RepoMirrors/ceph
1
0
Fork 0
mirror of https://github.com/ceph/ceph synced 2024-12-16 16:39:21 +00:00
Code Issues Packages Projects Releases Wiki Activity
77,199 Commits 211 Branches 394 Tags 2 GiB
bd27b2e76a
Commit Graph

77199 Commits

This Branch
This Branch
All Branches
Author SHA1 Message Date
Sage Weil
ba3602b216 Merge pull request #17005 from liewegas/wip-verify-slowness 
qa/suites/rados/verify/validater/valgrind: whitelist PG_
 2017-08-12 13:19:23 -05:00
Sage Weil
41e5a85308 qa/suites/rados/verify/validater/valgrind: whitelist PG_ 
Peering might be slow due to valgrind.

Signed-off-by: Sage Weil <sage@redhat.com>
 2017-08-12 14:18:59 -04:00
Sage Weil
f04360c917 Merge pull request #17004 from liewegas/wip-clock-skew-whitelist 
qa/suites/rados/multimon/tasks/mon_lock_with_skew: whitelist PG_
 2017-08-12 13:15:50 -05:00
Sage Weil
12007044b1 qa/suites/rados/multimon/tasks/mon_lock_with_skew: whitelist PG_ 
Default pool pgs not up because mons too broken for OSDs to peer.

Signed-off-by: Sage Weil <sage@redhat.com>
 2017-08-12 14:15:15 -04:00
Sage Weil
5dd89cc17a Merge pull request #15921 from joscollin/wip-CID1411830-Uninitialized-pointer-field 
test/osd: Non-static class members not initialized in UnsetRedirectOp

Reviewed-by: xie xingguo <xie.xingguo@zte.com.cn>
 2017-08-12 13:03:58 -05:00
mychoxin
e6855210e0 os/bluestore: function collect_metadata, use 'const string&' to reduce copy 
Signed-off-by: mychoxin <mychoxin@gmail.com>
 2017-08-12 16:49:10 +08:00
mychoxin
a3b27da52e os/bluestore: move aio_callback and aio_callback_priv to class BlockDevice 
Signed-off-by: mychoxin <mychoxin@gmail.com>
 2017-08-12 16:48:44 +08:00
Jos Collin
b1b6688b50 test/osd: Non-static class member not initialized in UnsetRedirectOp 
Fixes the coverity scan report:
CID 1411830: Uninitialized pointer field (UNINIT_CTOR)
2. uninit_member: Non-static class member completion is not initialized in this constructor nor in any functions that it calls.
4. uninit_member: Non-static class member comp is not initialized in this constructor nor in any functions that it calls.

Signed-off-by: Jos Collin <jcollin@redhat.com>
 2017-08-12 13:36:08 +05:30
Jos Collin
0ec42b03c9 Merge pull request #15922 from joscollin/wip-CID1411829-Uninitialized-scalar-field 
test/osd: initialize Non-static class members in WeightedTestGenerator

Reviewed-by: Xie Xingguo <xie.xingguo@zte.com.cn>
 2017-08-12 07:50:05 +00:00
Sage Weil
4f6ae418c0 Merge pull request #16924 from liewegas/wip-20923 
os/bluestore: fail early on very large objects

Reviewed-by: xie xingguo <xie.xingguo@zte.com.cn>
 2017-08-11 21:55:34 -05:00
Sage Weil
ba66977a34 test/osd/safe-to-destroy.sh: test 'osd safe-to-destroy' 
This is hard with workunits/cephtool/test.sh because we don't
control the whole cluster.

Signed-off-by: Sage Weil <sage@redhat.com>
 2017-08-11 22:54:33 -04:00
Sage Weil
bf9380457b mgr: implement 'osd safe-to-destroy' and 'ok-to-stop' commands 
An osd is safe to destroy if

- we have osd_stat for it
- osd_stat indicates no pgs stored
- all pgs are known
- no pgs map to it

An osd is ok ot stop if

- we have pg stats
- no pgs will drop below min_size

Signed-off-by: Sage Weil <sage@redhat.com>
 2017-08-11 22:54:33 -04:00
Sage Weil
a7612d33d1 mgr/ClusterState: record osd_stat for out osds too 
I'm not quite sure why we were doing this. :/

Signed-off-by: Sage Weil <sage@redhat.com>
 2017-08-11 22:47:50 -04:00
Sage Weil
c2945479f4 osd/osd_types: include number of locally stored PGs in osd_stat_t 
Signed-off-by: Sage Weil <sage@redhat.com>
 2017-08-11 22:47:50 -04:00
Sage Weil
6fc33a046b osd/OSDMap: add parse_osd_id_list helper 
Signed-off-by: Sage Weil <sage@redhat.com>
 2017-08-11 22:47:50 -04:00
Sage Weil
8619fb6a2a Merge pull request #16992 from liewegas/wip-20977 
os/bluestore: do not segv on kraken upgrade debug print

Reviewed-by: xie xingguo <xie.xingguo@zte.com.cn>
 2017-08-11 21:36:53 -05:00
Sage Weil
f1d8a40122 Merge pull request #16993 from liewegas/wip-mimic 
common: kick off mimic

Reviewed-by: Gregory Farnum <gfarnum@redhat.com>
Reviewed-by: Abhishek Lekshmanan <abhishek.lekshmanan@gmail.com>
 2017-08-11 21:36:32 -05:00
Sage Weil
ae739b31f1 Merge pull request #16994 from liewegas/wip-20983 
os/bluestore: fix clone dirty_range again

Reviewed-by: xie xingguo <xie.xingguo@zte.com.cn>
 2017-08-11 21:35:35 -05:00
Sage Weil
93b1b5ca49 Merge pull request #16995 from liewegas/wip-20738 
os/memstore: memstore_page_set=false

Reviewed-by: Adam Emerson <aemerson@redhat.com>
 2017-08-11 21:34:48 -05:00
Sage Weil
300e395c75 Merge pull request #16996 from liewegas/wip-mgr-vs-clock-skew 
qa/suites/rados/multimon: whitelist mgr down vs clock skew test
 2017-08-11 21:34:07 -05:00
Greg Farnum
6f7545e0b5 mon: correctly print out mds versions (instead of mon ones) 
Signed-off-by: Greg Farnum <gfarnum@redhat.com>
 2017-08-11 15:31:43 -07:00
Greg Farnum
a9921a2fda osd: be more precise about our asserts and cases when rebuilding missing sets and handling divergent priors 
Fixes: http://tracker.ceph.com/issues/20985

Signed-off-by: Greg Farnum <gfarnum@redhat.com>
 2017-08-11 15:10:01 -07:00
Gregory Farnum
4e69d80e6b Merge pull request #16898 from amitkumar50/cov-messages-6 
messages: Initialization of member variables

Reviewed-by: Greg Farnum <gfarnum@redhat.com>
 2017-08-11 14:25:16 -07:00
Sage Weil
50523a225b os/bluestore: do not segv on kraken upgrade debug print 
When loading an onode from kraken we have a compat path that calls
get_ref before the SharedBlob pointer is initialized.  This is fine except
that if debugging is enabled the operator<< on the Blob will segv on
printing *b.shared_blob (which is NULL).

Fix operator<< to print something else if it is NULL.  shared_blob does
get set up right after the call to decode() so having it be NULL at this
point is otherwise harmless.

Fixes: http://tracker.ceph.com/issues/20977
Signed-off-by: Sage Weil <sage@redhat.com>
 2017-08-11 16:44:48 -04:00
David Zafman
75b425671a osd: In scrub's be_select_auth_object() detect multiple errors better 
Signed-off-by: David Zafman <dzafman@redhat.com>
 2017-08-11 11:37:32 -07:00
David Zafman
4c949b6258 osd, rados: Adding ss_attr_missing and ss_attr_corrupt errors to list-inconsistent-obj 
Signed-off-by: David Zafman <dzafman@redhat.com>
 2017-08-11 11:37:32 -07:00
David Zafman
5f58301a13 osd, rados: Improve size scrub error handling 
Fixes: http://tracker.ceph.com/issues/20243

Signed-off-by: David Zafman <dzafman@redhat.com>
 2017-08-11 11:37:32 -07:00
David Zafman
437e5cf106 osd: Compare all object info even when can't consider for auth copy 
Signed-off-by: David Zafman <dzafman@redhat.com>
 2017-08-11 11:37:32 -07:00
David Zafman
8e2b9a07e0 osd: Change a check to an assert() since it can't happen anymore 
Signed-off-by: David Zafman <dzafman@redhat.com>
 2017-08-11 11:37:03 -07:00
David Zafman
8ad4b29113 osd: Add whether shard is primary in list-inconsistent-obj 
Add new field in the client interface
Update test case

Fixes: http://tracker.ceph.com/issues/18836

Signed-off-by: David Zafman <dzafman@redhat.com>
 2017-08-11 11:37:03 -07:00
David Zafman
c0606b9eea test: Add undocumented corrupt-size for testing 
Signed-off-by: David Zafman <dzafman@redhat.com>
 2017-08-11 11:37:03 -07:00
David Zafman
8325639a32 osd: Fix test op error message 
Signed-off-by: David Zafman <dzafman@redhat.com>
 2017-08-11 11:37:03 -07:00
Sage Weil
ad23d7dc1f qa/suites/rados/multimon: whitelist mgr down vs clock skew test 
Clock skew might make us fail the mgr.

Signed-off-by: Sage Weil <sage@redhat.com>
 2017-08-11 13:42:02 -04:00
Sage Weil
7c96868797 os/memstore: memstore_page_set=false 
This regularly returns bad results, see http://tracker.ceph.com/issues/20738

Signed-off-by: Sage Weil <sage@redhat.com>
 2017-08-11 13:31:51 -04:00
Sage Weil
d5ba7061ee os/bluestore: fix clone dirty_range again 
If we are cloning a blob for a 1 byte logical extent then dirty_range_begin
will equal _end and we won't dirty the source onode (with possibly newly
shared blobs).

Fix by using a separate flag to indicate whether we are dirtying instead
of overloading the begin/end markers for this.  Note that even if they
are equal dirty_range will still dirty the shard in question.

This is a result of 0ae5d92d42.

Fixes: http://tracker.ceph.com/issues/20983
Signed-off-by: Sage Weil <sage@redhat.com>
 2017-08-11 12:46:25 -04:00
Sage Weil
3bcb4b3e8d include/rados: nautilus is next 
Signed-off-by: Sage Weil <sage@redhat.com>
 2017-08-11 12:14:59 -04:00
Sage Weil
7d56c7082b include/ceph_features: define SERVER_NAUTILUS 
Nautilus will be the next major release after luminous.

Signed-off-by: Sage Weil <sage@redhat.com>
 2017-08-11 12:14:59 -04:00
Sage Weil
4a1a15478d include/ceph_features: incarnation 3 can begin! 
With all upgrades passing through luminous, we can now start
reusing bits retired in luminous.  Our sentinel bitmask will be the
combination of SERVER_MIMIC and SERVER_JEWEL (i.e.,
CEPH_FEATUREMASK_SERVER_MIMIC).

Signed-off-by: Sage Weil <sage@redhat.com>
 2017-08-11 12:14:59 -04:00
Sage Weil
7b81212a6d include/ceph_features: SERVER_M -> SERVER_MIMIC 
Signed-off-by: Sage Weil <sage@redhat.com>
 2017-08-11 12:14:59 -04:00
Sage Weil
9cec473e0c ceph_release: kick off mimic 13.y.z 
Signed-off-by: Sage Weil <sage@redhat.com>
 2017-08-11 12:14:59 -04:00
Matt Benjamin
50724db8e5 Merge pull request #16628 from theanalyst/rgw-policy-list-buckets 
rgw: policy: support for s3 conditionals in ListBucket Op
 2017-08-11 11:53:01 -04:00
Sage Weil
410a23bcdb Merge pull request #16968 from liewegas/wip-store-test-expect-death 
ceph_test_objectstore: drop expect regex
 2017-08-11 08:45:02 -05:00
Sage Weil
863468803d ceph_test_objectstore: drop expect regex 
If logging is enabled (as it now is in teuthology) this won't match the
forked output.

Signed-off-by: Sage Weil <sage@redhat.com>
 2017-08-11 09:43:20 -04:00
Sage Weil
9c42597f09 qa/workunits/mon/crush_ops.sh: test weight sets vs device classes 
Signed-off-by: Sage Weil <sage@redhat.com>
 2017-08-11 09:40:11 -04:00
Sage Weil
49c468b788 mon/OSDMonitor: remove choose_args when pool is removed 
Signed-off-by: Sage Weil <sage@redhat.com>
 2017-08-11 09:40:11 -04:00
Sage Weil
0ed55e6150 crush/CrushWrapper: fill in weight-sets when we build shadow trees 
When we build the shadow buckets for the class hierarchies, we need
to fill in the weight-sets for each shadow bucket too.

Skip the ids vector for now since it's not yet used by anything.

Fixes: http://tracker.ceph.com/issues/20939
Signed-off-by: Sage Weil <sage@redhat.com>
 2017-08-11 09:40:11 -04:00
Sage Weil
2649b27d13 crush/CrushWrapper: remove unused 'unused' arg for trim_roots_with_classes 
Signed-off-by: Sage Weil <sage@redhat.com>
 2017-08-11 09:39:49 -04:00
Sage Weil
1f62ee8408 crush: do add/remove before updating weight-sets 
Signed-off-by: Sage Weil <sage@redhat.com>
 2017-08-11 09:39:49 -04:00
Sage Weil
9ecf2cbaed Merge pull request #16991 from liewegas/wip-doc-bluestore-migration 
doc/rados/operations/bluestore-migration: typos and whitespace
 2017-08-11 08:37:28 -05:00
Sage Weil
0a1512b023 doc/rados/operations/bluestore-migration: typos and whitespace 
Signed-off-by: Sage Weil <sage@redhat.com>
 2017-08-11 09:36:59 -04:00