds found in either the + crush map or the osd map. but the newly set crushmap is not validated for + the dangling references, so we need to check to see if any item in new crush + map is referencing unknown type/name when a new crush map is sent to + monitor, reject it if any. + + Fixes: #11680 + Signed-off-by: Kefu Chai + (cherry picked from commit a955f36a509e5412b1f72632a1a956d99e768e35) + +commit ff29a7f9dd21505c681881e609183aed9ac3250e +Author: Kefu Chai +Date: Tue May 26 12:08:36 2015 +0800 + + crushtool: add the "--check-names" option + + * so one is able to verify that the "ceph osd tree" won't chock on the + new crush map because of dangling name/type references + + Signed-off-by: Kefu Chai + (cherry picked from commit d6b46d4c7b722945ce24ac2930381a109b1e3dda) + +commit 960ea49699f421ceb89c9e0c9430378a35f09a9a +Author: Kefu Chai +Date: Tue May 26 12:08:09 2015 +0800 + + crush/CrushTester: add check_name_maps() method + + * check for dangling bucket name or type names referenced by the + buckets/items in the crush map. + * also check for the references from Item(0, 0, 0) which does not + necessarily exist in the crush map under testing. the rationale + behind this is: the "ceph osd tree" will also print stray OSDs + whose id is greater or equal to 0. so it would be useful to + check if the crush map offers the type name indexed by "0" + (the name of OSDs is always "OSD.{id}", so we don't need to + look up the name of an OSD item in the crushmap). + + Signed-off-by: Kefu Chai + (cherry picked from commit b75384d73958faf81d45847a7dfa56f4fa347e6f) + +commit 1cffe8c46a39142ee0da4e2279eda7276df262e1 +Merge: 6ffb1c4 ef6641c +Author: Loic Dachary +Date: Fri Jul 10 09:39:15 2015 +0200 + + Merge pull request #4667 from SUSE/wip-11611-hammer + + ceph.spec: update OpenSUSE BuildRequires + + Reviewed-by: Loic Dachary + +commit 5e72479b0882ac13597d7a613698e583dcb2b932 +Author: Samuel Just +Date: Tue Jul 7 11:43:01 2015 -0700 + + OSDMonitor: allow addition of cache pool with non-empty snaps with config + + We need to be able to allow the version of ceph_test_* from earlier + versions of ceph to continue to work. This patch also adjusts the + work unit to use a single rados snap to test the condition without + --force-nonempty to ensure that we don't need to be careful about + the config value when running that script. + + Signed-off-by: Samuel Just + +commit 6ffb1c4ae43bcde9f5fde40dd97959399135ed86 +Merge: c7ebf96 524f4a5 +Author: Gregory Farnum +Date: Wed Jul 8 16:52:12 2015 +0100 + + Merge pull request #5123 from theanalyst/wip-11979-hammer + + MDSMonitor: handle MDSBeacon messages properly + + Reviewed-by: Kefu Chai + +commit ecac1a458bc7331ed8d667f20ba31995d74892d3 +Author: Thorsten Behrens +Date: Fri Feb 6 01:26:40 2015 +0100 + + Always provide summary for non-healthy cluster + + This fixes a problem, wherein calamari does not provide + popup drill-downs for warnings or errors, should the summary + be missing. + + Calamari gets health info from /api/v1/cluster/$FSID/health. + If the data here has a summary field, this summary is provided + in a popup window: + + /api/v1/cluster/$FSID/health is populated (ultimately) with + status obtained via librados python bindings from the ceph + cluster. In the case where there's clock skew, the summary + field supplied by the ceph cluster is empty. + + No summary field, no popup window with more health details. + + Signed-off-by: Thorsten Behrens + (cherry picked from commit eaf6e0cf48488fe604d0ef0db164d44948d4e8d4) + +commit c7ebf96a9a4a6143b112c8606d5ee346fb800cec +Merge: b163728 1a321e4 +Author: Loic Dachary +Date: Wed Jul 8 15:36:38 2015 +0200 + + Merge pull request #4862 from SUSE/wip-11874-hammer + + Bucket header is enclosed by quotes + + Reviewed-by: Yehuda Sadeh + +commit b1637289ec4ff99d923457577893b4c4a8d2e9fe +Merge: e33af22 54f4e7d +Author: Loic Dachary +Date: Wed Jul 8 15:35:53 2015 +0200 + + Merge pull request #4885 from theanalyst/wip-11755-hammer + + Object copy bug + + Reviewed-by: Yehuda Sadeh + +commit e33af22dec32467f229ca2dc1dc9668702a44ce8 +Merge: 5696b0f 9dfef60 +Author: Loic Dachary +Date: Wed Jul 8 15:34:45 2015 +0200 + + Merge pull request #4884 from theanalyst/wip-11722-hammer + + Keystone PKI token expiration is not enforced + + Reviewed-by: Yehuda Sadeh + +commit 5696b0ff2a981ae8afe72df796ba7d7da47bb7d1 +Merge: 72ecd52 ed5442b +Author: Loic Dachary +Date: Wed Jul 8 15:30:48 2015 +0200 + + Merge pull request #4875 from ceph/wip-11770-hammer + + librbd: aio calls may block + + Reviewed-by: Jason Dillaman + Reviewed-by: Loic Dachary + +commit 1a32379dd6cb56ed69b8f448d3612506c8131fbe +Author: Kefu Chai +Date: Mon May 11 19:30:30 2015 +0800 + + mon/PGMap: add more constness + + Signed-off-by: Kefu Chai + (cherry picked from commit e1f1c56ce88ef3ad05e881d650fac637931ce3fe) + +commit 84ebc3d320bdf871ccf85e555951cea421b56021 +Author: Kefu Chai +Date: Mon May 11 19:29:13 2015 +0800 + + mon/PGMap: sort pg states by the states in "pg ls" spec + + Signed-off-by: Kefu Chai + (cherry picked from commit 990dfb6cd45438bc8293ac37882daa413860a2f8) + +commit e310461aecc667cf26806bc5a3bbabb05696bdfc +Author: Kefu Chai +Date: Mon May 11 17:02:41 2015 +0800 + + mon: s/recovery/recoverying/ in "pg ls*" commands' spec + + * also translate "repair" if specified as "states" + * update test_mon_pg in cephtool-test-mon.sh + + Fixes: #11569 + Signed-off-by: Kefu Chai + (cherry picked from commit 89f89ca3477eddcae11a05fbd58a8f3658eb1fc1) + +commit 524f4a52d115ecda8cd7793d0f8bea148eff92af +Author: Kefu Chai +Date: Fri May 15 22:50:36 2015 +0800 + + mon: always reply mdsbeacon + + the MDS (Beacon) is always expecting the reply for the mdsbeacon messages from + the lead mon, and it uses the delay as a metric for the laggy-ness of the + Beacon. when it comes to the MDSMonitor on a peon, it will remove the route + session at seeing a reply (route message) from leader, so a reply to + mdsbeacon will stop the peon from resending the mdsbeacon request to the + leader. + + if the MDSMonitor re-forwards the unreplied requests after they are + outdated, there are chances that the requests reflecting old and even wrong + state of the MDSs mislead the lead monitor. for example, the MDSs which sent + the outdated messages could be dead. + + Fixes: #11590 + Signed-off-by: Kefu Chai + (cherry picked from commit b3555e9c328633c9e1fbc27d652c004b30535e5b) + +commit 413e407dea446bebb9c36abb3025ada450dd4fe9 +Author: Kefu Chai +Date: Tue Jun 2 23:20:21 2015 -0700 + + mon/MDSMonitor: rename labels to a better name + + * s/ignore/reply/ + * s/out/ignore/ + + Signed-off-by: Kefu Chai + (cherry picked from commit f00ecb8b3df73ce6337985bc6d43bce5143ee537) + +commit a03968ad584a3ff8e351cc5dce053e535fcdc454 +Author: Kefu Chai +Date: Tue Jun 2 12:55:06 2015 +0800 + + mon: send no_reply() to peon to drop ignored mdsbeacon + + so the peon can remove the ignored mdsbeacon request from the + routed_requets at seeing this reply, and hence no longer resend the + request. + + Fixes: #11590 + Signed-off-by: Kefu Chai + (cherry picked from commit 72a37b3a8e145d8522ea67fc14ce2c5510b6852b) + +commit 39f34596b0ec6f769f507e2b372204f8551f7ee0 +Author: Kefu Chai +Date: Tue Jun 2 12:22:26 2015 +0800 + + mon: remove unnecessary error handling + + msg.get_session() should always return a non-zero pointer in + Monitor.dispatch() + + Signed-off-by: Kefu Chai + (cherry picked from commit 16e8e2cc82a90c49cd8aa3d0e3acc4694ba659a0) + +commit 0e5e7e1d259579571c1fc05660f6af8e295e733b +Author: Kefu Chai +Date: Fri May 8 15:21:20 2015 +0800 + + mon: remove unused variable + + * as a side effect, this change silences + http://tracker.ceph.com/issues/11576 + + Fixes: #11576 + Signed-off-by: Kefu Chai + (cherry picked from commit e7b196a4a091c0ea258866559ba06e7ed0cc4247) + +commit 70347209260688f1a067354c744569499adb6920 +Author: Samuel Just +Date: Wed Apr 1 16:37:51 2015 -0700 + + ReplicatedPG::finish_promote: handle results->snaps is empty case + + If results->snaps winds up empty after filtering removed snaps, + we need to treat the object as if we had gotten an ENOENT. + + PartialFix: #11296 + Backport: firefly, hammer + Signed-off-by: Samuel Just + (cherry picked from commit 6150757dbe0fa11cceb14460865b859a7c8164c7) + +commit 3e44dc16ed3fbda053996e44826aa3d90042a234 +Author: Samuel Just +Date: Wed Apr 1 16:25:22 2015 -0700 + + ReplicatedPG::finish_promote: fix snap promote head snaps + + If the snaps vector is: 10=[9,5,2]:[4]+head, the backing pool's snaps + vector is 3=[2]:[]+head, and we request clone 4 from the backing pool, + the backing pool will send us head with an empty results->snaps vector. + Actually, clone 4 should be trimmed, but the cache pool does not know + that. Thus, we should construct an empty snaps vector for that clone. + + PartialFix: #11296 + Backport: firefly, hammer + Signed-off-by: Samuel Just + (cherry picked from commit a45a698372def1623323470c6a1c4eb70e0bb79f) + +commit 5ef08466abf1b3934fcad0a0ca46f3a4380d6dbd +Author: John Spray +Date: Wed Jun 3 10:04:26 2015 +0100 + + tools: chunk reads in Dumper + + Previously tried to read entire journal + into memory in one go, which was problematic + for large journals. + + Fixes: #11746 + Signed-off-by: John Spray + (cherry picked from commit e3ddcb894ad09326698999d42de0ce3feb69f28e) + +commit 408880bed296e5cbf05864fa6744a5b00a245272 +Author: Greg Farnum +Date: Tue Jun 16 08:13:41 2015 -0700 + + qa: update to newer Linux tarball + + This should make newer gcc releases happier in their default configuration. + kernel.org is now distributing tarballs as .xz files so we change to that + as well when decompressing (it is supported by Ubuntu Precise so we should + be all good). + + Fixes: #11758 + + Signed-off-by: Greg Farnum + (cherry picked from commit 1ea3f47ab806d48ca7b045c2731d344eae3900e1) + +commit 56c2688b87d7d78831f8e147fc67cc0651ab644c +Author: Yehuda Sadeh +Date: Fri Apr 24 14:45:40 2015 -0700 + + rgw: simplify content length handling + + Signed-off-by: Yehuda Sadeh + (cherry picked from commit e97fd5052cab83c5f699531a8c960b93437a8f9f) + +commit d9bbef3e470c6406bb65dc40e7e9c08c5d599f73 +Author: Robin H. Johnson +Date: Fri Apr 24 10:49:16 2015 -0700 + + rgw: make compatability deconfliction optional. + + Per request from Yehuda, the deconfliction for having both + HTTP_CONTENT_LENGTH and CONTENT_LENGTH set is now optional, and + controlled by new configuration boolean, which defaults to false. + rgw content length compat + + X-URL: https://github.com/ceph/ceph/pull/4436#issuecomment-95994887 + Signed-off-by: Robin H. Johnson + (cherry picked from commit 79d17af1a1ec0659884f768945a7bac6282b5e0b) + +commit 0260abd5d265bd63ea9c89f4082c31ba1c5ae8fa +Author: Robin H. Johnson +Date: Wed Apr 22 12:53:06 2015 -0700 + + rgw: improve content-length env var handling + + The FastCGI specification, section 6.3 on Authorizers, describes a case + where HTTP_CONTENT_LENGTH will be set in the environment and + CONTENT_LENGTH will NOT be set. + + Further documention in the code. + + Fixes: #11419 + Signed-off-by: Robin H. Johnson + (cherry picked from commit 3e38eab44bfb082fdd2b6f29b8b0357f8f5c11bb) + +commit 8abc46a157e4c1431a92a1e52ab694dccff5d514 +Author: wuxingyi +Date: Wed Jun 10 06:57:57 2015 +0000 + + rgw: fix data corruption when race condition + + We should delete the object in the multipart namespace lastly to prevent a previous upload + wrongly deleting objects belong to the following upload. + + Fixes: #11749 + Signed-off-by: wuxingyi + (cherry picked from commit ac1e729a75b5d995028bbc223bcf5ecce0d112cc) + +commit 72ecd522941156c8a7e5303531944b0735dcbeb8 +Merge: 59f37a9 d723e11 +Author: Abhishek L +Date: Wed Jul 1 18:09:46 2015 +0530 + + Merge pull request #4886 from theanalyst/wip-11737-hammer + + Reviewed-by: Greg Farnum + +commit 59f37a9bafc095181b3f41ec5d93ac58e2cda604 +Merge: 53a2143 89d0266 +Author: Kefu Chai +Date: Sat Jun 27 17:11:11 2015 +0800 + + Merge pull request #5095 from ceph/wip-fix-doc-hammer + + doc: fix doc build + + Reviewed-by: Loic Dachary + +commit 89d0266a60729d5d9747867e8c30abf71a891231 +Author: Kefu Chai +Date: Sat Jun 27 14:44:55 2015 +0800 + + doc: add the corresponding @endcond command for @cond + + * they are used to applease asphyxiate, as it + is not able to handle "enum" sections + + Signed-off-by: Kefu Chai + +commit 2aa77b33a97e5a3ca134c9c555aa6e7a69ef50f7 +Author: Kefu Chai +Date: Sat Jun 27 14:43:01 2015 +0800 + + doc: remove orphan: directive in ceph-create-keys.rst + + * it is used to silence the sphinx warning, but conf.py + does not like it. + + Signed-off-by: Kefu Chai + +commit ad66e40e8bd598da7c9738cb44abb543008c90a3 +Author: Kefu Chai +Date: Sat Jun 27 14:41:59 2015 +0800 + + doc: let doxygen ignore src/tracing + + Signed-off-by: Kefu Chai + +commit 53a2143eb7e549de1185b01ed0bde841ffa5235a +Merge: 6f7cd04 bfb1442 +Author: Samuel Just +Date: Fri Jun 26 14:19:40 2015 -0700 + + Merge pull request #4902 from theanalyst/wip-11908-hammer + + Fixes for rados ops with snaps + + Reviewed-by: Samuel Just + +commit 6f7cd0460d5729c15966119e0177ddc56a361d8e +Merge: 78d894a 356bd2c +Author: Loic Dachary +Date: Thu Jun 25 20:03:42 2015 -0400 + + Merge pull request #5069 from dachary/wip-11806-hammer + + ceph / ceph-dbg steal ceph-objecstore-tool from ceph-test / ceph-test-dbg + + Reviewed-by: Ken Dreyer + +commit 356bd2c68ca730e766d06c46a0364784f5d72275 +Author: Loic Dachary +Date: Wed Jun 24 14:58:47 2015 -0400 + + debian: ceph-dbg steals ceph-objectstore-tool from ceph-test-dbg (take 2) + + 968573b8930a7c8485bf53e3a989ce2f7d0a2fff incorrectly backported + 6f11fbf41fab10924b1e0e41fcf27864779d4073. It should instead reflect that + ceph-dbg in 0.94.2 and above will break ceph-test-dbg in all versions + prior to 0.94.2. + + In other words, 0.94-XXXX being lower than 0.94.1, upgrading from + 0.94.1 to 0.94.2 will not notice that ceph-dbg breaks ceph-test-dbg. + + $ dpkg --compare-versions 0.94-XXXX lt 0.94.1 && echo yes || echo no + yes + $ dpkg --compare-versions 0.94.2 lt 0.94.1-xxx && echo yes || echo no + no + + http://tracker.ceph.com/issues/11806 Fixes: #11806 + + Signed-off-by: Loic Dachary + +commit c5c627fdd3de669ee94ab7ecd6939c29a4a68011 +Author: wuxingyi +Date: Wed Mar 11 17:34:40 2015 +0800 + + rgw/logrotate.conf: Rename service name + + The service name for ceph rados gateway was changed to "ceph-radosgw", + the previous version of service name "radosgw" would cause a failed reload, + and finally make it impossible to write any log data to the log file. + + Signed-off-by: wuxingyi + (cherry picked from commit 9df3f798179481fe8ae6ae873dcb793de7d8f367) + +commit 582cf731e05baabd2cd79567af89d7f005c24925 (refs/remotes/gh/wip-12109-hammer) +Author: Jason Dillaman +Date: Tue May 5 13:08:21 2015 -0400 + + tests: add librbd watch/notify version compatibility test + + Fixes: #11405 + Backport: hammer + Signed-off-by: Jason Dillaman + (cherry picked from commit 41e4cbe032e32762e3a9e8bc3eff8ece19f91a54) + +commit 43b9aef11c1281d8b83b659a523dba365d79add4 +Author: Jason Dillaman +Date: Tue May 5 11:22:55 2015 -0400 + + qa/workunits/rbd: add notify_master/slave bootstrap scripts + + Backport: hammer + Signed-off-by: Jason Dillaman + (cherry picked from commit 124b1d3d2d0dbd685bbd717856d29b316e62e660) + +commit f995fb50d3d7734161fa498db9e204eaded651b6 +Author: Jason Dillaman +Date: Tue May 5 10:34:48 2015 -0400 + + qa/workunits/rbd: add new test_librbd_api workunit + + This only tests the public librbd API for use during upgrade tests. + + Backport: hammer + Signed-off-by: Jason Dillaman + (cherry picked from commit 9039955f42d01044cfcf20c56ca2181e51c317ee) + +commit a09da2a2c6c12c6bf2f8e17f2096b3ab24587007 +Author: Jason Dillaman +Date: Tue May 5 10:27:38 2015 -0400 + + tests: create librbd API-only integration test suite + + The QA client-upgrade test suite requires a librbd test + that is dynamically linked to librbd. Since the current + ceph_test_librbd includes tests against the non-public API, + it is statically linked against librbd and thus cannot be + used to test a client upgrade scenario. + + Backport: hammer + Signed-off-by: Jason Dillaman + (cherry picked from commit 6fe94c8cbb924c31c1007e2d0c76f28ce9dbca57) + +commit e149916cd40a58ce5db5f63a2b4efd82a8c4e1ba +Author: Owen Synge +Date: Tue Apr 21 11:31:24 2015 +0200 + + Increase max files open limit for OSD daemon. + + Under heavy load the number of file descriptors opened + by the OSD can go beyond the 64K file limit. This patch + increases the default to 128K. + + Signed-off-by: Owen Synge + (cherry picked from commit ebda4ba1c67172852587e47a8e6fb538809a1b1c) + +commit 22f58ce2665f1ea6b38f6016dc585202a4595322 +Author: Sage Weil +Date: Mon Jun 1 15:10:14 2015 -0700 + + Makefile: install ceph-post-file keys with mode 600 + + Otherwise ssh (may) prompt for a password. + + Signed-off-by: Sage Weil + (cherry picked from commit 106a1c3081d02446aa3d8e13865da0c3393bae90) + +commit 3e65a10bd2351744da199a4b076659191b4378a2 +Author: Joseph McDonald +Date: Mon Jun 1 15:05:32 2015 -0700 + + ceph-post-file: improve check for a source install + + Signed-off-by: Joseph McDonald + (cherry picked from commit ee170eadcdcb4b54d36a7d474558484de9d917eb) + +commit c1f6743940250b04ae6dbea30d8805571add39b6 +Author: Joseph McDonald +Date: Mon Jun 1 15:00:39 2015 -0700 + + ceph-post-file: behave when sftp doesn't take -i + + Fixes: #11836 + Signed-off-by: Joseph McDonald + (cherry picked from commit b84031ed5eaace1222e14d3c4076a3ab1155da96) + +commit 38d36b1174160ad104704aaa7ff5290d1ec3e782 +Author: Tim Serong +Date: Thu Apr 30 10:55:38 2015 +1000 + + packaging: move SuSEfirewall2 templates out of src + + Better to have static config like this that needs to be installed in + /etc in a separate subdirectory of the project. + + Signed-off-by: Tim Serong + (cherry picked from commit 70292658d5febb6c5f94af4df9c3e93551772d12) + +commit 24bc9f2c58ca78c8023935b8b69f8b01adbc11af +Author: Tim Serong +Date: Wed Apr 29 13:12:38 2015 +1000 + + packaging: add SuSEfirewall2 service files + + This adds SuSEfirewall2 service files for Ceph MON, OSD and MDS, for use + on SLES and openSUSE. The MON template opens port 6789 and the OSD/MDS + template opens the range 6800-7300 as per + http://docs.ceph.com/docs/master/rados/configuration/network-config-ref/ + + Signed-off-by: Tim Serong + (cherry picked from commit 77685f5b787c56bcb1c4d9f1e058e25312fa62fe) + +commit 8acfb994f22efa07beeecccda300cbd50d683566 +Author: Owen Synge +Date: Thu May 7 12:02:41 2015 +0200 + + Bug fix to ceph systemV compatability script. + + Was failing with more than one OSD / MON deamon on a single node. + Fixes suse bugzilla #927862 + + Signed-off-by: Owen Synge + (cherry picked from commit dfda3ff8741fcdbac3150456ca7614cf75ef1776) + +commit bd3fd929e6f95e6d9840317aa3ac02f52711c94b +Author: Owen Synge +Date: Thu Jun 18 14:16:03 2015 +0200 + + Fixes to rcceph script + + - only start OSDs if mon daemons are also present + - adds support for mask and unmask + - removes support for cluster with non default cluster name, + as this was very limited and inconsistent + - Reapplied from a patch as could not cherry-pick + 66cb46c411d874be009c225450eea5021cf1219b from Mon Jan 12 + as this produced issues with src/gmock + + Signed-off-by: Owen Synge + (cherry picked from commit bfa0c4a626fdbb2bf978ccfab783ac428156144b) + +commit 37d77d3680d2c8adda35eddf9625a331a45ece11 +Author: Owen Synge +Date: Tue Apr 21 11:31:24 2015 +0200 + + Increase max files open limit for OSD daemon. + + Under heavy load the number of file descriptors opened + by the OSD can go beyond the 64K file limit. This patch + increases the default to 128K. + + Signed-off-by: Owen Synge + (cherry picked from commit ebda4ba1c67172852587e47a8e6fb538809a1b1c) + +commit 78d894a634d727a9367f809a1f57234e5e6935be +Author: Sage Weil +Date: Wed Jun 17 09:35:28 2015 -0700 + + qa/workunits/rados/test-upgarde-v9.0.1: fix exclude syntax + + It's -, then a list of all exclusions separated by :. There are just 2. + + Signed-off-by: Sage Weil + +commit 3e8d60a80ce31860eac76a1f6489a35e1795a0c0 +Author: Sage Weil +Date: Tue Jun 16 21:05:29 2015 -0700 + + qa/workunits/rados/test-upgrade-v9.0.1: skip one more evict test + + Signed-off-by: Sage Weil + +commit 348a3d3c9880e7d022e71a2faafe51c8f771406e +Author: Josh Durgin +Date: Mon Jun 15 15:12:43 2015 -0700 + + qa: add compatibility filtered rados api tests for upgrades + + Post-9.0.1, the evict op returns success when an object doesn't exist + in the cache tier. Skip the tests that are incompatible across + versions. + + Fixes: #11548 + Signed-off-by: Josh Durgin + +commit d2b80966b8f74de818a671c90b4c821a4b0782db +Merge: 70bba62 f68bf94 +Author: Sage Weil +Date: Mon Jun 15 12:38:57 2015 -0700 + + Merge pull request #4961 from ceph/wip-11493-hammer + + backport 11493 fixes, and test, prevetning ec cache pools + + Reviewed-by: Samuel Just + +commit f68bf94e370fb11a3047ec2762a972a0b7a7c0bb +Author: Samuel Just +Date: Fri May 15 13:05:40 2015 -0700 + + OSDMonitor: disallow ec pools as tiers + + Fixes: 11650 + Signed-off-by: Samuel Just + (cherry picked from commit 11b7801bb57cb25cd2d26d58722d49691747725b) + +commit 13c8d58da1303cc68d99da19f79d625f91f99d43 +Author: Sage Weil +Date: Wed Apr 29 12:34:25 2015 -0700 + + mon: prevent pool with snapshot state from being used as a tier + + If we add a pool with snap state as a tier the snap state gets clobbered + by OSDMap::Incremental::propogate_snaps_to_tiers(), and may prevent OSDs + from starting. Disallow this. + + Include a test. + + Fixes: #11493 + Backport: hammer, giant, firefly + Signed-off-by: Sage Weil + (cherry picked from commit bbec53edf9e585af4e20bbc9ba9057d6fdfda342) + +commit 58e62662f6ef04ac76470090d1d958467e34194a +Author: Samuel Just +Date: Fri May 8 10:26:48 2015 -0700 + + test/librados/tier.cc: destroy and recreate cache pool on every test + + Namespaces are not sufficient with the checks for 11493 in the mon. + + Signed-off-by: Samuel Just + (cherry picked from commit bef09e0cdb274cb1c87335a2af9ee532d14a4596) + +commit 70bba6226a64090dcf41cd90b23fdf5aed8cd0ca (refs/remotes/jashan/hammer) +Merge: 3b6977b 5a60a03 +Author: Loic Dachary +Date: Mon Jun 15 17:26:07 2015 +0200 + + Merge pull request #4846 from SUSE/wip-11862-hammer + + missing man pages for ceph-create-keys, ceph-disk-* + + Reviewed-by: Kefu Chai + Reviewed-by: Ken Dreyer + +commit 3b6977b706dbc99cac25bec1b71a628c398c6ff1 +Merge: 5fb8561 3db1026 +Author: Sage Weil +Date: Thu Jun 11 13:54:35 2015 -0700 + + Merge pull request #4934 from dachary/wip-releases-hammer + + doc/release-notes: v0.94.2 + +commit 3db1026f3706e6f5a5c25013cb6646a0298057d8 +Author: Sage Weil +Date: Wed Jun 10 12:48:41 2015 -0700 + + doc/release-notes: v0.94.2 + + Signed-off-by: Sage Weil + (cherry picked from commit 306345b29c259ab04a58ed5d40f801645485b29d) + + Conflicts: + doc/release-notes.rst + +commit ed5442b1057dcc4fb1f9404c805dabe2bbde2252 (refs/remotes/jashan/wip-11770-hammer, refs/remotes/gh/wip-11770-hammer) +Author: Jason Dillaman +Date: Tue Jun 2 10:33:35 2015 -0400 + + tests: verify librbd blocking aio code path + + Signed-off-by: Jason Dillaman + (cherry picked from commit 4cf41486e9c9e1efcb863960a8f3e0326ffca7e5) + + Conflicts: + src/test/librbd/test_librbd.cc: trival resolution + +commit 20e104869f3d17ce672438144700a4d984d487b4 +Author: Jason Dillaman +Date: Mon Jun 1 22:56:11 2015 -0400 + + librbd: new rbd_non_blocking_aio config option + + Setting this option to false reverts librbd to legacy behavior + where AIO operations could potentially block. + + Signed-off-by: Jason Dillaman + (cherry picked from commit 769cad12716b85d87eacc1069dd9f5c21cad3915) + +commit b4571b3e238efc39767f753e0ec1622c8bd6d6e6 +Author: Jason Dillaman +Date: Thu Apr 9 20:34:28 2015 -0400 + + PendingReleaseNotes: document changes to librbd's aio_read methods + + Signed-off-by: Jason Dillaman + +commit 9ea1edd0ca9e385f823ad04b05bc887d77aa5136 +Author: Jason Dillaman +Date: Thu Apr 9 13:33:09 2015 -0400 + + librbd: AioRequest::send no longer returns a result + + The librados calls used by AioRequest::send should always return + zero unless there is a bug. + + Signed-off-by: Jason Dillaman + (cherry picked from commit c77bce3311ab62892eb8c1d883263ba7ed663b20) + + Conflicts: + src/librbd/AioRequest.cc: trivial resolution + src/librbd/AsyncFlattenRequest.cc: trivial resolution + +commit 272df2aed79a95dd9c45db4e0953e9b321f7b0f5 +Author: Jason Dillaman +Date: Wed Apr 8 21:55:36 2015 -0400 + + tests: update librbd AIO tests to remove result code + + Signed-off-by: Jason Dillaman + (cherry picked from commit 948b15eb52fd5d9ce842fa12ee0cecda17353b01) + + Conflicts: + src/test/librbd/test_internal.cc: trivial resolution + src/test/librbd/test_librbd.cc: trivial resolution + +commit dd2e4c13ff6d88edb25f90af62af16ba825c15c9 +Author: Jason Dillaman +Date: Wed Apr 8 21:37:50 2015 -0400 + + librbd: internal AIO methods no longer return result + + All failures should be returned via the AioCompletion. + + Signed-off-by: Jason Dillaman + (cherry picked from commit 9ab42d613128ab08c688ddbea93df4c95068b9cd) + + Conflicts: + src/librbd/AioRequest.cc: trivial resolution + src/librbd/internal.cc: trivial resolution + +commit dbd4e293d7124c89a22148e8fa5f425a995c900c +Author: Jason Dillaman +Date: Wed Apr 8 21:48:21 2015 -0400 + + Throttle: added pending_error method to SimpleThrottle + + Allow the client of SimpleThrottle to detect an async error + so that it can exit early. + + Signed-off-by: Jason Dillaman + (cherry picked from commit b88b88c5df91325fb713c2031a56bffe421268e0) + +commit 7df6091a30b1b94d764240262195e971175554b3 +Author: Jason Dillaman +Date: Wed Apr 8 20:18:50 2015 -0400 + + librbd: add new fail method to AioCompletion + + Helper method to handle passing fatal errors generated within + librbd (not from the OSDs) back to the client. + + Signed-off-by: Jason Dillaman + (cherry picked from commit 6d1d0c867855a96bee4c13a0c0a39a0e002ccd12) + +commit cf6e1f50ea7b5c2fd6298be77c06ed4765d66611 +Author: Jason Dillaman +Date: Wed Apr 8 19:06:52 2015 -0400 + + librbd: avoid blocking AIO API methods + + Enqueue all AIO API methods within the new librbd thread pool to + reduce the possibility of any blocking operations. To maintain + backwards compatibility with the legacy return codes of the API's + AIO methods, it's still possible to block attempting to acquire + the snap_lock. + + Fixes: #11056 + Signed-off-by: Jason Dillaman + (cherry picked from commit 3a7b5e30efdb21aa1a0aeb68a5d02a1ac2a5faf3) + +commit e61974aed09a3f81e1f65a4bbaed43e3f22b27b4 +Author: Jason Dillaman +Date: Wed Apr 8 17:24:08 2015 -0400 + + librbd: add task pool / work queue for requests + + Signed-off-by: Jason Dillaman + (cherry picked from commit afb896d91f886b647baf38f7ec94cc3739f6d2a9) + + Conflicts: + src/librbd/ImageCtx.cc: trivial resolution + src/librbd/ImageCtx.h: trivial resolution + +commit bfb144268b803340efad29cd6c627b170ea32402 +Author: Samuel Just +Date: Wed May 20 16:10:02 2015 -0700 + + ReplicatedPG::release_op_ctx_locks: requeue in scrub queue if blocked + + Otherwise we can reorder an op around another op which got blocked by a + scrub which started after the first blocked on an obc. + + Fixes: #11691 + Signed-off-by: Samuel Just + (cherry picked from commit be873eb8da7b29ecefaa5a99b88de7ddcca711ee) + +commit c7b6a6370a69149ea94f9e35d536aa90f06e7659 +Author: Samuel Just +Date: Tue May 19 10:56:11 2015 -0700 + + ReplicatedPG::finish_ctx: take excl lock if operation is rw + + Fixes: #11677 + Signed-off-by: Samuel Just + (cherry picked from commit 5c2b795724423ed484ab451de855ddcfc085342b) + +commit 1550a569dab120ce28396fe365565e8e4acd9801 +Author: Samuel Just +Date: Thu May 21 12:13:43 2015 -0700 + + RadosModel: randomly prefix delete with assert_exists + + Signed-off-by: Samuel Just + (cherry picked from commit 4fe7d2abdff2fce359e5e992206644cc03825ee0) + +commit 4cdc5f7d6b3ec488c79c09cb44a43d4d9398b74c +Author: Samuel Just +Date: Thu May 21 11:36:42 2015 -0700 + + RadosModel: assert exists on subsequent writes + + Signed-off-by: Samuel Just + (cherry picked from commit 121aa3bc612b86281535ac3bcfe98bc99bc99ace) + +commit 25c730bda74b94f2c894c508ab09988dbd528c4e +Author: Samuel Just +Date: Tue May 19 10:23:01 2015 -0700 + + test/librados/snapshots.cc: add test for 11677 + + Signed-off-by: Samuel Just + (cherry picked from commit c2d17b927f8a222164b3bf2922a4ff337696f566) + +commit 3d74164d3d6caaa5099abd9a1d1920482d3e05c2 +Author: Samuel Just +Date: Wed May 27 11:14:15 2015 -0700 + + ReplicatedPG::trim_object: write filtered snapset while we're at it + + If we trimmed an object, we might as well remove the obsolete snaps + as well. + + Signed-off-by: Samuel Just + (cherry picked from commit 90eb7768f99ea249952df195a844a3a7c9a59b78) + +commit a1161540bc0094a951021d4ca651b95ec045213e +Author: Samuel Just +Date: Wed May 27 11:00:54 2015 -0700 + + ReplicatedPG: start_flush: use filtered snapset + + Otherwise, we might send our deletes based on deleted snaps. This is + problematic since we may have trimmed the clones to which those snaps + belong, causing us to send them at an earlier snap than we used before. + + The specific situation was + + 78:[78, 70, 63, 5a, 58, 57]:[64(63), 58(58, 57)] + + with 58 already clean. To flush 64, we send: + + delete@58 + delete@59 + copyfrom@62 + + Then, snap 63 is trimmed leaving us with a snapset of: + + 78:[78, 70, 63, 5a, 58, 57]:[58(58, 57)] + + since trim_object doesn't filter the head object snapset snaps. This + isn't really a bug since in general all snapset users must be aware + that there may be trimmed snaps in snapset::snaps. However, here + it becomes a problem when we go to flush head: + + delete@58 -- ignored due to snapc + delete@59 -- ignored due to snapc + copyfrom@78 -- not ignored + + The base pool head is at snap seq 62, so it clones that value into + clone 78(78, 70) instead of forgetting it. What should have happened + is that we should have based our flushes on filtered snapset: + + 78:[78, 70, 58, 57]:[58(58, 57)] + + Causing us to instead send: + + delete@58 -- ignored due to snapc + delete@69 -- not ignored, causes no clone to be made + copyfrom@78 -- not ignored, updates head such that a subsequent clone + will leave 70 out of the clone snaps vector. + + Fixes: 11787 + Signed-off-by: Samuel Just + (cherry picked from commit 6051e255ac062985ada1989edb7f23cd750915e2) + +commit 82988d611bad6226138b94590275faadbca3554a +Author: islepnev +Date: Fri Apr 17 22:33:01 2015 +0300 + + ceph-disk: support NVMe device partitions + + Linux nvme kernel module v0.9 enumerate devices as following: + + /dev/nvme0 - characted revice + /dev/nvme0n1 - whole block device + /dev/nvme0n1p1 - first partition + /dev/nvme0n1p2 - second partition + + http://tracker.ceph.com/issues/11612 Fixes: #11612 + + Signed-off-by: Ilja Slepnev + (cherry picked from commit 9b62cf254d02d30609793be8b1cb8a94f38891f1) + +commit bd91fb027ab91d487b1d61d25516c13590735d89 +Author: Sage Weil +Date: Tue May 12 16:37:56 2015 -0700 + + mon: prevent bucket deletion when referenced by a rule + + If a rule references a bucket with 'take', prevent deletion. + + Fixes: #11602 + Signed-off-by: Sage Weil + (cherry picked from commit 3d591afef90b0601572c748f13faac029d05f5a0) + +commit 56565ee1cdb06a7705d1c3f26f5592b10399324a +Author: Sage Weil +Date: Tue May 12 14:03:49 2015 -0700 + + crush: fix crash from invalid 'take' argument + + Verify that the 'take' argument is a valid device or bucket. Otherwise, + ignore it (do not add the value to the working vector). + + Backport: hammer, firefly + Fixes: #11602 + Reported-by: shiva rkreddy + Signed-off-by: Sage Weil + (cherry picked from commit 9324d0a1af61e1c234cc48e2175b4e6320fff8f4) + +commit 1440122d61d5b0a3f8360f4e2101db1018109799 +Author: Kefu Chai +Date: Wed Apr 29 03:28:18 2015 -0700 + + common/config: detect overflow of float values + + Signed-off-by: Kefu Chai + (cherry picked from commit 1ff409ef8d022a1a84d034bd3db976c4d769e993) + +commit 9b947fa320b77e0055a581005353c2561a12a198 +Author: Kefu Chai +Date: Wed Apr 29 15:41:08 2015 +0800 + + common/config: detect overflow of int values + + * #include "strtol.h" in strtol.cc, to ensure the function defintions + are consistent. + * add a test accordingly + * fix the testcase of converting 1024E. + * do not accept integers overflow after adding SI suffix + * do not accept integers underflow (i.e. negative values) + + Fixes: #11484 + Signed-off-by: Kefu Chai + (cherry picked from commit d62f80dc7b25d312ff05b65b7be854aae15b66a8) + +commit d723e1156e70a492d633e43b86e7c373e5750065 +Author: Yan, Zheng +Date: Tue May 12 14:52:30 2015 +0800 + + mds: clear CDir::STATE_REJOINUNDEF after fetching dirfrag + + Fixes: #11541 + Signed-off-by: Yan, Zheng + (cherry picked from commit ab1e5394dc778f6799472bd79a4d9ba7197107c2) + +commit 54f4e7d4a534448293c74612f8140cf34b9cf9f8 +Author: Javier M. Mellid +Date: Fri May 15 14:22:29 2015 +0200 + + rgw: Use attrs from source bucket on copy + + On copy objects, when bucket source is the same as the destination, use attrs + from source bucket. + + Fixes: #11639 + + Signed-off-by: Javier M. Mellid + (cherry picked from commit 1dac80df1d4a2364154ed8b404d13609936c257b) + +commit 9dfef6004d208af3730634796aad199391707826 +Author: Anton Aksola +Date: Fri Apr 10 13:25:21 2015 +0300 + + rgw: always check if token is expired + + Fixes: #11367 + + Currently token expiration is only checked by the token cache. With PKI + tokens no expiration check is done after decoding the token. This causes + PKI tokens to be valid indefinitely. UUID tokens are validated by + keystone after cache miss so they are not affected by this bug. + + This commit adds explicit token expiration check to + RGWSwift::validate_keystone_token() + + Signed-off-by: Anton Aksola + Reported-by: Riku Lehto + (cherry picked from commit 2df069390ea3bbcfbab5022750e89f51d197cc11) + +commit fdb43ebe2b1211acdb454836a64dbd589feeef45 +Author: Nathan Cutler +Date: Fri May 15 21:43:34 2015 +0200 + + ceph.spec.in: tweak ceph-common for SUSE/openSUSE + + ceph-common needs python-argparse in SUSE/openSUSE and + needs redhat-lsb-core only in RHEL/CentOS/Fedora. + + http://tracker.ceph.com/issues/11638 Fixes: #11638 + + Signed-off-by: Nathan Cutler + (cherry picked from commit 363d957d8fdd15a1674befbd8e485fd89b76d716) + +commit ba1a0167cc809081eda7e6cc9ecfb971e439d696 +Author: John Spray +Date: Wed Jun 3 10:09:09 2015 +0100 + + admin/build-doc: fix dependency checks + + http://tracker.ceph.com/issues/11857 Fixes: #11857 + + Signed-off-by: John Spray + (cherry picked from commit 539c1ba7211f579bad4f59ae824f1e68e620ecbd) + + Conflicts: + admin/build-doc + Insert lines at appropriate spot + +commit 5a60a034bf3015eaf468e5e3f9d8feb08b8fdd95 +Author: Nathan Cutler +Date: Wed Jun 3 10:41:27 2015 +0200 + + man/ceph-create-keys.8: add missing file + + This is the generated manpage, taken from a local build of master + + Signed-off-by: Nathan Cutler + (cherry picked from commit bcda61fcbe07ee36cd5172d80018f287591660ec) + +commit 19305b897ac147602049752a2dfbe1fd39ba562c +Author: Kefu Chai +Date: Tue Apr 21 14:59:32 2015 +0800 + + doc: add ceph-create-keys.8 + + Fixes: #10725 + Signed-off-by: Kefu Chai + (cherry picked from commit 27cee2f1f46a9f47cda9dfeb56ff1259e982960c) + + Conflicts: + doc/man/8/ceph-create-keys.rst + Includes fixes from https://github.com/ceph/ceph/pull/4855 + +commit ffd0933dcc790d7cedc1048b664bf4e8c40464a3 +Author: Jason Dillaman +Date: Mon May 11 17:05:49 2015 -0400 + + WorkQueue: added virtual destructor + + Signed-off-by: Jason Dillaman + (cherry picked from commit b3f5a75332c058816dc39b71e9d2b36e752159f4) + +commit a28adfbdd8abc86e7766c303bc610c0c252910f7 +Author: Jason Dillaman +Date: Wed Apr 8 16:46:34 2015 -0400 + + WorkQueue: add new ContextWQ work queue + + The queue holds a collection of Context pointers that will + be completed by the thread pool. + + Signed-off-by: Jason Dillaman + (cherry picked from commit 24a33e977f7b71962adeeb48f75d488a76e70fa9) + +commit 1a321e477effa23bffbb6cb057aa955586197345 +Author: Wido den Hollander +Date: Wed Jun 3 13:13:33 2015 +0200 + + rgw: Do not enclose the Bucket header in quotes + + Fixes: #11860 + Signed-off-by: Wido den Hollander + (cherry picked from commit 8af25faed93fe02d3dad585b8579ce8b041cc4e6) + +commit a62c3aa1df9e0f79ac75d94083d10b902dbba382 +Author: Kefu Chai +Date: Fri Apr 24 14:04:30 2015 +0800 + + ceph.in: handle unknown Exception correctly + + * in case parse_cmdargs() throws + + Signed-off-by: Kefu Chai + (cherry picked from commit 7789eefc01deb9ca7fe90f5521aece3e36c3c350) + +commit cc7f7441c1d0457e840bfdc3413f9bc6d5f5134a +Author: Kefu Chai +Date: Fri Apr 24 01:27:44 2015 +0800 + + ceph.in: improve the interactive mode + + * if ceph is not reading from a tty, expect EOF instead of "quit" + as the end of input. + * do not panic at seeing the EOF + * update the test case test_mon_injectargs_SI(). since we disables + "ceph injectargs " in a458bd83, in which the arguments + of "injectargs" are supposed to be consumed by "tell" instead. + so "ceph injectargs ..." is taken as an incomplete command, and + this command will bring ceph cli into the interactive mode, + redirecting its stdin to /dev/null helps ceph cli quit the loop, + but in a way of throwing EOFError exception. this change handles + the EOF, so the "ceph injectargs ..." does not throws anymore. + but the side effect is that the test fails since it expects a + non-zero return code. so replace it with an equivalent "tell" + command which also fails but due to the non-SI postfix. + + Signed-off-by: Kefu Chai + (cherry picked from commit da9d2b4077ab1dceeed979ab71f0d9ed59b14266) + +commit eb26388cb985456e8f6e07a52a20fa912cf4efaa +Author: Kefu Chai +Date: Fri Apr 24 00:50:37 2015 +0800 + + ceph.in: parse quote correctly in interactive mode + + Fixes: #11279 + Signed-off-by: Kefu Chai + (cherry picked from commit bc7d8c99d2e16a141a8b575281ba12c67628dac3) + +commit 558d6391ecdfd5d716558341ce5d84ce1f7ec9c3 +Author: Kefu Chai +Date: Fri May 8 12:01:16 2015 +0800 + + json_sprit: fix the FTBFS on old gcc + + Fixes: #11574 + Signed-off-by: Kefu Chai + (cherry picked from commit 6b68b27146852f057a3373cd04b08cd6917f3eea) + +commit 678b3e6082729698ce3575ba70313dd8399b1aed +Author: Tim Serong +Date: Sat May 2 01:59:53 2015 +1000 + + json_spirit: use utf8 intenally when parsing \uHHHH + + When the python CLI is given non-ASCII characters, it converts them to + \uHHHH escapes in JSON. json_spirit parses these internally into 16 bit + characters, which could only work if json_spirit were built to use + std::wstring, which it isn't; it's using std::string, so the high byte + ends up being zero'd, leaving the low byte which is effectively garbage. + + This hack^H^H^H^H change makes json_spirit convert to utf8 internally + instead, which can be stored just fine inside a std::string. + + Note that this implementation still assumes \uHHHH escapes are four hex + digits, so it'll only cope with characters in the Basic Multilingual + Plane. Still, that's rather a lot more characters than it could cope + with before ;) + + (For characters outside the BMP, Python seems to generate escapes in the + form \uHHHHHHHH, i.e. 8 hex digits, which the current implementation + doesn't expect to see) + + Fixes: #7387 + + Signed-off-by: Tim Serong + (cherry picked from commit 8add15b86e7aaef41397ab8fa9e77ee7957eb607) + +commit ef6641c0aa47c7f559aa56d7c35a5815afc2ba49 +Author: Loic Dachary +Date: Mon Mar 23 21:26:23 2015 +0100 + + ceph.spec: update OpenSUSE BuildRequires + + OpenSUSE 13.2 needs libsnappy-devel but not bzip2-devel. + + Signed-off-by: Loic Dachary + (cherry picked from commit bdac3dc3fb189688af52f60b0b34339dd8fea251) + +commit 0b6d4427ac7234fd509a9541adf45e303e40cb16 (refs/remotes/jashan/wip-hammer-11535-admin-socket, refs/remotes/gh/wip-hammer-11535-admin-socket) +Author: Jon Bernard +Date: Fri May 8 11:54:06 2015 -0400 + + common/admin_socket: close socket descriptor in destructor + + Long-running processes that do not reuse a single client connection will + see accumulating file descriptors as a result of not closing the + listening socket. In this case, eventually the system will reach + file-max and subsequent connections will fail. + + Fixes: #11535 + + Signed-off-by: Jon Bernard + (cherry picked from commit 88fabb1ee6cd3c7d211ca46919055195e32245db)