RepoMirrors/ceph
1
0
Fork 0
mirror of https://github.com/ceph/ceph synced 2025-01-03 01:22:53 +00:00
Code Issues Packages Projects Releases Wiki Activity
75,875 Commits 212 Branches 394 Tags 1.5 GiB
126cf70c85
Commit Graph

75875 Commits

This Branch
This Branch
All Branches
Author SHA1 Message Date
Sage Weil
70263dae67 mon: 'osd crush weight-set {ls,dump,create[-compat],rm[-compat],reweight[-compat]}' commands 
Signed-off-by: Sage Weil <sage@redhat.com>
 2017-07-21 13:50:52 -04:00
Sage Weil
a7917a9dbf crush/CrushWrapper: choose_arg_update_item_weight 
Signed-off-by: Sage Weil <sage@redhat.com>
 2017-07-21 13:50:51 -04:00
Sage Weil
65068e2402 common/cmdparse: allow vector<double> 
Signed-off-by: Sage Weil <sage@redhat.com>
 2017-07-21 13:50:51 -04:00
Sage Weil
8c00e00f22 crush/CrushWrapper: {create,rm}_choose_args(index) 
Signed-off-by: Sage Weil <sage@redhat.com>
 2017-07-21 13:50:51 -04:00
Sage Weil
5b1e859efd test/mon/osd-crush: remove xml osd-crush-tree.rng test 
I don't know how to generate or update this, and am not *that* interested
in learning more xml.

Signed-off-by: Sage Weil <sage@redhat.com>
 2017-07-21 13:50:51 -04:00
Sage Weil
fa4608ea5d test/cli/crushtool: fix osd tree output 
Signed-off-by: Sage Weil <sage@redhat.com>
 2017-07-21 13:50:51 -04:00
Sage Weil
b196b27eaf test/ceph_objectstore_tool.py: fix parsing of osd tree output 
Signed-off-by: Sage Weil <sage@redhat.com>
 2017-07-21 13:50:50 -04:00
Sage Weil
f5b3ec9703 crush: dump weight-set values with 'osd crush tree' 
For plain, include a column for each pool.  Show only the first weight if
it is positional.

For formatted, include the full set of weights.

As far of this we switch 'ceph osd crush tree' to have a plaintext mode.

Signed-off-by: Sage Weil <sage@redhat.com>
 2017-07-21 13:50:50 -04:00
Sage Weil
673c114e26 crush/CrushTreeDumper: include parent in Item 
This gives us info we need for the weight_sets...

Signed-off-by: Sage Weil <sage@redhat.com>
 2017-07-21 13:50:50 -04:00
Sage Weil
87f64a951d crush/CrushWrapper: make a compat/default choose_args 
We can make a backward-compatible crush map for legacy clients only if we
have a single global choose_args for the whole map.  Add a special
slot to store that of -1.  (Normally the index is the pool id, so -1 is
unused by ceph.)

Signed-off-by: Sage Weil <sage@redhat.com>
 2017-07-21 13:50:50 -04:00
Sage Weil
f7db29026d crush/CrushWrapper: have_choose_args(index) 
Signed-off-by: Sage Weil <sage@redhat.com>
 2017-07-21 13:50:49 -04:00
Sage Weil
8f033d4a27 crush/CrushWrapper: has_non_straw2_buckets() 
Signed-off-by: Sage Weil <sage@redhat.com>
 2017-07-21 13:50:49 -04:00
Sage Weil
b8287a783e crush/CrushWrapper: whitespace cleanup in decode 
Signed-off-by: Sage Weil <sage@redhat.com>
 2017-07-21 13:50:49 -04:00
Sage Weil
ecf79d566e osd/OSDMap: fix CHOOSE_ARGS feature detection in get_features() 
We need to include the mask in the CEPH_FEATURES_CRUSH value, and | in the
FEATUREMASK later, because this is a second-generation feature bit usage.

This fixes get_min_compat_client(), which was returning 'jewel' when
incompat choose_args were in use instead of 'luminous'.  Now it is correct.

Signed-off-by: Sage Weil <sage@redhat.com>
 2017-07-21 13:50:49 -04:00
Sage Weil
5430c0d944 crush/CrushWrapper: fix decoding of choose_args weight-set 
Assert that ids_size matches the bucket while we are at it.

Signed-off-by: Sage Weil <sage@redhat.com>
 2017-07-21 13:50:48 -04:00
Sage Weil
cfa327279b common/options: remove mon_warn_osd_usage_min_max_delta from options.cc too 
Signed-off-by: Sage Weil <sage@redhat.com>
 2017-07-21 13:36:23 -04:00
Kefu Chai
10b88b5d82 test: create asok files in a temp directory under $TMPDIR 
to shorten the pathname of unix domain socket created for admin socket,
so it does not exceed the limit of 107 on GNU/Linux:

* ceph-helper.sh: the temp directory is named ${TMPDIR:-/tmp}/ceph-asok.$$
* vstart.sh: the temp directory is named `mktemp -u -d "${TMPDIR:-/tmp}/ceph-asok.XXXXXX"`

Fixes: http://tracker.ceph.com/issues/16895
Signed-off-by: Kefu Chai <kchai@redhat.com>
 2017-07-22 01:05:29 +08:00
Yuri Weinstein
9fdd2455d3 Merge pull request #16408 from yanghonggang/master 
osd: fix pg ref leaks when osd shutdown

Reviewed-by: Kefu Chai <kchai@redhat.com>
Reviewed-by: Josh Durgin <jdurgin@redhat.com>
 2017-07-21 09:41:02 -07:00
Sage Weil
3bce6128b5 Merge pull request #16449 from smithfarm/wip-obsolete-xio 
build/ops: rpm: Drop legacy libxio support

Reviewed-by: Ken Dreyer <kdreyer@redhat.com>
Reviewed-by: Kefu Chai <kchai@redhat.com>
 2017-07-21 11:01:39 -05:00
Sage Weil
cb084a55f6 Merge pull request #16453 from liewegas/wip-workloadgen 
crush: enforce buckets-before-rules rule

Reviewed-by: Joao Eduardo Luis <joao@suse.de>
 2017-07-21 11:01:22 -05:00
Sage Weil
33daf6f746 Merge pull request #16455 from liewegas/wip-20667 
osd,mds,mgr: do not dereference null rotating_keys

Reviewed-by: Bassam Tabbara <Bassam.Tabbara@Quantum.com>
 2017-07-21 11:00:42 -05:00
Sage Weil
9bd678e88a Merge pull request #16459 from jdurgin/wip-20704 
osd: fix a couple bugs with persisting the missing set when it contains deletes

Reviewed-by: Sage Weil <sage@redhat.com>
 2017-07-21 10:59:36 -05:00
Sage Weil
29996ea673 Merge pull request #16461 from liewegas/wip-skewed 
mon/PGMap: remove skewed utilizatoin warning

Reviewed-by: xie xingguo <xie.xingguo@zte.com.cn>
 2017-07-21 10:58:33 -05:00
Sage Weil
7832c53375 mon/PGMap: remove skewed utilizatoin warning 
This has a few problems:

1- It does not do it's analysis over CRUSH rule roots/classes, which
means that an innocent user of classes will see skewed usage (bc hdds are
more full than ssds, say)

2- It does not take degraded clusters into account, which means the warning
will appear when a fresh OSD is added.

See http://tracker.ceph.com/issues/20730

Signed-off-by: Sage Weil <sage@redhat.com>
 2017-07-21 11:58:08 -04:00
Sage Weil
f236b5e783 mgr/DaemonState: add per DaemonState lock 
Signed-off-by: Sage Weil <sage@redhat.com>
 2017-07-21 11:30:02 -04:00
Sage Weil
410ab9a25d Merge pull request #16406 from dmick/wip-mgr-counters 
mgr: perf schema fns/change notification and Prometheus plugin

Reviewed-by: Sage Weil <sage@redhat.com>
 2017-07-21 10:28:53 -05:00
Sage Weil
7d59d0d0b0 doc/man/8/ceph: document 'ceph mgr *' commands 
Signed-off-by: Sage Weil <sage@redhat.com>
 2017-07-21 11:25:26 -04:00
Sage Weil
3bc029f469 mgr: use rvalue ref for MMgrBeacon metadata 
Signed-off-by: Sage Weil <sage@redhat.com>
 2017-07-21 11:25:26 -04:00
Sage Weil
75ac7d85da qa/workunits/cephtool/test.sh: add a few tests 
Signed-off-by: Sage Weil <sage@redhat.com>
 2017-07-21 11:25:05 -04:00
Sage Weil
293b7ba956 mgr/ServiceMap: fix typo in service status summary 
Signed-off-by: Sage Weil <sage@redhat.com>
 2017-07-21 11:25:04 -04:00
Sage Weil
80a42122a0 mon: 'versions' command to show running versions for daemons of all types 
Easier!

Signed-off-by: Sage Weil <sage@redhat.com>
 2017-07-21 11:25:04 -04:00
Sage Weil
4f1db4c437 mon: factor out count_metadata w/ counter helper 
Signed-off-by: Sage Weil <sage@redhat.com>
 2017-07-21 11:25:04 -04:00
Sage Weil
831e2a75a6 mon/MgrMonitor: 'mgr metadata', 'mgr versions', 'mgr count-metadata' 
Signed-off-by: Sage Weil <sage@redhat.com>
 2017-07-21 11:25:04 -04:00
Sage Weil
1bf4a89890 mon/MgrMonitor: store mgr daemon metadata 
Signed-off-by: Sage Weil <sage@redhat.com>
 2017-07-21 11:24:44 -04:00
Sage Weil
8d177dbb84 Merge pull request #16484 from jecluis/wip-20371-qa-suites 
qa: flush out monc's dropped msgs on msgr failure injection
 2017-07-21 09:46:53 -05:00
Joao Eduardo Luis
6f6fbe7870 qa: flush out monc's dropped msgs on msgr failure injection 
We have a few open tickets regarding the mgr being down during suites
involving messenger failure injection. There are a few suspicions that
this may be related with the monclient, but we'll need more logs to
validate those suspicions and, more, to validate we're actually fixing
the issue.

Signed-off-by: Joao Eduardo Luis <joao@suse.de>
 2017-07-21 15:29:21 +01:00
Casey Bodley
6929f21c4e Merge pull request #16448 from linuxbox2/wip-supriti-ord 
rgw_file: properly & |'d flags

Reviewed-by: Casey Bodley <cbodley@redhat.com>
 2017-07-21 10:11:46 -04:00
Jos Collin
5128e7b135 Merge pull request #16475 from amitkumar50/branchAffliation 
Add amitkumar50 affiliation to .organizationmap

Reviewed-by: Jos Collin <jcollin@redhat.com>
 2017-07-21 13:19:09 +00:00
amitkuma
a087849603 Add amitkumar50 affiliation to .organizationmap 
Signed-off-by: Amit Kumar <amitkuma@redhat.com>
 2017-07-21 20:36:00 +05:30
Nikita Gerasimov
424166db02 ceph-disk: Fix for missing 'not' in *_is_diskdevice checks 
In 85d7f8817e was introdused
is_diskdevice() tests but in some 'if' statements 'not' are lost.

Fixes: http://tracker.ceph.com/issues/20706
Signed-off-by: Nikita Gerasimov <nikita.gerasimov@oracle.com>
 2017-07-21 15:34:28 +03:00
Jos Collin
fae6dc4786 Merge pull request #16430 from yuriw/wip_add_luminous 
qa: Added luminous to the mix in schedule_subset.sh

Reviewed-by: Kefu Chai <kchai@redhat.com>
Reviewed-by: Jos Collin <jcollin@redhat.com>
 2017-07-21 12:11:29 +00:00
John Spray
60642bf7f5 common: remove config opt conversion utility 
Now that the new options.cc stuff has landed, this is no longer
needed.

Signed-off-by: John Spray <john.spray@redhat.com>
 2017-07-21 08:06:47 -04:00
John Spray
223c8ce27f Merge pull request #16211 from jcsp/wip-options-jcsp 
common: Revamp config option definitions

Reviewed-by: Sage Weil <sage@redhat.com>
Reviewed-by: Kefu Chai <kchai@redhat.com>
 2017-07-21 13:02:16 +01:00
Jason Dillaman
ee958a5930 Merge pull request #16313 from Songweibin/wip-mirror-peer 
rbd: make it more understandable when adding peer returns error

Reviewed-by: Jason Dillaman <dillaman@redhat.com>
 2017-07-21 08:00:50 -04:00
Kefu Chai
6a821eea8e Merge pull request #16462 from wjwithagen/wjw-warning-unused-unsetdumpable 
common: prevent unset_dumpable from generating warnings

Reviewed-by: Kefu Chai <kchai@redhat.com>
 2017-07-21 19:50:45 +08:00
xie xingguo
cb1616b7e4 mon/HealthMonitor: avoid sending unnecessary MMonHealthChecks to leader 
If there is no historic warnings and no new warnings is generated,
skip sending MMonHealthChecks(for peon) or updating quorum_checks(for leader).

Signed-off-by: xie xingguo <xie.xingguo@zte.com.cn>
 2017-07-21 19:50:11 +08:00
Kefu Chai
4599eb7963 Merge pull request #16454 from liewegas/wip-fix-ceph-scrub 
qa/tasks/ceph_manager: wait for osd to start after objectstore-tool sequence

Reviewed-by: Kefu Chai <kchai@redhat.com>
 2017-07-21 19:31:19 +08:00
Kefu Chai
3aa3b4b60d mon/HealthMonitor: trigger a proposal if stat updated 
leader should always propose if the peon update it with new health stats

Signed-off-by: Kefu Chai <kchai@redhat.com>
 2017-07-21 18:31:53 +08:00
John Spray
b519ed0519 common/options: update for latest added 
Signed-off-by: John Spray <john.spray@redhat.com>
 2017-07-21 06:30:27 -04:00
John Spray
85779cba67 common: fix erasure_code_dir definition 
This was missing its `safe` flag, causing some attempts
to set it during testing to fail.

Signed-off-by: John Spray <john.spray@redhat.com>
 2017-07-21 06:27:30 -04:00