Commit Graph

37170 Commits

Author SHA1 Message Date
Greg Farnum
4d2f63e73d mailmap: Update Noah Watkins' attribution to Red Hat
Signed-off-by: Greg Farnum <gfarnum@redhat.com>
2014-11-23 21:04:53 -08:00
Greg Farnum
4ee3a87691 mailmap: Update Neil Levine's attribution to Red Hat
Signed-off-by: Greg Farnum <gfarnum@redhat.com>
2014-11-23 21:02:38 -08:00
Greg Farnum
12e511b537 mailmap: Update Mark Nelson's attribution to Red Hat
Signed-off-by: Greg Farnum <gfarnum@redhat.com>
2014-11-23 21:01:56 -08:00
Greg Farnum
bdf534a543 mailmap: Update Ken Dreyer's attribution to Red Hat
Signed-off-by: Greg Farnum <gfarnum@redhat.com>
2014-11-23 21:00:55 -08:00
Greg Farnum
df19410610 mailmap: Update Josh Durgin's attribution to Red Hat
Signed-off-by: Greg Farnum <gfarnum@redhat.com>
2014-11-23 21:00:34 -08:00
Greg Farnum
d9ee8d601c mailmap: Update John Wilkins' attribution to Red Hat
Signed-off-by: Greg Farnum <gfarnum@redhat.com>
2014-11-23 20:59:49 -08:00
Greg Farnum
c84aa39420 mailmap: Update Joao Luis' attribution to Red Hat
Signed-off-by: Greg Farnum <gfarnum@redhat.com>
2014-11-23 20:56:00 -08:00
Greg Farnum
806a478edc mailmap: Update Ilya Dryomov's attribution to Red Hat
Signed-off-by: Greg Farnum <gfarnum@redhat.com>
2014-11-23 20:54:38 -08:00
Greg Farnum
eaa9889bde mailmap: Update Dan Mick's attribution
Point his addresses at Red Hat now, but revert the Inktank->Redhat mappings
so we have more historical accuracy.

Signed-off-by: Greg Farnum <gfarnum@redhat.com>
2014-11-23 20:49:23 -08:00
Greg Farnum
330b09de1a mailmap: Update Alfredo Deza's attribution to Red Hat
Signed-off-by: Greg Farnum <gfarnum@redhat.com>
2014-11-23 20:47:01 -08:00
Greg Farnum
1853af3dcd mailmap: Update Greg Farnum's attribution to Red Hat
Signed-off-by: Greg Farnum <gfarnum@redhat.com>
2014-11-23 20:42:50 -08:00
Sage Weil
47bc8da329 Merge remote-tracking branch 'gh/next' 2014-11-23 18:55:32 -08:00
Sage Weil
ca13ce1b35 Merge pull request #2983 from tchaikov/wip-fix-lock-dep
lockdep: do not use $CEPH_LOCKDEP for g_lockdep

Reviewed-by: Sage Weil <sage@redhat.com>
2014-11-23 12:59:42 -08:00
Kefu Chai
5470914f73 lockdep: do not use $CEPH_LOCKDEP for g_lockdep
* a non-zero CEPH_LOCKDEP brings ceph down because g_lockdep_ceph_ctx
  is still being constructed when dout_impl() dereferences it.
* fix a typo in comment.
* remove dead code.

Signed-off-by: Kefu Chai <tchaikov@gmail.com>
2014-11-24 03:19:40 +08:00
Yehuda Sadeh
9ba17a321d rgw-admin: create subuser if needed when creating user
Fixes: #10103
Backport: firefly, giant
This turned up after fixing #9973. Earlier we also didn't create the
subuser in this case, but we didn't really read the subuser info when it
was authenticating. Now we do that as required, so we end up failing the
authentication. This only applies to cases where a subuser was created
using 'user create', and not the 'subuser create' command.

Reviewed-by: Sage Weil <sage@redhat.com>
Signed-off-by: Yehuda Sadeh <yehuda@redhat.com>
2014-11-22 09:16:03 -08:00
Samuel Just
eb5d2117fd Merge pull request #2895 from yuyuyu101/wip-10022
AsyncMessenger: Fix several problems

Reviewed-by: Sage Weil <sage@redhat.com>
2014-11-21 16:18:25 -08:00
Josh Durgin
6070f4ad68 Merge pull request #2980 from ceph/wip-10123
librbd: protect list_children from invalid child pool IoCtxs

Reviewed-by: Josh Durgin <jdurgin@redhat.com>
2014-11-21 14:35:24 -08:00
Haomai Wang
e0ab267c75 Merge pull request #2981 from yuyuyu101/quick-fix
GenericObjectMap: Avoid master branch compile error
2014-11-21 17:27:10 +08:00
Haomai Wang
25fe088b75 GenericObjectMap: Fix typo
Signed-off-by: Haomai Wang <haomaiwang@gmail.com>
2014-11-21 17:25:06 +08:00
Sage Weil
4884d3e3d1 Merge pull request #2966 from yuyuyu101/wip-10119
KeyValueStore: Fix EC+ KV OSDs crashing

Reviewed-by: Sage Weil <sage@redhat.com>
2014-11-20 21:41:23 -08:00
Sage Weil
7d299528b5 Merge pull request #2978 from majianpeng/fix4
blkdev: using strncpy instead of strcpy.

Reviewed-by: Sage Weil <sage@redhat.com>
2014-11-20 20:51:38 -08:00
Jianpeng Ma
6c7bb8c068 blkdev: using strncpy instead of strcpy.
Coverity Scan reported this bug:
> New defect(s) Reported-by: Coverity Scan Showing 1 of 1 defect(s)
> *** CID 1255369:  Copy into fixed size buffer  (STRING_OVERFLOW)
> /common/blkdev.cc: 34 in block_device_support_discard(const char *)()
> 28     bool block_device_support_discard(const char *devname)
> 29     {
> 30       bool can_trim = false;
> 31       char *p = strstr((char *)devname, "sd");
> 32       char name[32] = {0};
> 33
> >>>     CID 1255369:  Copy into fixed size buffer  (STRING_OVERFLOW)
> >>>     You might overrun the 32 byte fixed-size string "name" by
> >>>     copying "p" without checking the length.
> 34       strcpy(name, p);
> 35       for (unsigned int i = 0; i < strlen(name); i++) {
> 36         if(isdigit(name[i])) {
> 37           name[i] = 0;
> 38           break;
> 39         }
>

Signed-off-by: Jianpeng Ma <jianpeng.ma@intel.com>
2014-11-21 09:55:02 +08:00
Sage Weil
cb001e7a6b Merge pull request #2974 from ceph/wip-sepia-update
Updated sepia hardware list.
2014-11-20 11:34:36 -08:00
Sandon Van Ness
627f138a72 Updated sepia hardware list.
Added missing details and additional list of hardware.

Signed-off-by: Sandon Van Ness <sandon@redhat.com>
2014-11-20 11:13:38 -08:00
Samuel Just
257ae0dc21 Merge pull request #2752 from dachary/wip-9817-deep-scrub-start
osd: log when scrub,deep scrub, repair starts

Reviewed-by: Samuel Just <sjust@redhat.com>
2014-11-20 09:45:33 -08:00
Samuel Just
e60ff7a1d0 Merge pull request #2891 from XinzeChi/master
OSD: batch trim for pg_log

Reviewed-by: Samuel Just <sjust@redhat.com>
Reviewed-by: Sage Weil <sage@redhat.com>
2014-11-20 09:44:09 -08:00
Samuel Just
c25e9126d6 Merge pull request #2885 from ceph/wip-flush-snaps
osd/ReplicatedPG: flush snaps immediately

Reviewed-by: Samuel Just <sjust@redhat.com>
2014-11-20 09:42:47 -08:00
Loic Dachary
b14dc28fbc Merge pull request #2973 from wonzhq/unused-parm
ReplicatedPG: remove unused parameter in function write_update_size_and_usage

Reviewed-by: Loic Dachary <ldachary@redhat.com>
2014-11-20 11:19:37 +01:00
Zhiqiang Wang
7a868fd64b ReplicatedPG: remove unused parameter in function write_update_size_and_usage
Signed-off-by: Zhiqiang Wang <zhiqiang.wang@intel.com>
2014-11-20 16:26:32 +08:00
Haomai Wang
5177759cd1 KeyValueStore: Fix parse_header_key
Signed-off-by: Haomai Wang <haomaiwang@gmail.com>
2014-11-20 11:03:17 +08:00
Samuel Just
47d6780849 Merge pull request #2926 from ceph/wip-10052
ceph_test_rados_api_tier: fix whiteout cleanup (10052)

Reviewed-by: Samuel Just <sjust@redhat.com>
2014-11-19 11:49:07 -08:00
David Zafman
9877beae0f Merge pull request #2967 from ceph/wip-10077
ceph_objectstore_tool shard feature handling

 Reviewed-by: Loic Dachary <ldachary@redhat.com>
2014-11-19 11:37:22 -08:00
Loic Dachary
0ea29a55d4 Merge pull request #2971 from dachary/wip-mailmap
mailmap: Loic Dachary name normalization
2014-11-19 20:30:38 +01:00
Loic Dachary
92662a9b57 mailmap: Loic Dachary name normalization
Signed-off-by: Loic Dachary <ldachary@redhat.com>
2014-11-19 20:25:24 +01:00
Loic Dachary
cb60a3e8dd Merge pull request #2962 from ceph/wip-9439
osd: Check filter ops for pgls and pgnls

Reviewed-by: Loic Dachary <ldachary@dachary.org>
2014-11-19 19:27:15 +01:00
Loic Dachary
5997635d0b Merge pull request #2969 from dachary/wip-erasure-code-documentation
doc: erasure-code developer documentation typo

Reviewed-by: Pavan Rallabhandi <Pavan.Rallabhandi@sandisk.com>
2014-11-19 11:24:01 +01:00
Loic Dachary
a4b42d66a0 doc: erasure-code developer documentation typo
Reported-by: Pavan Rallabhandi <Pavan.Rallabhandi@sandisk.com>
Signed-off-by: Loic Dachary <ldachary@redhat.com>
2014-11-19 11:06:31 +01:00
Haomai Wang
2601a25a2e StoreTest: Add scrub situation to tests
Signed-off-by: Haomai Wang <haomaiwang@gmail.com>
2014-11-19 15:07:31 +08:00
Haomai Wang
b4fa384fc0 KeyValueStore: Add KEY_ENDING sign to the end of key
Keys stored in alphabetical order and need to follow ghobject_t comparison
rule. "generation" and "shard_id" are optional fields for object key, but
a default ghobject with UINT64_MAX generation(by default) will larger than
the same ghobject with other generation. GenericObjectMap rejects to store
generation if generation is UINT64_MAX in order to reduce too much words
in key. So we need to add a MAX sign to the end of key to make ordering
is same with ghobject's comparison rule.

For example:

_GHOBJTOSEQ_:1%e1ds0_head!D168A7E8!!1!!benchmark_last_metadata!head
_GHOBJTOSEQ_:1%e1ds0_head!D168A7E8!!1!!benchmark_last_metadata!head!78!0

The first key should larger than the second in ghobject_t ordering because
of generation. But the first key less than the second in GenericObjectMap.

_GHOBJTOSEQ_:1%e1ds0_head!D168A7E8!!1!!benchmark_last_metadata!head
_GHOBJTOSEQ_:1%e1ds0_head!D168A7E8!!1!!benchmark_last_metadata!head!78!0

After we add KEY_ENDING, '!' is (21) in hexadecimal:

_GHOBJTOSEQ_:1%e1ds0_head!D168A7E8!!1!!benchmark_last_metadata!head(21)78!0(FF)
_GHOBJTOSEQ_:1%e1ds0_head!D168A7E8!!1!!benchmark_last_metadata!head(FF)

Fix bug #10119
Signed-off-by: Haomai Wang <haomaiwang@gmail.com>
2014-11-19 15:07:16 +08:00
Sage Weil
9c1b199aee Merge pull request #2953 from yuyuyu101/fix-keyvaluestore-remove-col-attr
KeyValueStore: Remove assert for collection_getattr method

Reviewed-by: Sage Weil <sage@redhat.com>
2014-11-18 21:49:31 -08:00
Haomai Wang
b2e82543f9 StoreTest: Clean up warning
Signed-off-by: Haomai Wang <haomaiwang@gmail.com>
2014-11-19 13:11:37 +08:00
Haomai Wang
dcecfb89fe MemStore: Return -ENODATA when collection_getattr hit nonexist attr
Signed-off-by: Haomai Wang <haomaiwang@gmail.com>
2014-11-19 13:11:31 +08:00
Haomai Wang
7a9b75577b KeyValueStore: Fix collection_getattrs problem when zero-length attr
Signed-off-by: Haomai Wang <haomaiwang@gmail.com>
2014-11-19 13:11:23 +08:00
Haomai Wang
877237f096 KeyValueStore: Return -ENODATA if collection_getattr is nonexist
Signed-off-by: Haomai Wang <haomaiwang@gmail.com>
2014-11-19 11:53:12 +08:00
Haomai Wang
00b275b0ae StoreTest: Add collection_getattr(s) tests
Signed-off-by: Haomai Wang <haomaiwang@gmail.com>
2014-11-19 11:52:52 +08:00
Haomai Wang
ffb6f78fd0 KeyValueStore: Remove assert for collection_getattr method
Merge 22b6c1a531 introduce
5dc990c1ca commit which will lookup "remove"
attr. It shouldn't exist for most of cases but KeyValueStore will crash
if attr isn't exist. It seemed that this assert isn't correct.

Signed-off-by: Haomai Wang <haomaiwang@gmail.com>
2014-11-19 11:52:33 +08:00
David Zafman
f3dab446fc ceph_objectstore_tool: Add feature called set-allow-sharded-objects
Uses --op set-allow-sharded-objects option
This operation will be rejected if on the target OSD's osdmap there is
    at least one OSD which does not support ERASURE CODES.
Prompt the user that they could import if sharded state allowed
Prompt the user to use new feature if sharded state found inconsistent

Fixes: #10077

Signed-off-by: David Zafman <dzafman@redhat.com>
2014-11-18 19:09:41 -08:00
David Zafman
b3021b0d33 ceph_objectstore_tool: Add utility routine get_osdmap()
Signed-off-by: David Zafman <dzafman@redhat.com>
2014-11-18 19:04:21 -08:00
David Zafman
86baf2d381 ceph_objectstore_tool: Clear ...INCOMPAT_SHARDS from feature if exporting replicated pg
Don't require importing OSD to have shards feature for replicated pg

http://tracker.ceph.com/issues/10077 Fixes: #10077

Signed-off-by: David Zafman <dzafman@redhat.com>
2014-11-18 19:04:11 -08:00
Loic Dachary
4f37c50074 Merge pull request #2648 from dachary/wip-9665-ceph-disk-partprobe
ceph-disk: run partprobe/partx after zap and data partition creation

Reviewed-by: Sage Weil <sage@redhat.com>
2014-11-19 02:54:10 +01:00