Tasks #10560
Updated by Loïc Dachary over 9 years ago
* already merged in https://github.com/ceph/ceph/tree/dumpling ** d73f0b86d3989d7b5924e984f949734a64ab04a9 commit:d73f0b86d3989d7b5924e984f949734a64ab04a9 If trusty, use older version of qemu ** 81891f26ab71acd457057c0ab2a0a2b2766e1cd1 commit:81891f26ab71acd457057c0ab2a0a2b2766e1cd1 mds: store backtrace for straydir ** 547bfcfd8b75bb415454d28c9aaaddd67a7c9d92 commit:547bfcfd8b75bb415454d28c9aaaddd67a7c9d92 librbd: don't close an already closed parent image upon failure *** http://tracker.ceph.com/issues/10030 ** acc3bf6cc9ccc1e2ab096dc7f4b7a6151f624dcd commit:acc3bf6cc9ccc1e2ab096dc7f4b7a6151f624dcd librbd: protect list_children from invalid child pool IoCtxs *** http://tracker.ceph.com/issues/10123 ** d6af0e4c0e6419738b520f9d24487d38c272c4d7 commit:d6af0e4c0e6419738b520f9d24487d38c272c4d7 ceph-disk: use update_partition in prepare_dev and main_prepare *** http://tracker.ceph.com/issues/9721 ** 6de6f57562e8dc8da488dc94a0693e8f6b1fd166 commit:6de6f57562e8dc8da488dc94a0693e8f6b1fd166 ceph-disk: run partprobe after zap *** http://tracker.ceph.com/issues/9665 ** f64d742e4f1ac64914975284817fd734c0f4b7f3 commit:f64d742e4f1ac64914975284817fd734c0f4b7f3 ceph-disk: encapsulate partprobe / partx calls ** 6e3b6f0e3f5acfb25b8c59bb9e049b4f600d8d29 commit:6e3b6f0e3f5acfb25b8c59bb9e049b4f600d8d29 common: Add cctid meta variable *** http://tracker.ceph.com/issues/6228 ** 62c75ef633a7347a9d964cb2ae2adc40a7038a15 commit:62c75ef633a7347a9d964cb2ae2adc40a7038a15 PGLog: allow for empty pg log in update_range if log_tail == eversion_t() *** http://tracker.ceph.com/issues/10168 ** e17d579f9cc35393b7f1f6626f6732430bc4a2f6 commit:e17d579f9cc35393b7f1f6626f6732430bc4a2f6 qa: allow small allocation diffs for exported rbds *** http://tracker.ceph.com/issues/10002 ** 496e561d81f2dd1bf92d588fc3afc2431e0a5b98 commit:496e561d81f2dd1bf92d588fc3afc2431e0a5b98 ReplicatedPG: don't move on to the next snap immediately *** http://tracker.ceph.com/issues/9487 ** 25fb957ed7576ffde2aa6568f62f89757d3ecd0b commit:25fb957ed7576ffde2aa6568f62f89757d3ecd0b osd: initialize purged_snap on backfill start; restart backfill if change *** http://tracker.ceph.com/issues/9487 ** 01efc44adea6a16bed24fb10b67fe0cbd6a7595c commit:01efc44adea6a16bed24fb10b67fe0cbd6a7595c ReplicatedPG: do not queue the snap trimmer constantly *** http://tracker.ceph.com/issues/9113 ** 1f9961221f7cbec2814612f444dc669f94bed69d commit:1f9961221f7cbec2814612f444dc669f94bed69d ReplicatedPG: clean out completed trimmed objects as we go *** http://tracker.ceph.com/issues/9113 ** fabd4b5795ce9ec6b6fb960f97f496d642c989da commit:fabd4b5795ce9ec6b6fb960f97f496d642c989da client: cast m->get_client_tid() to compare to 16-bit Inode::flushing_cap_tid *** http://tracker.ceph.com/issues/9869 ** 309c74d847f88f906b42bbd12af6c79cc59e645b commit:309c74d847f88f906b42bbd12af6c79cc59e645b CrushWrapper: pick a ruleset same as rule_id *** http://tracker.ceph.com/issues/9675 *** splitting a pool doesn't start when rule_id != ruleset_id ** 06a62dbb552981f86e58d83c58eeb254d633e74a commit:06a62dbb552981f86e58d83c58eeb254d633e74a qa/workunits/rbd/import_export.sh: case insensitive grep ** 3f6630e6b3ec2932df15c7c532de04217dbd3c1b commit:3f6630e6b3ec2932df15c7c532de04217dbd3c1b test/cli-integration/rbd: hide stderr ** 1db47d8052638a7d72d7af68be2491aa042686ef commit:1db47d8052638a7d72d7af68be2491aa042686ef mds: Locker: remove unused variable ** 1dbda2ec996fc928b86f714bc1af49cd7811c334 commit:1dbda2ec996fc928b86f714bc1af49cd7811c334 mds: Locker: fix a NULL deref in _update_cap_fields ** 9be014c7fe40ff7f0a28f22ba0ef6e857c288fc8 commit:9be014c7fe40ff7f0a28f22ba0ef6e857c288fc8 ReplicatedPG::do_op: remove unused backfill_target_info ** 50f676143ea23cb66de02197ffdfc01fb1ac4e69 commit:50f676143ea23cb66de02197ffdfc01fb1ac4e69 osd: Fix problems in ReplicatedPG::do_op() logic ** a081531110ee09a41208f57adad8fa2a6cd9c7b2 commit:a081531110ee09a41208f57adad8fa2a6cd9c7b2 ReplicatedPG::recover_backfill: update last_backfill to max() when backfill is complete ** eac9632229365cea22f966e1543685eb6a4974f9 commit:eac9632229365cea22f966e1543685eb6a4974f9 ReplicatedPG: recover_backfill: don't prematurely adjust last_backfill ** 3c889cd118419ef5174245d3d4fa3b922e7c4339 commit:3c889cd118419ef5174245d3d4fa3b922e7c4339 Revert "ReplicatedPG::recover_backfill: do not update last_backfill prematurely" ** b9c6f1f84b6949835c20f6907525d91be1d84c2b commit:b9c6f1f84b6949835c20f6907525d91be1d84c2b ReplicatedPG: replace backfill_pos with last_backfill_started ** 7406f72f42b81efba692d53b3334b19b9b9abb93 commit:7406f72f42b81efba692d53b3334b19b9b9abb93 PG,ReplicatedPG: remove the waiting_for_backfill_peer mechanism ** 0642ca16c6ff6f93186d7aedb6b9d28426c4b3d4 commit:0642ca16c6ff6f93186d7aedb6b9d28426c4b3d4 PG::BackfillInfo: introduce trim_to ** f955c51cbb30b878dac5851c161fc8acdc0e123a commit:f955c51cbb30b878dac5851c161fc8acdc0e123a PG::BackfillInterval: use trim() in pop_front() ** fce6524b9d759be18b9e2030f226f82a4bdcb55b commit:fce6524b9d759be18b9e2030f226f82a4bdcb55b ReplicatedPG::prepare_transaction: info.last_backfill is inclusive ** 1f9e137b51d580bce6505773217bcca008225a47 commit:1f9e137b51d580bce6505773217bcca008225a47 ReplicatedPG: remove the other backfill related flushes ** 0cb797b3e5064ec6b092dc28f23a85fcdbf96526 commit:0cb797b3e5064ec6b092dc28f23a85fcdbf96526 ReplicatedPG: don't rescan the local collection if we can avoid it ** 41c6f4c32fb8c949b5b474317c4d7e10ade2000e commit:41c6f4c32fb8c949b5b474317c4d7e10ade2000e rbd: ObjectCacher reads can hang when reading sparse files ** b2efe7bbddcc7eabb16ced15fa751c891fd1a44e commit:b2efe7bbddcc7eabb16ced15fa751c891fd1a44e Enforce cache size on read requests ** 5f601f099be98c2b061cc94fb06917e7543f3efe commit:5f601f099be98c2b061cc94fb06917e7543f3efe Locker: accept ctime updates from clients without dirty write caps *** http://tracker.ceph.com/issues/9514 ** 285a13a1ecb2c9af65cfe39c7d017e277aeb0b00 commit:285a13a1ecb2c9af65cfe39c7d017e277aeb0b00 libcephfs: get osd location on -1 should return EINVAL ** d6a53ea03622bb95460d6927021011c06d26f506 commit:d6a53ea03622bb95460d6927021011c06d26f506 common/buffer.cc: fix rebuild_page_aligned typo *** http://tracker.ceph.com/issues/6003 ** 1d7a228d85117bf877f99f37bcfce502c4886b7b commit:1d7a228d85117bf877f99f37bcfce502c4886b7b common: rebuild_page_aligned sometimes rebuilds unaligned *** http://tracker.ceph.com/issues/6614 ** 3ebc35026c697ed2b895fdc85f49056a007c7571 commit:3ebc35026c697ed2b895fdc85f49056a007c7571 osdc/Objecter: only post_rx_buffer if no op timeout *** http://tracker.ceph.com/issues/9582 ** 503f865d6432bead72aac0ffba0539d807f078c4 commit:503f865d6432bead72aac0ffba0539d807f078c4 osd/ReplicatedPG: respect RWORDERED rados flag ** 1d3c0c6e3b351f584785c5f603d7b3c52423f79c commit:1d3c0c6e3b351f584785c5f603d7b3c52423f79c crush: fix get_full_location_ordered ** c1d9d8c5c3c87b59588cbc9632f2563b8a939a0d commit:c1d9d8c5c3c87b59588cbc9632f2563b8a939a0d crush/CrushWrapper: simplify get_full_location_ordered() ** bddeb954bd38e58a2dec518d3e3f11234b0cd3a3 commit:bddeb954bd38e58a2dec518d3e3f11234b0cd3a3 sharedptr_registry.hpp: removed ptrs need to not blast contents *** http://tracker.ceph.com/issues/5951 ** 49e8e466bd68a7381ee4547d2b8ea872b9a96fab commit:49e8e466bd68a7381ee4547d2b8ea872b9a96fab SharedPtrRegistry: get_next must not delete while holding the lock ** 478ec880c87bd9dc445bde1472376206a8b940cb commit:478ec880c87bd9dc445bde1472376206a8b940cb sharedptr_registry: add a variant of get_next() and the empty() method ** e0f4162ef07a520a53ad70b6376cf93c94d6e673 commit:e0f4162ef07a520a53ad70b6376cf93c94d6e673 replace in_method_t with a counter ** d236a93b5f0283012134b8cc100741162c548cb0 commit:d236a93b5f0283012134b8cc100741162c548cb0 unit tests for sharedptr_registry * included and tested in integration branch https://github.com/ceph/ceph/tree/dumpling-backports * pull requests waiting to be integrated * issues that need backporting ** * http://tracker.ceph.com/issues/9987 *** ** mon: min_last_epoch_complete tracking broken *** ** Author Sage Weil *** ** https://github.com/ceph/ceph/commit/3fb731b722c50672a5a9de0c86a621f5f50f2d06 ** * http://tracker.ceph.com/issues/10299 *** ** "ObjectCacher.cc: 530: FAILED assert(i->empty())" in rbd-firefly-distro-basic-multi run *** ** Assigned to Jason Dillaman *** ** https://github.com/ceph/ceph/commit/37a99694b1b4637b3b41b312809660d5139e8a23 ** * http://tracker.ceph.com/issues/9675 *** ** splitting a pool doesn't start when rule_id != ruleset_id *** ** Assigned to Joao Luis *** https://github.com/ceph/ceph/commit/b9450b532ab7ad23ec6e2c22ed7cf55e6e1cc4c0 ** https://github.com/ceph/ceph/commit/78e84f34da83abf5a62ae97bb84ab70774b164a6 ** * http://tracker.ceph.com/issues/9806 *** ** Objecter: resend linger ops on split *** ** Assigned to Josh Durgin *** ** https://github.com/ceph/ceph/commit/cb9262abd7fd5f0a9f583bd34e4c425a049e56ce ** * http://tracker.ceph.com/issues/6756 *** ** journal full hang on startup *** ** Assigned to Sage Weil *** ** https://github.com/ceph/ceph/commit/d8d27f13e11dcaefd3aa1c049b97c980283da575 ** * http://tracker.ceph.com/issues/9854 *** ** librbd: reads contending for cache space can cause livelock *** ** Assigned to Jason Dillaman *** ** https://github.com/ceph/ceph/commit/cd784eab22a329cd3eb3798feb9d975a4506c57e