Sage Weil
73981ad807
qa/suites: remove 'fs' facet from all tests
...
The objectstore facet now covers bluestore, filestore(xfs),
and filestore(btrfs).
Signed-off-by: Sage Weil <sage@redhat.com>
2017-03-28 11:57:21 -04:00
Sage Weil
208675afb1
qa/objectstore: split filestore into xfs and btrfs variants
...
Signed-off-by: Sage Weil <sage@redhat.com>
2017-03-28 11:54:56 -04:00
Sage Weil
35b60ae057
Merge pull request #13984 from liewegas/wip-osdmap-remap
...
mon,osd: explicitly remap some pgs
Reviewed-by: Josh Durgin <jduring@redhat.com>
2017-03-28 10:47:22 -05:00
Sage Weil
eda8d28c37
Merge pull request #13687 from Liuchang0812/wip-19106-eliminate-unsafe-option
...
common/config: eliminate config_t::set_val unsafe option
Reviewed-by: Sage Weil <sage@redhat.com>
2017-03-28 10:45:01 -05:00
Sage Weil
ba97f1f562
Merge pull request #13855 from yuyuyu101/wip-bluestore-rm-range
...
kv/RocksDBStore: implement rm_range_keys operator interface and test
Reviewed-by: Sage Weil <sage@redhat.com>
2017-03-28 10:44:44 -05:00
Sage Weil
3f107041f4
Merge pull request #13879 from liewegas/wip-19076
...
osd/ReplicatedBackend: clear pull source once we are done with it
Reviewed-by: Sage Weil <sage@redhat.com>
2017-03-28 10:44:16 -05:00
Sage Weil
1c954f4b64
Merge pull request #13961 from wangzhengyong/network
...
msg/SimpleMessenger: error out misplace in set_socket_options
Reviewed-by: Greg Farnum <gfarnum@redhat.com>
2017-03-28 10:43:37 -05:00
Sage Weil
936966601a
Merge pull request #14093 from hjwsm1989/sparse-read
...
osd/PrimaryLogPG: set return value if sparse read failed
Reviewed-by: Sage Weil <sage@redhat.com>
2017-03-28 10:43:07 -05:00
Sage Weil
d999e36c02
Merge pull request #14099 from tchaikov/wip-denc-arm32
...
include/denc: add {encode,decode}_nohead for denc_traits<basic_string>
Reviewed-by: Sage Weil <sage@redhat.com>
2017-03-28 10:42:40 -05:00
Sage Weil
fe9d54ac2c
Merge pull request #14129 from SUSE/wip-18764
...
mgr: fix crash on missing 'ceph_version' in daemon metadata (fixes #18764 )
Reviewed-by: Sage Weil <sage@redhat.com>
2017-03-28 10:42:23 -05:00
Mykola Golub
31f456b1f0
Merge pull request #14003 from dillaman/wip-19287
...
librbd: is_exclusive_lock_owner API should ping OSD
Reviewed-by: Mike Christie <mchristi@redhat.com>
Reviewed-by: Mykola Golub <mgolub@mirantis.com>
2017-03-28 18:03:54 +03:00
Haomai Wang
b164ac5990
Merge pull request #14179 from Adirl/rdma-cm-3
...
msg/async/rdma: RDMA-CM Initialize device on first connect
Reviewed-by: Haomai Wang <haomai@xsky.com>
2017-03-28 23:01:50 +08:00
Mykola Golub
e01dcbd9ee
Merge pull request #14166 from dillaman/wip-19388
...
librbd: random unit test failures due to shut down race
Reviewed-by: Mykola Golub <mgolub@mirantis.com>
2017-03-28 17:56:46 +03:00
Sage Weil
2a08cbbed5
qa/tasks/thrashosds,ceph_manager: thrash pg_remap[_items]
...
Signed-off-by: Sage Weil <sage@redhat.com>
2017-03-28 10:12:10 -04:00
Sage Weil
bd49dc0863
test/cli/osdmaptool/remap.t: add --remap test
...
Signed-off-by: Sage Weil <sage@redhat.com>
2017-03-28 10:12:09 -04:00
Sage Weil
ae557d0557
test/cli/osdmaptool: fix tests
...
Signed-off-by: Sage Weil <sage@redhat.com>
2017-03-28 10:12:09 -04:00
Sage Weil
d40825377e
osdmaptool: add --remap to run the remap_pgs() method
...
Output ceph cli commands to realize the new remaps entries.
Signed-off-by: Sage Weil <sage@redhat.com>
2017-03-28 10:12:09 -04:00
Sage Weil
0364b3f173
osd/OSDMap: clean_pg_remaps
...
Helper to clean out no-op pg_remap cruft:
- pg_remap entries that match the raw result
- pg_remap_items entries where the 'from' part of the (from,to)
set isn't present.
Signed-off-by: Sage Weil <sage@redhat.com>
2017-03-28 10:12:09 -04:00
Sage Weil
7f1934cb1e
osd/OSDMap: implement remap_pgs
...
Run a specified number of iterations trying to install new
pg_remap_items entries that improve the PG distribution.
Signed-off-by: Sage Weil <sage@redhat.com>
2017-03-28 10:12:09 -04:00
Sage Weil
06ec9d41eb
crush: implement try_remap_rule
...
Simulate a CRUSH mapping but try to identify alternative OSD
choices (based on an underfull list and overfull set) that still
respect the CRUSH rule constraints.
Signed-off-by: Sage Weil <sage@redhat.com>
2017-03-28 10:12:09 -04:00
John Spray
af3b8ed617
Merge pull request #14021 from batrick/i19291
...
mds: set ceph-mds name uncond for external tools
Reviewed-by: John Spray <john.spray@redhat.com>
2017-03-28 09:29:29 -04:00
John Spray
dbe1f60aed
Merge pull request #14017 from jcsp/wip-19282
...
mds: don't assert on read errors in RecoveryQueue
Reviewed-by: Patrick Donnelly <pdonnell@redhat.com>
2017-03-28 09:28:52 -04:00
John Spray
48120683cf
Merge pull request #14015 from jcsp/wip-19288
...
mds: remove legacy "mds tell" command
Reviewed-by: Patrick Donnelly <pdonnell@redhat.com>
2017-03-28 09:28:13 -04:00
John Spray
afe7d0d78f
Merge pull request #13904 from jcsp/wip-16709
...
mon: fix hiding mdsmonitor informative strings
Reviewed-by: Patrick Donnelly <pdonnell@redhat.com>
2017-03-28 09:27:38 -04:00
John Spray
d4959004d1
Merge pull request #12074 from batrick/i17604
...
mds: warn if insufficient standbys exist
Reviewed-by: John Spray <john.spray@redhat.com>
2017-03-28 09:25:24 -04:00
Amir Vadai
1890e924ca
msg/async/rdma: Extract sockets stuff from RDMAStack.h
...
This is a preparation commit, in order to make review easier. In this
commit I move code from RDMAStack.h into the new file
RDMAConnectedSocketImpl.h - without changing the code.
In the next commit, the actual logic changes will be done and socket
classes will be split into a base RDMAConnected classes and child
classes with TCP connection establishment specific code.
Issue: 995322
Change-Id: I639fda490a6fbd02addb95d3158c5ac1e7390ef0
Signed-off-by: Amir Vadai <amir@vadai.me>
2017-03-28 09:52:07 +03:00
Amir Vadai
94eddb1ec5
msg/async/rdma: Initialize device on first connect
...
Allocate Device's IB resources only when first connect for a device is
called.
Also, removed Infiniband::MemoryManager::Chunk::post_srq() which is not
used anywhere in the code.
Issue: 995322
Change-Id: I2ac1e36aff645ad7d8bbc06c87530bf33c4b3ecf
Signed-off-by: Amir Vadai <amir@vadai.me>
2017-03-28 09:49:41 +03:00
David Zafman
4c8ec8a2c3
Merge pull request #13423 from dzafman/wip-18533-tool
...
Fixes: http://tracker.ceph.com/issues/18533
Reviewed-by: Josh Durgin <jdurgin@redhat.com>
2017-03-27 15:15:43 -07:00
Dan Mick
64ee8e8dde
Merge pull request #14132 from dmick/wip-arm-crc
...
arch: fix cmake's ARM CRC intrinsics test to handle duplicitous gcc 4.8.5
Reviewed-by: Kefu Chai <kchai@redhat.com>
2017-03-27 12:50:57 -07:00
Dan Mick
ac4815f74b
cmake/modules/SIMDExt.cmake: armv8 crypto intrinsics
...
Test not only for -march support, but also the actual
presence of the intrinsic routines. Not sure why, but gcc
4.8.5 passes the first but not the second.
Fixes: http://tracker.ceph.com/issues/19386
Signed-off-by: Dan Mick <dan.mick@redhat.com>
2017-03-27 09:35:22 -07:00
Jason Dillaman
f3984f00de
librbd: random unit test failures due to shut down race
...
Fixes: http://tracker.ceph.com/issues/19389
Signed-off-by: Jason Dillaman <dillaman@redhat.com>
2017-03-27 11:41:35 -04:00
David Zafman
1704f62c08
filestore, tools: Fix logging of DBObjectMap check() repairs
...
Signed-off-by: David Zafman <dzafman@redhat.com>
2017-03-27 08:33:34 -07:00
David Zafman
738156a99e
osd: Simplify DBObjectMap by no longer creating complete tables
...
Bump the version for new maps to 3
Make clone less efficient but simpler
Add rename operation (use instead of clone/unlink)
For now keep code that understands version 2 maps
Signed-off-by: David Zafman <dzafman@redhat.com>
2017-03-27 08:33:34 -07:00
David Zafman
1dda0411f4
ceph-osdomap-tool: Fix seg fault with large amount of check error output
...
Signed-off-by: David Zafman <dzafman@redhat.com>
2017-03-27 08:33:34 -07:00
David Zafman
4cd3c74c92
osd: Add automatic repair for DBObjectMap bug
...
Add repair command to ceph-osdomap-tool too
Under some situations the previous rm_keys() code would
generated a corrupt complete table. There is no way to
figure out what the table should look like now. By removing
the entries we fix the corruption and aren't much worse off
because the corruption caused some deleted keys to re-appear.
This doesn't breaking the parent/child relationship during
repair because some of the keys may still be contained
in the parent.
Signed-off-by: David Zafman <dzafman@redhat.com>
2017-03-27 08:33:34 -07:00
David Zafman
666f14ed90
ceph-osdomap-tool: Fix tool exit status
...
Signed-off-by: David Zafman <dzafman@redhat.com>
2017-03-27 08:33:05 -07:00
Samuel Just
c4dffb68ea
DBObjectMap: rewrite rm_keys and merge_new_complete
...
Leverage the updated in_complete_region and needs_parent to simplify
these methods.
Signed-off-by: Samuel Just <sjust@redhat.com>
2017-03-27 08:33:05 -07:00
Samuel Just
97b35f4d7d
DBObjectMap: strengthen in_complete_region post condition
...
Previously, in_complete_region didn't guarantee anything about
where it left complete_iter pointing. It will be handy for
complete_iter to be pointing at the lowest interval which ends
after to_test. Make it so.
Signed-off-by: Samuel Just <sjust@redhat.com>
2017-03-27 08:33:05 -07:00
Samuel Just
74a7631d09
DBObjectMap: fix next_parent()
...
The previous implementation assumed that
lower_bound(parent_iter->key()) always leaves the iterator
on_parent(). There isn't any guarantee, however, that that
key isn't present on the child as well.
Signed-off-by: Samuel Just <sjust@redhat.com>
2017-03-27 08:33:05 -07:00
Samuel Just
f131dbcf5b
test_object_map: add tests to trigger some bugs related to 18533
...
Signed-off-by: Samuel Just <sjust@redhat.com>
2017-03-27 08:33:05 -07:00
David Zafman
0e97a01bd7
test: Add ceph_test_object_map to make check tests
...
Signed-off-by: David Zafman <dzafman@redhat.com>
2017-03-27 08:33:05 -07:00
David Zafman
5fb2b2d139
ceph-osdomap-tool: Add --debug and only show internal logging if enabled
...
Signed-off-by: David Zafman <dzafman@redhat.com>
2017-03-27 08:33:05 -07:00
David Zafman
fcf1e17c64
osd: DBOjectMap::check: Dump complete mapping when inconsistency found
...
Signed-off-by: David Zafman <dzafman@redhat.com>
2017-03-27 08:33:05 -07:00
David Zafman
053a273cbc
test_object_map: Use ASSERT_EQ() for check() so failure doesn't stop testing
...
Signed-off-by: David Zafman <dzafman@redhat.com>
2017-03-27 08:33:05 -07:00
David Zafman
e5e8eb962d
tools: Check for overlaps in internal "complete" table for DBObjectMap
...
Changed check to return an error count and fix tool error message
Signed-off-by: David Zafman <dzafman@redhat.com>
2017-03-27 08:33:04 -07:00
David Zafman
f4101591ad
tools: Add dump-headers command to ceph-osdomap-tool
...
Signed-off-by: David Zafman <dzafman@redhat.com>
2017-03-27 08:32:28 -07:00
David Zafman
2d94889e9e
tools: Add --oid option to ceph-osdomap-tool
...
Signed-off-by: David Zafman <dzafman@redhat.com>
2017-03-27 08:31:12 -07:00
David Zafman
937e6a03ea
osd: Remove unnecessary assert and assignment in DBObjectMap
...
Fix and add comment(s)
Signed-off-by: David Zafman <dzafman@redhat.com>
2017-03-27 08:31:12 -07:00
Kefu Chai
52468ae1f9
Merge pull request #13754 from liewegas/wip-objectstore-random
...
os: allow 'osd objectstore = random' to pick either filestore or bluestore
Reviewed-by: Kefu Chai <kchai@redhat.com>
2017-03-27 22:58:29 +08:00
Jianpeng Ma
903b3fcf4a
os/bluestore: remove unuse perfcounter:l_bluestore_read_wait_flush_lat.
...
Now read don't need flush(). See commit 83e33a32fd
.
Signed-off-by: Jianpeng Ma <jianpeng.ma@intel.com>
Reviewed-by: Sage Weil <sage@redhat.com>
2017-03-27 10:38:36 -04:00