1 commit 95292699291242794510b39ffde3f4df67898d3a (tag: refs/tags/v0.94.4, refs/remotes/gh/hammer)
2 Author: Jenkins <jenkins@ceph.com>
3 Date: Mon Oct 19 07:43:50 2015 -0700
7 commit 5764e233e56be08a59ffe6292f6fba9a76288aee
9 Author: Sage Weil <sage@redhat.com>
10 Date: Thu Oct 15 16:35:10 2015 -0400
12 Merge pull request #6287 from jcsp/wip-test-fixes
14 git path backports for hammer
16 commit b2039797638057dd74a8a47c99091c2c892b042e
17 Author: Sage Weil <sage@redhat.com>
18 Date: Thu Apr 16 08:29:41 2015 -0700
20 use git://git.ceph.com
22 Signed-off-by: Sage Weil <sage@redhat.com>
23 (cherry picked from commit 34c467ba06345eacd2fd0e79c162dfd0b22f0486)
25 commit 0f4ef19047c58aacfb359bdd9d104f059aa2f5b4
26 Author: Sage Weil <sage@redhat.com>
27 Date: Thu Sep 24 18:10:46 2015 -0400
29 qa: http://ceph.com/qa -> http://download.ceph.com/qa
31 Signed-off-by: Sage Weil <sage@redhat.com>
32 (cherry picked from commit d4d65fbd6ad15320339751c604613ac63511e645)
34 commit 7f485ed5aa620fe982561663bf64356b7e2c38f2
35 Merge: 3dc1de2 294f016
36 Author: Loic Dachary <loic@dachary.org>
37 Date: Tue Oct 13 00:28:52 2015 +0200
39 Merge pull request #6161 from dachary/wip-13227-hammer
41 With root as default user, unable to have multiple RGW instances running
43 Reviewed-by: Sage Weil <sage@redhat.com>
44 Reviewed-by: Yehuda Sadeh <ysadehwe@redhat.com>
45 Reviewed-by: Ken Dreyer <kdreyer@redhat.com>
47 commit 294f016ff84c7efef36ca94daef39fe2686df690
48 Author: Sage Weil <sage@redhat.com>
49 Date: Thu May 7 15:40:05 2015 -0700
51 init-radosgw.sysv: remove
53 Signed-off-by: Sage Weil <sage@redhat.com>
54 (cherry picked from commit 9d8c115b0ab05cfc1db6670d1d367737e61fe0ea)
56 commit 3dc1de22552065bd930cc3e42a486e29639927ba
57 Merge: aa74de9 7de65e7
58 Author: Yehuda Sadeh <yehuda@redhat.com>
59 Date: Mon Oct 12 08:16:17 2015 -0700
61 Merge pull request #6166 from dachary/wip-13046-hammer
63 RGW : setting max number of buckets for user via ceph.conf option
65 commit aa74de9d34801552d7d3e0aced07b6bc5bd129b9
66 Merge: d3e49eb 77cb503
67 Author: Yehuda Sadeh <yehuda@redhat.com>
68 Date: Mon Oct 12 08:15:21 2015 -0700
70 Merge pull request #6039 from ceph/wip-13015-hammer
72 rgw: don't preserve acls when copying object
74 commit d3e49ebce00d749c36fe8dc46a5c432b1b907126
75 Merge: 417e948 424fc1c
76 Author: Yehuda Sadeh <yehuda@redhat.com>
77 Date: Mon Oct 12 08:14:47 2015 -0700
79 Merge pull request #6042 from dreamhost/wip-hammer-crossftp
81 rgw: remove trailing :port from host for purposes of subdomain matching
83 commit 417e948898d260a1ad6354244f9967ccbd75d768
84 Merge: b5badcb 9ab9c44
85 Author: Yehuda Sadeh <yehuda@redhat.com>
86 Date: Mon Oct 12 08:14:09 2015 -0700
88 Merge pull request #6160 from dachary/wip-13228-hammer
90 rgw: segments are read during HEAD on Swift DLO
92 commit b5badcb47e565cd17145adca49abf07e167963e0
93 Merge: e3f8df9 6119b15
94 Author: Yehuda Sadeh <yehuda@redhat.com>
95 Date: Mon Oct 12 08:13:25 2015 -0700
97 Merge pull request #6162 from dachary/wip-13226-hammer
99 Keystone Fernet tokens break auth
101 commit e3f8df9e12b434714b67036838f37b75ed1f9a25
102 Merge: 8078233 ad83304
103 Author: Yehuda Sadeh <yehuda@redhat.com>
104 Date: Mon Oct 12 08:12:26 2015 -0700
106 Merge pull request #6163 from dachary/wip-13054-hammer
108 rgw: region data still exist in region-map after region-map update
110 commit 8078233ca1dda8eef3c59d6a92760365634cf41a
111 Merge: dd64c78 4b0686f
112 Author: Yehuda Sadeh <yehuda@redhat.com>
113 Date: Mon Oct 12 08:12:07 2015 -0700
115 Merge pull request #6164 from dachary/wip-13053-hammer
117 rgw: GWWatcher::handle_error -> common/Mutex.cc: 95: FAILED assert(r == 0)
119 commit dd64c7813bbdd2a6a4ffde6767c5d060ee1b9290
120 Merge: 135c112 e80bd0a
121 Author: Yehuda Sadeh <yehuda@redhat.com>
122 Date: Mon Oct 12 08:11:16 2015 -0700
124 Merge pull request #5718 from dachary/wip-12859-hammer
126 testGetContentType and testHead failed
128 commit 135c112789d4925a735229394c0ec9e2b728d3ca
129 Merge: fd17f3c 3b2affc
130 Author: Yehuda Sadeh <yehuda@redhat.com>
131 Date: Mon Oct 12 08:10:55 2015 -0700
133 Merge pull request #5860 from lebauce/wip-12960-hammer
135 rgw: add delimiter to prefix only when path is specified
137 commit fd17f3ccbf98832098bd1cd35fe2a5d567b51f4c
138 Merge: 036c718 297c04d
139 Author: Yehuda Sadeh <yehuda@redhat.com>
140 Date: Mon Oct 12 08:08:59 2015 -0700
142 Merge pull request #6165 from dachary/wip-13052-hammer
144 rgw: init_rados failed leads to repeated delete
146 commit 036c7186f9304e4e48ecf50a4299cf74d55875fe
147 Merge: 629b631 698d75c
148 Author: Loic Dachary <loic@dachary.org>
149 Date: Fri Oct 9 09:17:19 2015 +0200
151 Merge pull request #6201 from dachary/wip-13410-hammer
153 TEST_crush_rule_create_erasure consistently fails on i386 builder
155 Reviewed-by: Kefu Chai <kchai@redhat.com>
157 commit 629b631488f044150422371ac77dfc005f3de1bc
158 Merge: 70e612a 0a5b856
159 Author: Sage Weil <sage@redhat.com>
160 Date: Thu Oct 8 12:18:49 2015 -0400
162 Merge pull request #5885 from Abhishekvrshny/wip-13034-hammer
164 osd: copy-from doesn't preserve truncate_{seq,size}
166 Reviewed-by: Sage Weil <sage@redhat.com>
168 commit 698d75c11fa116d6546b09d14484d094fae8f6d7
169 Author: Loic Dachary <ldachary@redhat.com>
170 Date: Wed Jul 22 10:42:32 2015 +0200
172 tests: robust test for the pool create crushmap test
174 The test that goes with f1e86be589803596e86acc964ac5c5c03b4038d8 to
175 verify that a bugous crush ruleset will prevent the creation of a pool
176 trying to use it was fragile. I depends on the implementation of the
177 erasure code lrc plugin and turns out to not work on i386.
179 The test is modified to use a fake crushtool that always returns false
180 and validate that it prevents the creation of a pool, which demonstrate
181 it is used for crushmap validation prior to the pool creation.
183 http://tracker.ceph.com/issues/12419 Fixes: #12419
185 Signed-off-by: Loic Dachary <ldachary@redhat.com>
186 (cherry picked from commit e47903727b5d53f795014bf29b417d357f2bd65a)
188 commit 70e612a395f7f5daf2fa289d8ae522824ef7a38f
189 Merge: 0650862 2a28114
190 Author: Loic Dachary <loic@dachary.org>
191 Date: Thu Oct 8 00:09:24 2015 +0200
193 Merge pull request #6192 from dachary/wip-13401-hammer
195 mon: fix crush testing for new pools
197 Reviewed-by: Sage Weil <sage@redhat.com>
199 commit 2a28114053fbf4ceee09558790ad3cf755270d5f
200 Author: Sage Weil <sage@redhat.com>
201 Date: Fri Aug 21 16:40:34 2015 -0400
203 crush/CrushTester: test fewer inputs when running crushtool
205 If there are a lot of crush rules (say, 100) then the test can
206 take a long time. 100 values per rule should be enough to catch
209 Signed-off-by: Sage Weil <sage@redhat.com>
210 (cherry picked from commit 0f82f461b33d93d868e185912a2c7e4074d06900)
213 src/crush/CrushTester.cc
214 in hammer the crushtool validation is via a shell
215 and not via an internal subprocess utility
217 commit abc5b5f139b05da0be0fbaa99282be16386980db
218 Author: Loic Dachary <ldachary@redhat.com>
219 Date: Sun Sep 20 23:42:45 2015 +0200
221 tests: update to match crushmap validation message
223 http://tracker.ceph.com/issues/13182 Fixes: #13182
225 Signed-off-by: Loic Dachary <ldachary@redhat.com>
226 (cherry picked from commit 21a1e75d8a7bad89a48cd9d36902c5d609be5015)
229 qa/workunits/cephtool/test.sh
230 the test structure in hammer is different, adapt
233 commit 25bd2778f8b7f5eb5245efebca56a7348dd064eb
234 Author: Sage Weil <sage@redhat.com>
235 Date: Fri Sep 18 09:42:47 2015 -0400
237 mon/OSDMonitor: fix crush injection error message
239 Signed-off-by: Sage Weil <sage@redhat.com>
240 (cherry picked from commit 1b3090d50e5bd5ca3e6e396b23d2d9826896c718)
243 src/mon/OSDMonitor.cc
244 hammer has a separate error message for EINTR
245 which is in the context of the fix for error message
247 commit 6635530aa94b1f21c3052e64d4dfe3952f98fe83
248 Author: Sage Weil <sage@redhat.com>
249 Date: Fri Sep 18 09:41:25 2015 -0400
251 mon/OSDMonitor: only test crush ruleset for the newly created pool
253 Otherwise, we test *all* crush rules.. which might be a lot, and
254 which is a big waste of time and effort.
256 Signed-off-by: Sage Weil <sage@redhat.com>
257 (cherry picked from commit 524b0bdcc45c2f4b95f2239c988e93250f337f3d)
260 src/mon/OSDMonitor.cc : trivial resolution
262 commit cc1fedda02fbf3c95d7a00f0b395db5439e96f90
263 Author: Sage Weil <sage@redhat.com>
264 Date: Fri Sep 18 09:40:13 2015 -0400
266 crush/CrushTester: allow testing by ruleset
268 Signed-off-by: Sage Weil <sage@redhat.com>
269 (cherry picked from commit aa238e5ed50f44a94caf84567267e4f6be8732a2)
272 src/crush/CrushTester.cc
273 in hammer the crushtool validation is via a shell
274 and not via an internal subprocess utility
275 src/tools/crushtool.cc
276 ceph_argparse_withint is preferred to ceph_argparse_witharg
278 commit 065086246290cfa02f67c37c95a2ccdb05872f21
279 Merge: 79385a8 3a50b90
280 Author: Sage Weil <sage@redhat.com>
281 Date: Tue Oct 6 11:11:14 2015 -0400
283 Merge pull request #5887 from Abhishekvrshny/wip-13044-hammer
285 LibCephFS.GetPoolId failure
287 Reviewed-by: Sage Weil <sage@redhat.com>
289 commit 79385a85beea9bccd82c99b6bda653f0224c4fcd
290 Merge: 5b25fc9 3228161
291 Author: Loic Dachary <loic@dachary.org>
292 Date: Sun Oct 4 15:33:49 2015 +0200
294 Merge pull request #6172 from dachary/wip-13354-hammer
296 qa/workunits/cephtool/test.sh: don't assume crash_replay_interval=45
298 Reviewed-by: Loic Dachary <ldachary@redhat.com>
300 commit 322816116ea14a8fd79616b1c55545ae7c62e7f6
301 Author: Sage Weil <sage@redhat.com>
302 Date: Mon Sep 28 13:13:40 2015 -0400
304 qa/workunits/cephtool/test.sh: don't assume crash_replay_interval=45
306 e.g., it's 5 in teuthology's ceph.conf.
308 Signed-off-by: Sage Weil <sage@redhat.com>
309 (cherry picked from commit 80b7237e5e74d12b9d8b1f96ea535c96bdff9c6f)
311 commit ad833044dee4208625213bc83c7847ce9d1a73aa
312 Author: dwj192 <duanweijun@h3c.com>
313 Date: Wed Sep 9 09:48:16 2015 +0800
315 rgw:add --reset-regions for regionmap update
319 Signed-off-by: Weijun Duan <duanweijun@h3c.com>
320 (cherry picked from commit 95685c19d6f1eab50b903e61273b5351bedc2980)
323 src/rgw/rgw_admin.cc: usage is on cerr in hammer, not cout
324 src/test/cli/radosgw-admin/help.t: remove extra empty line
326 commit 5b25fc9c8431423c6239da2eafcf7e8863d2262f
327 Merge: e72bdc3 9026c4a
328 Author: Loic Dachary <loic@dachary.org>
329 Date: Sat Oct 3 09:39:12 2015 +0200
331 Merge pull request #6156 from dachary/wip-13170-hammer
333 doc: update docs to point to download.ceph.com
335 Reviewed-by: Abhishek Varshney <abhishek.varshney@flipkart.com>
337 commit 7de65e7ede52f3e417c354f180cfd80f1407c6cb
338 Author: Vikhyat Umrao <vumrao@redhat.com>
339 Date: Thu Sep 3 12:02:05 2015 +0530
341 rgw : setting max number of buckets for users via ceph.conf option
343 This patch adds a new option "rgw_user_max_buckets" for setting
344 max number of buckets for users via ceph.conf.
348 Signed-off-by: Vikhyat Umrao <vumrao@redhat.com>
349 (cherry picked from commit f65267c96cbd4cd25036b6bf399692e77bbb9436)
352 src/rgw/rgw_admin.cc: usage is on cerr in hammer, not cout
354 commit 297c04d856c4864d21b2241ad01560afa61dde40
355 Author: Xiaowei Chen <cxwshawn@gmail.com>
356 Date: Tue Sep 8 06:58:57 2015 -0400
358 rgw: init_rados failed leads to repeated delete
362 Signed-off-by: Xiaowei Chen <chen.xiaowei@h3c.com>
363 (cherry picked from commit ab4232baa7bcc86e90746e13312ac9bda1772872)
365 commit 4b0686f18f323b95adb1fc5cffe7c1e55e8f1c63
366 Author: Yehuda Sadeh <yehuda@redhat.com>
367 Date: Wed Aug 26 15:08:18 2015 -0700
369 rgw: delete finisher only after finalizing watches
373 The watch error path might try to schedule a finisher work, delete finisher
374 only after watch destruction.
376 Signed-off-by: Yehuda Sadeh <yehuda@redhat.com>
377 (cherry picked from commit 97aed59fb9e865d30d31d2b7f4e93fc9727c96fa)
379 commit 6119b152412ddfa2a614c1862aad52a123a4fe8e
380 Author: Abhishek Lekshmanan <abhishek.lekshmanan@ril.com>
381 Date: Mon Aug 24 23:11:35 2015 +0530
383 rgw: be more flexible with iso8601 timestamps
385 make parsing 8601 more flexible by not restricting the length of seconds
386 to 5, this allows timestamp to be specified both as ms or us. Newer
387 keystone backends such as fernet token backend default to microseconds
388 when publishing iso8601 timestamps, so this allows these timestamps to
389 be allowed when specifying the token expiry time.
392 Reported-by: Ian Unruh <ianunruh@gmail.com>
393 Signed-off-by: Abhishek Lekshmanan <abhishek.lekshmanan@ril.com>
394 (cherry picked from commit 136242b5612b8bbf260910b1678389361e86d22a)
396 commit 607904e8d5616bd7df20e1e387a14d55e4ca43b6
397 Author: Sage Weil <sage@redhat.com>
398 Date: Tue Sep 22 14:12:02 2015 -0400
400 init-radosgw: specify pid file to start-stop-daemon
402 This fixes restart when multiple instances are running.
405 Tested-by: Pavan Rallabhandi <pavan.rallabhandi@sandisk.com>
406 Signed-off-by: Sage Weil <sage@redhat.com>
407 (cherry picked from commit e17c8e1a7a06db0e2680dedc7e954021610320d1)
409 commit f51ab26d8721bde6852e5c169869d7ae16bd0574
410 Author: Vikhyat Umrao <vumrao@redhat.com>
411 Date: Thu Jul 23 18:10:06 2015 +0530
413 rgw: fix radosgw start-up script.
415 radosgw init script is unable to start radosgw daemon.
416 as it is relies on requiretty being disabled.
417 once init script start this daemon with sudo it fails
420 changing 'sudo' to 'su', it will fix this issue and
421 will also help running radosgw daemon with our new
426 Signed-off-by: Vikhyat Umrao <vumrao@redhat.com>
427 (cherry picked from commit c83542296aeb8f5dd96f5d9e09327e7a89370742)
429 commit 544a98fa9dd47d02778f622de5f7ebe1acccc12f
430 Author: Sage Weil <sage@redhat.com>
431 Date: Thu May 7 15:39:22 2015 -0700
433 init-radosgw: unify init-radosgw[.sysv]
435 Signed-off-by: Sage Weil <sage@redhat.com>
436 (cherry picked from commit 1c45f512a91ee49e9253ba6fd3857af9bd2ae57a)
438 commit 2a733e9e720bcdb8ea342a96be679d6f9efa3b2b
439 Author: Sage Weil <sage@redhat.com>
440 Date: Thu May 7 15:29:11 2015 -0700
442 init-radosgw: look in /var/lib/ceph/radosgw
444 Signed-off-by: Sage Weil <sage@redhat.com>
445 (cherry picked from commit a4bb992426b39188e750fce9198052ca0d510117)
447 commit d00c52b874ddc74aa5f4c2e07b70b881ce963114
448 Author: Nathan Cutler <ncutler@suse.cz>
449 Date: Sun May 10 18:17:53 2015 +0200
451 doc: rgw: fix typo in comments
453 Signed-off-by: Nathan Cutler <ncutler@suse.cz>
454 (cherry picked from commit 05cf0db22dc3a9cb2bf4b86b8f240683c23f41ac)
456 commit eb001d3a7275cc040bea716b8bdb0cf8b97efbc9
457 Author: Dmitry Yatsushkevich <dyatsushkevich@mirantis.com>
458 Date: Tue Mar 17 15:44:52 2015 -0700
460 rgw: init script waits until the radosgw stops
463 Init script waits in stop action until the radowgw daemons stop.
465 Signed-off-by: Dmitry Yatsushkevich <dyatsushkevich@mirantis.com>
466 (cherry picked from commit 1cca0c1e91a1e18f82f4d22855e96b2fc947f5ea)
468 commit 9ab9c44583c96f1679db3dbcef322a706548a7dd
469 Author: Yehuda Sadeh <yehuda@redhat.com>
470 Date: Thu Sep 17 09:10:39 2015 -0700
472 rgw: don't read actual data on user manifest HEAD
475 We unconditionally read all the data, which is not needed
476 when doing HEAD operation on user manifest objects.
478 Signed-off-by: Yehuda Sadeh <yehuda@redhat.com>
479 Signed-off-by: Radoslaw Zarzynski <rzarzynski@mirantis.com>
480 (cherry picked from commit 4a8b08d1dd2511e28e999f1625103fb919fb5465)
482 commit 9026c4a7c7699a9a013d4fe207e6681403f934a6
483 Author: Alfredo Deza <adeza@redhat.com>
484 Date: Fri Sep 18 14:20:12 2015 -0400
486 doc: remove mention of ceph-extra as a requirement
488 Signed-off-by: Alfredo Deza <adeza@redhat.com>
489 (cherry picked from commit a6f07e9fa3bd5e50a2d8cfa387b44a39d383fa65)
491 commit 45ed24da879fe12b921cf03e4a921ccc0582b8d2
492 Author: Alfredo Deza <adeza@redhat.com>
493 Date: Fri Sep 18 14:13:21 2015 -0400
495 doc: remove ceph-extras
497 Signed-off-by: Alfredo Deza <adeza@redhat.com>
498 (cherry picked from commit c0ef84fd22aac58b6ec8e58eb4f8dffed306d39e)
500 commit faccdce79e003318c751dd9203773af7ed31d052
501 Author: Alfredo Deza <adeza@redhat.com>
502 Date: Fri Sep 18 14:13:02 2015 -0400
504 doc: correct links to download.ceph.com
506 Signed-off-by: Alfredo Deza <adeza@redhat.com>
507 (cherry picked from commit 387d7800359154950431d0984c756f43f21dd9b4)
509 commit e9f4aecf3f406b42b74f840edb398be9243da85a
510 Author: ritz303 <ritz_303@yahoo.com>
511 Date: Tue Aug 25 10:02:49 2015 -0500
513 doc: Added "Hammer" in the list of major releases.
517 Signed-off-by: ritz303 <ritz_303@yahoo.com>
518 (cherry picked from commit 582f0f64455079290ad3b4ae7338b716e170911f)
520 commit 424fc1ccb244f5084f94904cb20cbf88a1bd4638
521 Author: Yehuda Sadeh <yehuda@redhat.com>
522 Date: Fri Sep 25 10:44:40 2015 -0700
524 rgw: set default value for env->get() call
528 This fixes a regression introduced at commit abe4ec293d08b0314bf5c081ace2456073f3a22c.
529 The host var is a string, env->get() returns a char pointer, shouldn't
532 Signed-off-by: Yehuda Sadeh <yehuda@redhat.com>
533 (cherry picked from commit 0d5730efc0054334897315dc23ba04f30548e36e)
534 Signed-off-by: Robin H. Johnson <robin.johnson@dreamhost.com>
536 commit e72bdc3bf1541e445f74def70b5aa185f1757468
537 Author: Sage Weil <sage@redhat.com>
538 Date: Mon Sep 21 11:41:03 2015 -0400
540 osd/ReplicatedPG: tolerate promotion completion with stopped agent
542 We may start a promotion, then get a pool update that disables the
543 agent, and then complete it. We should not segfault in this case.
546 Signed-off-by: Sage Weil <sage@redhat.com>
547 (cherry picked from commit fd9ce66e8b76787f69c5e337332d57aa72c8b8d5)
549 commit a3afb3f59435050efa711436134b4abe63a8f5cf
550 Author: Sage Weil <sage@redhat.com>
551 Date: Tue Jun 9 14:15:10 2015 -0400
553 rgw: remove trailing :port from host for purposes of subdomain matching
555 Some clients (ahem, CrossFTP) include the :port in the HTTP_HOST header.
558 Switch req_info field to a std::string and avoid copying it in preprocess.
560 Signed-off-by: Sage Weil <sage@redhat.com>
561 (cherry picked from commit abe4ec293d08b0314bf5c081ace2456073f3a22c)
563 commit fe0fc4f18f6f275b34e8bbfa868a7310ecc3842f
564 Merge: e26f8bc 8ba6b2f
565 Author: Yehuda Sadeh <yehuda@redhat.com>
566 Date: Thu Sep 24 14:01:26 2015 -0700
568 Merge pull request #5715 from dachary/wip-12592-hammer
570 RGW returns requested bucket name raw in Bucket response header
572 commit e26f8bc2dbc82b6ff31d2bd2d5c890aa12da9f4a
573 Merge: 19e549e 6acf36f
574 Author: Yehuda Sadeh <yehuda@redhat.com>
575 Date: Thu Sep 24 14:00:55 2015 -0700
577 Merge pull request #5719 from dachary/wip-12855-hammer
579 segmentation fault when rgw_gc_max_objs > HASH_PRIME
581 commit 19e549e9c2fbacdbf747344c1dafcc6b66902339
582 Merge: 2b11a4c a13c7fd
583 Author: Yehuda Sadeh <yehuda@redhat.com>
584 Date: Thu Sep 24 14:00:09 2015 -0700
586 Merge pull request #5720 from dachary/wip-12854-hammer
588 the arguments 'domain' should not be assigned when return false
590 commit 77cb503ba057f033825aa40d57dee42e838c67ac (refs/remotes/gh/wip-13015-hammer)
591 Author: Yehuda Sadeh <yehuda@redhat.com>
592 Date: Wed Sep 9 15:41:02 2015 -0700
594 rgw: preserve all attrs if intra-zone copy
597 Intra zone copy requires that all objects' attributes are preserved.
599 Signed-off-by: Yehuda Sadeh <yehuda@redhat.com>
601 commit b9f2ed303eedfa0b747884f8e66fbe97cfeeb0d5
602 Author: Yehuda Sadeh <yehuda@redhat.com>
603 Date: Thu Aug 13 10:59:32 2015 -0700
605 rgw: don't preserve acls when copying object
609 When copying an object we need to use the acls we calculated earlier,
610 and not the source acls.
611 This was broken at e41d97c8e38bb60d7e09e9801c0179efe7af1734.
613 Signed-off-by: Yehuda Sadeh <yehuda@redhat.com>
614 (cherry picked from commit fb020247449d3daf033885bab5b3a10c0e5df4a5)
616 commit 2b11a4c07a9416a2f75c8ce62a923994977d799c
617 Merge: 0022c03 b3822f1
618 Author: Loic Dachary <loic@dachary.org>
619 Date: Tue Sep 15 10:23:49 2015 +0200
621 Merge pull request #5930 from liewegas/wip-11798-hammer
623 upstart: limit respawn to 3 in 30 mins (instead of 5 in 30s)
625 Reviewed-by: Loic Dachary <ldachary@redhat.com>
627 commit 0022c035476eff471a777b23a619d024f632a1bf
628 Merge: 23fb811 0d6a8c6
629 Author: Loic Dachary <loic@dachary.org>
630 Date: Mon Sep 14 23:57:13 2015 +0200
632 Merge pull request #5908 from liewegas/wip-simple-connect-hammer
634 Pipe: Drop connect_seq increase line
636 Reviewed-by: Greg Farnum <gfarnum@redhat.com>
638 commit b3822f113e07547194b844f647bcb7d45513b25f (refs/remotes/me/wip-11798-hammer)
639 Author: Sage Weil <sage@redhat.com>
640 Date: Wed Jun 3 14:57:34 2015 -0400
642 upstart: limit respawn to 3 in 30 mins (instead of 5 in 30s)
644 It may take tens of seconds to restart each time, so 5 in 30s does not stop
645 the crash on startup respawn loop in many cases. In particular, we'd like
646 to catch the case where the internal heartbeats fail.
648 This should be enough for all but the most sluggish of OSDs and capture
649 many cases of failure shortly after startup.
652 Signed-off-by: Sage Weil <sage@redhat.com>
653 (cherry picked from commit eaff6cb24ef052c54dfa2131811758e335f19939)
655 commit 0d6a8c631f55367ac4f6f792a10ba62cd97e9fed (refs/remotes/me/wip-simple-connect-hammer)
656 Author: Haomai Wang <haomaiwang@gmail.com>
657 Date: Wed Feb 25 23:38:54 2015 +0800
659 Pipe: Drop connect_seq increase line
661 Revert commit 0fc47e267b6f8dcd4511d887d5ad37d460374c25.
663 When accepting and "connect.connect_seq == existing->connect_seq",
664 existing->state maybe STATE_OPEN, STATE_STANDBY or STANDY_CONNECTING.
665 This commit only fix partial problem and want to assert
666 "(existing->state == STATE_CONNECTING)".
668 So later we added codes to catch
669 "(existing->state == STATE_OPEN || existing->state == STATE_STANDBY)"
672 Backport: dumpling, firefly, giant
674 Signed-off-by: Haomai Wang <haomaiwang@gmail.com>
675 (cherry picked from commit 67225cb3ee1f6d274a02293724942bdb25cec8ca)
677 commit 23fb811303971152f1c348e2a02de4e2bc6ed53d
678 Merge: c1849ec 6c4ccc8
679 Author: Loic Dachary <loic@dachary.org>
680 Date: Sun Sep 13 14:10:23 2015 +0200
682 Merge pull request #5767 from dachary/wip-12597-hammer
684 Crash during shutdown after writeback blocked by IO errors
686 Reviewed-by: Jason Dillaman <dillaman@redhat.com>
688 commit c1849ecfb2ae55b99b6db75ee49af19886c246b2
689 Merge: f35c53d f028389
690 Author: Sage Weil <sage@redhat.com>
691 Date: Sun Sep 13 07:42:36 2015 -0400
693 Merge pull request #5687 from liewegas/wip-hammer-feature-hammer
695 include/ceph_features: define HAMMER_0_94_4 feature
697 commit f35c53df7c155d7ff4a316a0817b11bc018c57ca
698 Merge: 6e22620 4be8a28
699 Author: Sage Weil <sage@redhat.com>
700 Date: Sat Sep 12 09:23:13 2015 -0400
702 Merge pull request #5892 from ceph/wip-13060-hammer
704 osd: allow peek_map_epoch to return an error
706 Reviewed-by: David Zafman <dzafman@redhat.com>
708 commit 4be8a28d5b5ba868d100300ab64ebb82b188222b
709 Author: Sage Weil <sage@redhat.com>
710 Date: Fri Sep 11 11:00:38 2015 -0400
712 osd/PG: peek_map_epoch: skip legacy PGs if infos object is missing
715 - osd is stopped before pg is fully removed
716 - on restart, we ignore/skip the pg because its epoch is too old
717 - we upgrade to hammer and convert other pgs, skipping this one, and then
718 remove the legacy infos object
719 - hammer starts, tries to parse the legacy pg, and fails because the infos
720 object is gone, crashing.
722 The fix is to continue ignoring the zombie pg.
725 Signed-off-by: Sage Weil <sage@redhat.com>
727 commit f237ed97228839a1b412ad213945f0343df05bf5
728 Author: Sage Weil <sage@redhat.com>
729 Date: Fri Sep 11 17:41:01 2015 -0400
731 osd: allow peek_map_epoch to return an error
733 Allow PG::peek_map_epoch to return an error indicating the PG
736 Signed-off-by: Sage Weil <sage@redhat.com>
737 (cherry picked from commit f15d9585edc5a12ac2d076951076247253b897c2)
739 [fixed *pepoch default of 0]
740 [fixed other return paths in peek_map_epoch]
742 commit 3a50b904a2ffd613b695ead1b26c93278044d7df
743 Author: Yan, Zheng <zyan@redhat.com>
744 Date: Wed Aug 5 15:19:13 2015 +0800
746 crypto: fix unbalanced ceph::crypto::init/ceph::crypto:shutdown
748 we may create a CephContext without calling common_init_finish(), then
749 delete the CephContext. In this case, ceph::crypto:init() is not called,
750 so CephContext::~CephContext() should not call ceph::crypto::shutdown().
753 Signed-off-by: Yan, Zheng <zyan@redhat.com>
754 (cherry picked from commit 98a85ec6464d8ec3fc7f0c647ac97c8cf002ebe2)
756 commit 0a5b8569ecc6efa42b3d4fa7a58c155b3c2dbbe8
757 Author: Samuel Just <sjust@redhat.com>
758 Date: Thu Jul 30 16:59:32 2015 -0700
760 ReplicatedPG,Objecter: copy_get should include truncate_seq and size
762 Otherwise, we break CephFS over cache tiers.
765 Signed-off-by: Samuel Just <sjust@redhat.com>
766 (cherry picked from commit 6f9ee7961eee9ee3a61a07cbe0d8d289ee98fa9a)
769 src/osd/ReplicatedPG.cc
771 commit 6e22620165872a88ab4b75ed2fb93f61beb65d12
772 Merge: 508733f 00e73ad
773 Author: Josh Durgin <jdurgin@redhat.com>
774 Date: Wed Sep 9 19:44:09 2015 -0400
776 Merge pull request #5769 from dachary/wip-12850-hammer
778 Crash during TestInternal.MultipleResize
780 Reviewed-by: Josh Durgin <jdurgin@redhat.com>
782 commit 508733fd48f2c704b79ad36841346965f4d0429b
783 Merge: 69a320e dc693fc
784 Author: Josh Durgin <jdurgin@redhat.com>
785 Date: Wed Sep 9 19:44:01 2015 -0400
787 Merge pull request #5768 from dachary/wip-12849-hammer
789 [ FAILED ] TestLibRBD.BlockingAIO
791 Reviewed-by: Josh Durgin <jdurgin@redhat.com>
793 commit 82ea02ab8d886c16c511c4bd7d03d826158f954b
794 Author: Yehuda Sadeh <yehuda@redhat.com>
795 Date: Wed Jun 17 11:35:18 2015 -0700
797 rgw: fix assignment of copy obj attributes
800 Clarify the confusing usage of set_copy_attrs() by switching the source and
801 destinatiion params (attrs, src_attrs). Switch to use attrs instead of
802 src_attrs afterwards. In one of the cases we originally used the wrong
805 Signed-off-by: Yehuda Sadeh <yehuda@redhat.com>
806 (cherry picked from commit e41d97c8e38bb60d7e09e9801c0179efe7af1734)
808 commit 3b2affce1a229cead3e0ffacfdfde69258a267cb
809 Author: Sylvain Baubeau <sbaubeau@redhat.com>
810 Date: Fri Sep 4 22:51:44 2015 +0200
812 rgw: add delimiter to prefix only when path is specified
814 http://tracker.ceph.com/issues/12960
817 Signed-off-by: Sylvain Baubeau <sbaubeau@redhat.com>
818 (cherry picked from commit 27cf257248ea55f8f4bc7851c3956611828bcae2)
820 commit 69a320e185f20da0824d0a62d3de77578c431a7a
821 Merge: 8039375 0bc909e
822 Author: Loic Dachary <loic@dachary.org>
823 Date: Wed Sep 9 16:58:56 2015 +0200
825 Merge pull request #5697 from tchaikov/wip-12638-hammer
827 mon: add a cache layer over MonitorDBStore
829 Reviewed-by: Joao Eduardo Luis <joao@suse.de>
831 commit 8039375327b5074a700193aae2d8b1922ca85398
832 Merge: 19ff928 256620e
833 Author: Loic Dachary <loic@dachary.org>
834 Date: Wed Sep 9 14:52:54 2015 +0200
836 Merge pull request #5381 from dachary/wip-12499-hammer
838 ceph-fuse 0.94.2-1trusty segfaults / aborts
840 Reviewed-by: Greg Farnum <gfarnum@redhat.com>
842 commit 19ff92806fd1e0fb866737f58e379aa8078b8017
843 Merge: 7614bf7 059bf98
844 Author: David Zafman <dzafman@redhat.com>
845 Date: Tue Sep 8 14:58:16 2015 -0700
847 Merge pull request #5757 from dachary/wip-12836-hammer
849 WBThrottle::clear_object: signal on cond when we reduce throttle values
851 Reviewed-by: David Zafman <dzafman@redhat.com>
853 commit 7614bf7d117fdd476931cdd8825265321cef3c4c
854 Merge: 2100631 bf72785
855 Author: Loic Dachary <loic@dachary.org>
856 Date: Mon Sep 7 10:06:30 2015 +0200
858 Merge pull request #5759 from dachary/wip-12841-hammer
860 recursive lock of md_config_t (0)
862 Reviewed-by: Sage Weil <sage@redhat.com>
864 commit 210063120708a7d6502c2b8d3e8c61731e37de5b
865 Merge: a2e1fe2 836f763
866 Author: Loic Dachary <loic@dachary.org>
867 Date: Mon Sep 7 10:05:41 2015 +0200
869 Merge pull request #5761 from dachary/wip-12843-hammer
871 long standing slow requests: connection->session->waiting_for_map->connection ref cycle
873 Reviewed-by: Sage Weil <sage@redhat.com>
875 commit a2e1fe26f602d0165999dfb72b97985f9bfe38ce
876 Merge: 8ccb771 2348a5b
877 Author: Loic Dachary <loic@dachary.org>
878 Date: Mon Sep 7 10:04:51 2015 +0200
880 Merge pull request #5762 from dachary/wip-12844-hammer
882 osd suicide timeout during peering - search for missing objects
884 Reviewed-by: Sage Weil <sage@redhat.com>
886 commit 8ccb7711d68dd3b95d2522c6e46f5021362c9088
887 Merge: e9db807 77624af
888 Author: Loic Dachary <loic@dachary.org>
889 Date: Mon Sep 7 10:04:03 2015 +0200
891 Merge pull request #5763 from dachary/wip-12846-hammer
893 osd/PGLog.cc: 732: FAILED assert(log.log.size() == log_keys_debug.size())
895 Reviewed-by: Sage Weil <sage@redhat.com>
897 commit e9db8075034dc4ac0ffe5d8dd9dd2153509237fa
898 Merge: 5ef999e aa00373
899 Author: Loic Dachary <loic@dachary.org>
900 Date: Mon Sep 7 10:03:10 2015 +0200
902 Merge pull request #5764 from dachary/wip-12847-hammer
904 common: do not insert emtpy ptr when rebuild emtpy bufferlist
906 Reviewed-by: Sage Weil <sage@redhat.com>
908 commit 5ef999e4a1dc04d7b7e58b1a61f85511099fd6d4
909 Merge: c243f7a cd11b88
910 Author: Loic Dachary <loic@dachary.org>
911 Date: Mon Sep 7 10:02:14 2015 +0200
913 Merge pull request #5373 from dachary/wip-12489-hammer
915 pg_interval_t::check_new_interval - for ec pool, should not rely on min_size to determine if the PG was active at the interval
917 Reviewed-by: Sage Weil <sage@redhat.com>
919 commit c243f7a33b0fff261bb92595a3cc7abf12b7e56d
920 Merge: e61fdc6 7bddf5d
921 Author: Loic Dachary <loic@dachary.org>
922 Date: Mon Sep 7 10:00:32 2015 +0200
924 Merge pull request #5383 from dachary/wip-12504-hammer
926 rest-bench common/WorkQueue.cc: 54: FAILED assert(_threads.empty())
928 Reviewed-by: Kefu Chai <kchai@redhat.com>
930 commit e61fdc67031b0358800335b638c96d22ecb85bd6
931 Merge: 62f543b 9f69660
932 Author: Loic Dachary <loic@dachary.org>
933 Date: Mon Sep 7 09:54:07 2015 +0200
935 Merge pull request #5765 from dachary/wip-12883-hammer
937 cache agent is idle although one object is left in the cache
939 Reviewed-by: Kefu Chai <kchai@redhat.com>
941 commit 62f543bbb092554746d3b89aab4b1ea75a5cdbf1
942 Merge: f25315c e471c5d
943 Author: Loic Dachary <loic@dachary.org>
944 Date: Mon Sep 7 09:53:14 2015 +0200
946 Merge pull request #5754 from dachary/wip-12588-hammer
948 Change radosgw pools default crush ruleset
950 Reviewed-by: Kefu Chai <kchai@redhat.com>
952 commit f25315c1e7e6e4700c2f39ba849dbcb25b0f1f89
953 Merge: 021abe7 43a72e4
954 Author: Loic Dachary <loic@dachary.org>
955 Date: Mon Sep 7 09:51:50 2015 +0200
957 Merge pull request #5377 from dachary/wip-12396-hammer
959 register_new_pgs() should check ruleno instead of its index
961 Reviewed-by: Kefu Chai <kchai@redhat.com>
963 commit 021abe742ccec876497468ac556b9dfaf6c9503d
964 Merge: 4125196 c94fd92
965 Author: Sage Weil <sage@redhat.com>
966 Date: Sun Sep 6 21:07:38 2015 -0400
968 Merge pull request #5758 from dachary/wip-12839-hammer
970 Mutex Assert from PipeConnection::try_get_pipe
972 commit 9f69660133f9534722589801a6fc827b74bead44
973 Author: Loic Dachary <ldachary@redhat.com>
974 Date: Thu Aug 13 19:41:47 2015 +0200
976 tests: tiering agent and proxy read
978 Verify that an object promoted to a cache tier because of a proxy read
979 is evicted as expected.
981 http://tracker.ceph.com/issues/12673 Refs: #12673
983 Signed-off-by: Loic Dachary <ldachary@redhat.com>
984 (cherry picked from commit 7924231930732bd297d3bd034c8295e96cb81088)
986 commit 5656eec0000e95ea790c796a9f572c7261163cfa
987 Author: Loic Dachary <ldachary@redhat.com>
988 Date: Thu Aug 13 13:47:24 2015 +0200
990 osd: trigger the cache agent after a promotion
992 When a proxy read happens, the object promotion is done in parallel. The
993 agent_choose_mode function must be called to reconsider the situation
994 to protect against the following scenario:
997 * agent_choose_mode finds no object exists and the agent
999 * object promotion happens
1000 * the agent does not reconsider and eviction does not happen
1003 http://tracker.ceph.com/issues/12673 Fixes: #12673
1005 Signed-off-by: Loic Dachary <ldachary@redhat.com>
1006 (cherry picked from commit e1f58feb9b1d20b72f2eb2eefdea5982e0cddccd)
1008 commit 4125196d31de8933656561b2f860e16a546ec4b9
1009 Merge: 47af509 f47ba4b
1010 Author: Loic Dachary <loic@dachary.org>
1011 Date: Sun Sep 6 23:17:22 2015 +0200
1013 Merge pull request #5276 from dachary/wip-11824-hammer
1015 implicit erasure code crush ruleset is not validated
1017 Reviewed-by: Loic Dachary <ldachary@redhat.com>
1019 commit 47af509003e551238702858ccd845b930e5ef1ed
1020 Merge: 2fd8ec8 1e05578
1021 Author: Loic Dachary <loic@dachary.org>
1022 Date: Sun Sep 6 17:24:43 2015 +0200
1024 Merge pull request #5382 from dachary/wip-12500-hammer
1026 segfault launching ceph-fuse with bad --name
1028 Reviewed-by: Sage Weil <sage@redhat.com>
1030 commit 2fd8ec8385a90daaa1bdb9de189ec73c7ba49c6c
1031 Merge: e6bdae9 fa19474
1032 Author: Loic Dachary <loic@dachary.org>
1033 Date: Sun Sep 6 17:23:19 2015 +0200
1035 Merge pull request #5367 from dachary/wip-12311-hammer
1037 read on chunk-aligned xattr not handled
1039 Reviewed-by: Sage Weil <sage@redhat.com>
1041 commit dc693fcaf2c596087b434f9e90fbc553394b9a14
1042 Author: Jason Dillaman <dillaman@redhat.com>
1043 Date: Wed Jul 29 12:46:24 2015 -0400
1045 lockdep: allow lockdep to be dynamically enabled/disabled
1047 librbd test cases attempt to enable lockdep coverage via the librados
1048 API. Use a configuration observer to register/unregister lockdep
1051 Signed-off-by: Jason Dillaman <dillaman@redhat.com>
1052 (cherry picked from commit af0cade0293be22e74a3293765676c7b227b06c4)
1054 commit 805732b143fd28603238969ba784adb57eff1a12
1055 Author: Jason Dillaman <dillaman@redhat.com>
1056 Date: Tue Jul 28 14:23:14 2015 -0400
1058 tests: librbd API test cannot use private md_config_t struct
1060 Remove all depencencies on md_config_t and instead use librados API
1061 methods to get/set configuration values.
1065 Signed-off-by: Jason Dillaman <dillaman@redhat.com>
1066 (cherry picked from commit 2c51aadb447d3447b5dace19c5cb6268e94247c8)
1068 commit 7ac01731bcb0f1d46cf98ecf623c7e8e6cb1a38e
1069 Author: Jason Dillaman <dillaman@redhat.com>
1070 Date: Thu Apr 9 15:06:27 2015 -0400
1072 tests: ensure old-format RBD tests still work
1074 Override the RBD default image format back to version 1
1075 to ensure tests properly cover the old format.
1077 Signed-off-by: Jason Dillaman <dillaman@redhat.com>
1078 (cherry picked from commit 3b95edb5bd373bd05f83516eccfd30a6b5cf4693)
1080 commit b68d7574d12944589b246281474bcad01f5347be
1081 Author: Jason Dillaman <dillaman@redhat.com>
1082 Date: Thu Jul 30 09:00:57 2015 -0400
1084 librados_test_stub: implement conf get/set API methods
1086 Signed-off-by: Jason Dillaman <dillaman@redhat.com>
1087 (cherry picked from commit 4d03c664f2a9362009c7e3077ed65e2989f0f21b)
1089 commit f0fa637e4a91a93a326ba303bc22101fadcc787d
1090 Author: Jason Dillaman <dillaman@redhat.com>
1091 Date: Tue Jul 28 13:14:29 2015 -0400
1093 crypto: use NSS_InitContext/NSS_ShutdownContex to avoid memory leak
1095 Switched to context-aware NSS init/shutdown functions to avoid conflicts
1096 with parent application. Use a reference counter to properly shutdown the
1097 NSS crypto library when the last CephContext is destroyed. This avoids
1098 memory leaks with the NSS library from users of librados.
1100 Signed-off-by: Jason Dillaman <dillaman@redhat.com>
1101 (cherry picked from commit 9fa0112dd9286178da1d76020158503b2062d252)
1103 commit 3f542aa2410139bece0f4192079e4e34260da782
1104 Author: Yan, Zheng <zyan@redhat.com>
1105 Date: Sat Mar 21 07:13:51 2015 +0800
1107 auth: use crypto_init_mutex to protect NSS_Shutdown()
1109 Signed-off-by: Yan, Zheng <zyan@redhat.com>
1110 (cherry picked from commit cf058174fe6be26129eb52f4fb24d0f7146ef7af)
1112 commit e487e8e3d84c7845ce7824ace3e375c6c389ba20
1113 Author: Yan, Zheng <zyan@redhat.com>
1114 Date: Sat Mar 21 01:02:42 2015 +0800
1116 auth: reinitialize NSS modules after fork()
1119 Signed-off-by: Yan, Zheng <zyan@redhat.com>
1120 (cherry picked from commit 4c24d0cc074462ae258b5cf901cd884bb4f50a53)
1122 commit 00e73adb03ef4453599268779fe354e4e8513e54
1123 Author: Jason Dillaman <dillaman@redhat.com>
1124 Date: Tue Aug 11 09:26:33 2015 -0400
1126 librbd: prevent race condition between resize requests
1128 It was possible that the same resize request could be sent twice
1129 if a completed resize op started a newly created resize op while
1130 it was also being concurrently started by another thread.
1134 Signed-off-by: Jason Dillaman <dillaman@redhat.com>
1135 (cherry picked from commit 6ec431bac1cf9446f3af90209d193dfcf003f2c5)
1137 commit 6c4ccc854fa8a8403b03785b06cb35a7174f4f42
1138 Author: Jianpeng Ma <jianpeng.ma@intel.com>
1139 Date: Mon Mar 9 14:23:23 2015 +0800
1141 librbd: Add a paramter:purge_on_error in ImageCtx::invalidate_cache().
1143 If bh_write met error, it will try again. For closing image, if met this
1144 issue, it will trigger a assert:
1145 >>2015-02-03 15:22:49.198292 7ff62d537800 -1 osdc/ObjectCacher.cc: In
1146 function 'ObjectCacher::~ObjectCacher()' thread 7ff62d537800 time
1147 >>2015-02-03 15:22:49.195927osdc/ObjectCacher.cc: 551: FAILED
1148 >>assert(i->empty())
1150 Now add purge_on_error, when shutdown_cache it set true.
1151 In ImageCtx::invalidate_cache, if met error and purge_on_error is true,
1154 Signed-off-by: Jianpeng Ma <jianpeng.ma@intel.com>
1155 (cherry picked from commit 35def5c81f7fc83d55d18320e4860c6a63d4c7f5)
1158 src/librbd/ImageCtx.cc : trivial resolution
1160 commit 05734916ab119c6d3879c2ce2bc9f9581907861a
1161 Author: Jianpeng Ma <jianpeng.ma@intel.com>
1162 Date: Thu Feb 5 11:28:50 2015 +0800
1164 librbd: Remvoe unused func ImageCtx::read_from_cache.
1166 Signed-off-by: Jianpeng Ma <jianpeng.ma@intel.com>
1167 (cherry picked from commit 101440a41253680770f94bc380af7072c7adaebf)
1169 commit 28838f20a722b72a6e926e53f104342d3b9f4791
1170 Author: Jianpeng Ma <jianpeng.ma@intel.com>
1171 Date: Thu Feb 5 11:03:10 2015 +0800
1173 osdc: clean up code in ObjectCacher::Object::map_write
1175 Signed-off-by: Jianpeng Ma <jianpeng.ma@intel.com>
1176 (cherry picked from commit 9f80c2909ace09cd51c24b49c98a093e0e864dca)
1178 commit 5c4f152efa8e8f57c59ea7decc05ae1a34f2a9ee
1179 Author: Jianpeng Ma <jianpeng.ma@intel.com>
1180 Date: Thu Feb 5 10:13:06 2015 +0800
1182 osdc: Don't pass mutex into ObjectCacher::_wait_for_write.
1184 Because the mutex is the same as ObjectCacher::lock.
1186 Signed-off-by: Jianpeng Ma <jianpeng.ma@intel.com>
1187 (cherry picked from commit d7cf7aeea5cba1ffa8e51ff1ad2424b1ec161a12)
1189 commit 86e7698a19745c26b9d3e7a12a16c87ea9a5d565
1190 Author: Jianpeng Ma <jianpeng.ma@intel.com>
1191 Date: Mon Mar 2 11:23:44 2015 +0800
1193 osdc: After write try merge bh.
1195 Signed-off-by: Jianpeng Ma <jianpeng.ma@intel.com>
1196 (cherry picked from commit 1a48a8a2b222e41236341cb1241f0885a1b0b9d8)
1198 commit c96541ad19fb142ed31ff3006f1dc25e0c1de86c
1199 Author: Jianpeng Ma <jianpeng.ma@intel.com>
1200 Date: Mon Mar 2 11:36:24 2015 +0800
1202 osdc: Make last missing bh to wake up the reader.
1204 Avoid wakeup early and wait again.
1206 Signed-off-by: Jianpeng Ma <jianpeng.ma@intel.com>
1207 (cherry picked from commit d582bda090b3339d03e25b0e6d0971ad0267f476)
1209 commit 4135b9a2d199583685a1bae3713347dcc1b872e6
1210 Author: Jianpeng Ma <jianpeng.ma@intel.com>
1211 Date: Fri Feb 6 09:53:36 2015 +0800
1213 osdc: For trust_enoent is true, there is only one extent.
1215 Now the judgement only in conditon which will return -ENOENT.
1216 But o->exists don't depend on the extent size. It only depend on
1217 trust_enoent. So move this judgement at the first of _readx().
1218 Make this bug ASAP occur.
1220 Signed-off-by: Jianpeng Ma <jianpeng.ma@intel.com>
1221 (cherry picked from commit 2449fddc13b5ce8bbf1bb4ecaa5d6937f54e54d1)
1223 commit 81376b6a1a5db9f8b45c58ea318ec924e932d990
1224 Author: Jianpeng Ma <jianpeng.ma@intel.com>
1225 Date: Tue Feb 3 14:46:39 2015 +0800
1227 osdc: In _readx() only no error can tidy read result.
1229 Signed-off-by: Jianpeng Ma <jianpeng.ma@intel.com>
1230 (cherry picked from commit 540346d4a901d8041c9fd74641c98cdfd2e1ab32)
1232 commit e6bdae903cdd2f702829adc31fa9a31c239a3b87
1233 Merge: 317770d cdde626
1234 Author: Ken Dreyer <kdreyer@redhat.com>
1235 Date: Fri Sep 4 15:38:43 2015 -0600
1237 Merge pull request #5223 from SUSE/wip-12305-hammer
1239 ceph.spec.in: running fdupes unnecessarily
1241 Reviewed-by: Ken Dreyer <kdreyer@redhat.com>
1243 commit e80bd0a2a39afdabcb134100e0cc8fa74cbd14c2
1244 Author: Radoslaw Zarzynski <rzarzynski@mirantis.com>
1245 Date: Thu Jun 25 18:28:50 2015 +0200
1247 rgw: send Content-Length in response for GET on Swift account.
1250 Signed-off-by: Radoslaw Zarzynski <rzarzynski@mirantis.com>
1251 (cherry picked from commit 5f47b11791cd46ce2771046e4888575869cc9685)
1253 src/rgw/rgw_rest_swift.cc
1255 commit 2e542456dee6c30b3200e9830d4ccae4b0325388
1256 Author: Orit Wasserman <owasserm@redhat.com>
1257 Date: Mon Jun 22 13:51:49 2015 +0200
1259 rgw: force content_type for swift bucket stats request
1263 Signed-off-by: Orit Wasserman <owasserm@redhat.com>
1264 (cherry picked from commit da259e4da1329e1e448d75c48c6ee3d1318e19b2)
1266 commit 5d57b63c32e43000ff95c2d8ce34347ed0579a8a
1267 Author: Orit Wasserman <owasserm@redhat.com>
1268 Date: Thu Aug 13 15:02:03 2015 +0200
1270 rgw: we should not overide Swift sent content type
1275 Signed-off-by: Orit Wasserman <owasserm@redhat.com>
1276 (cherry picked from commit 423cf136f15df3099c9266f55932542d303c2713)
1278 commit b8aafbc6f9f80f029606a577d73f8685a80225b8
1279 Author: Radoslaw Zarzynski <rzarzynski@mirantis.com>
1280 Date: Tue Jun 23 11:54:09 2015 +0200
1282 rgw: enforce Content-Type in Swift responses.
1284 Swift sends Content-Type HTTP header even if the response
1285 doesn't contain body. We have this behaviour implemented
1286 until applying some changes in end_header() function.
1287 Unfortunately, lack of Content-Type causes early exits in
1288 many Tempest's tests for Swift API verification.
1291 Signed-off-by: Radoslaw Zarzynski <rzarzynski@mirantis.com>
1292 (cherry picked from commit 106aeba206736d4080326f9bc191876bed63370b)
1294 commit 143cfc331efcc7bed8b1cf90d7e3188d0ac7604b
1295 Author: Orit Wasserman <owasserm@redhat.com>
1296 Date: Tue Apr 21 12:31:19 2015 +0200
1298 rgw: force content-type header for swift account responses without body
1300 This regression was introduced in commit 4cfeca03ade21861ab70ca759f94eb244c16cb39
1304 Signed-off-by: Orit Wasserman <owasserm@redhat.com>
1305 (cherry picked from commit 009664e560ee37ef6dc170a0155d6c749e151b06)
1307 commit b5420d6f1a6a1da34c7bba339e6763c4b369d349
1308 Author: Orit Wasserman <owasserm@redhat.com>
1309 Date: Thu Apr 9 20:04:59 2015 +0200
1311 rgw: shouldn't return content-type: application/xml if content length is 0
1315 Signed-off-by: Orit Wasserman <owasserm@redhat.com>
1316 (cherry picked from commit 5cc92bb27cfbaee4af24dc312c9b98331f899794)
1318 commit 317770dd39510221cbc0fc7ee50960992137eaaf
1319 Merge: 6b02be7 6b36514
1320 Author: Loic Dachary <loic@dachary.org>
1321 Date: Thu Sep 3 12:20:38 2015 +0200
1323 Merge pull request #5716 from dachary/wip-12851-hammer
1325 Ensure that swift keys don't include backslashes
1327 Reviewed-by: Orit Wasserman <owasserm@redhat.com>
1329 commit 6b02be7395ad8f41f70458e7bd11630ddd498cc5
1330 Merge: fa74ea1 f1c7c62
1331 Author: Loic Dachary <loic@dachary.org>
1332 Date: Thu Sep 3 12:11:24 2015 +0200
1334 Merge pull request #5717 from dachary/wip-12591-hammer
1336 rgw: create a tool for orphaned objects cleanup
1338 Reviewed-by: Orit Wasserman <owasserm@redhat.com>
1340 commit fa74ea10ac25fe93f359ccf0fd036d567e9610fe
1341 Merge: 011e4c5 a478385
1342 Author: Loic Dachary <loic@dachary.org>
1343 Date: Wed Sep 2 23:35:58 2015 +0200
1345 Merge pull request #5755 from dachary/wip-12589-hammer
1347 ceph-disk zap should ensure block device
1349 Reviewed-by: Loic Dachary <ldachary@redhat.com>
1351 commit 011e4c58ffb41aa5ae3240a40cbeba8266b89001
1352 Merge: e7c2269 b80859e
1353 Author: Loic Dachary <loic@dachary.org>
1354 Date: Wed Sep 2 23:22:59 2015 +0200
1356 Merge pull request #5732 from ceph/wip-11455-hammer
1358 rgw: init some manifest fields when handling explicit objs
1360 Reviewed-by: Loic Dachary <ldachary@redhat.com>
1362 commit e7c2269dbebd5337ae5aad127d6ddaa4260a8f53
1363 Merge: 92b641f da00bed
1364 Author: Loic Dachary <loic@dachary.org>
1365 Date: Wed Sep 2 23:19:02 2015 +0200
1367 Merge pull request #5721 from dachary/wip-12853-hammer
1369 RGW Swift API: X-Trans-Id header is wrongly formatted
1371 Reviewed-by: Radoslaw Zarzynski <rzarzynski@mirantis.com>
1373 commit 92b641f301df33a38ccfcf018c5f7657db8548ae
1374 Merge: 845d2ae 3ab5d82
1375 Author: Loic Dachary <loic@dachary.org>
1376 Date: Wed Sep 2 23:08:24 2015 +0200
1378 Merge pull request #5498 from ceph/wip-12432-hammer
1380 rgw: HTTP return code is not being logged by CivetWeb
1382 Reviewed-by: Loic Dachary <ldachary@redhat.com>
1384 commit 836f7636f76b2778812e4fa055206a887fd27dad
1385 Author: Samuel Just <sjust@redhat.com>
1386 Date: Tue Jul 21 11:31:12 2015 -0700
1388 OSD: break connection->session->waiting message->connection cycle
1391 Signed-off-by: Samuel Just <sjust@redhat.com>
1392 (cherry picked from commit 7f15e22be291df139c85909b96b2366f6dbadf37)
1394 commit 77624aff03a8df7805ac0c1c6a7a2085f41fb6a3
1395 Author: Sage Weil <sage@redhat.com>
1396 Date: Sun Aug 9 10:46:10 2015 -0400
1398 osd/PGLog: dirty_to is inclusive
1400 There are only two callers of mark_dirty_to who do not pass max,
1401 and they are both in the merge_log extending tail path. In that
1402 case, we want to include the last version specified in the log
1403 writeout. Fix the tail extending code to always specify the
1404 last entry added, inclusive.
1407 Signed-off-by: Sage Weil <sage@redhat.com>
1408 (cherry picked from commit f0ca14df0641daa04eee39d98d8bd0faf46e4e6d)
1410 commit aa003736d2aeea937ce0d8663b159cf06b2dda3e
1411 Author: Xinze Chi <xmdxcxz@gmail.com>
1412 Date: Mon Aug 24 23:02:10 2015 +0800
1414 common: fix code format
1416 Signed-off-by: Xinze Chi <xmdxcxz@gmail.com>
1417 (cherry picked from commit 15e5ebe14787bbbc6b906d72ccd2d418d762cfad)
1419 commit aab35da90d7d50750fda6355d5a124a45d66f1c5
1420 Author: Xinze Chi <xmdxcxz@gmail.com>
1421 Date: Mon Aug 24 22:59:40 2015 +0800
1423 test: add test case for insert empty ptr when buffer rebuild
1425 Signed-off-by: Xinze Chi <xmdxcxz@gmail.com>
1426 (cherry picked from commit 2d2f0eb338906742d516a2f8e6d5d05505be8f47)
1428 commit 2b0b7ae6443eeca271813415a91ab749c886fdf0
1429 Author: Xinze Chi <xmdxcxz@gmail.com>
1430 Date: Mon Aug 24 23:01:27 2015 +0800
1432 common: fix insert empty ptr when bufferlist rebuild
1435 Signed-off-by: Xinze Chi <xmdxcxz@gmail.com>
1436 (cherry picked from commit fb1b6ddd7bfe40d4a38d5ec59b1c3f59f2a83f23)
1439 src/common/buffer.cc : because invalidate_crc did not exist
1442 commit 845d2aed9e9a9893678226b8661fbd8ff87c090f
1443 Merge: a6517d2 3396a96
1444 Author: Loic Dachary <loic@dachary.org>
1445 Date: Wed Sep 2 12:00:11 2015 +0200
1447 Merge pull request #5527 from SUSE/wip-12585-hammer
1449 OSD crash creating/deleting pools
1451 Reviewed-by: Kefu Chai <kchai@redhat.com>
1453 commit 2348a5b5c96d065fbc2188bd9e8e7ff20c75ecaa
1454 Author: Guang G Yang <yguang@renownedground.corp.gq1.yahoo.com>
1455 Date: Wed Jul 29 21:47:17 2015 +0000
1457 osd: copy the RecoveryCtx::handle when creating a new RecoveryCtx instance from another one
1461 Signed-off-by: Guang Yang <yguang@yahoo-inc.com>
1462 (cherry picked from commit 6ab9efe799b947f2ef6113df8cc542c73760107d)
1464 commit bf72785e1e11b18df9a03ebfa7c284fe4f43eaa0
1465 Author: Josh Durgin <jdurgin@redhat.com>
1466 Date: Mon Aug 24 15:40:39 2015 -0700
1468 config: skip lockdep for intentionally recursive md_config_t lock
1470 lockdep can't handle recursive locks, resulting in false positive
1471 reports for certain set_val_or_die() calls, like via
1472 md_config_t::parse_argv() passed "-m".
1475 Signed-off-by: Josh Durgin <jdurgin@redhat.com>
1476 (cherry picked from commit 415226995ebe841cc9e8639a3233150a75aecaed)
1478 commit c94fd926239a31dea1389a760f55f332a9b6814a
1479 Author: David Zafman <dzafman@redhat.com>
1480 Date: Thu Jul 23 16:36:19 2015 -0700
1482 osd: Keep a reference count on Connection while calling send_message()
1486 Signed-off-by: David Zafman <dzafman@redhat.com>
1487 (cherry picked from commit a140085f467889f2743294a3c150f13b62fcdf51)
1489 commit 059bf98d40fa7e1eddb7aef2ecfdfb6bc96c14fb
1490 Author: Samuel Just <sjust@redhat.com>
1491 Date: Tue Jul 7 08:49:54 2015 -0700
1493 WBThrottle::clear_object: signal if we cleared an object
1496 Signed-off-by: Samuel Just <sjust@redhat.com>
1497 (cherry picked from commit 8447b08d60d5cf7dce20f8b839cc2cb5a93782de)
1499 commit a478385334323071b5cfa4d546eef5b0ef43dce7
1500 Author: Loic Dachary <ldachary@redhat.com>
1501 Date: Sun May 31 19:42:45 2015 +0200
1503 ceph-disk: always check zap is applied on a full device
1505 zap only makes sense on a full device and this is verified for
1507 ceph-disk prepare --zap-disk
1513 Unify the two and add a test to check that it fails as expected when
1514 trying to zap a directory.
1516 http://tracker.ceph.com/issues/11272 Fixes: #11272
1518 Signed-off-by: Loic Dachary <ldachary@redhat.com>
1519 (cherry picked from commit 8e1bcf21f090fc98ac9b495d76614e3012f5872a)
1522 src/test/ceph-disk.sh : trivial resolution
1524 commit e471c5dc16847ba35d3c3c8c74da1bf7c4c946e0
1525 Author: Yuan Zhou <yuan.zhou@intel.com>
1526 Date: Wed Jul 8 10:35:49 2015 +0800
1528 librados: Make librados pool_create respect default_crush_ruleset
1530 Make pool_create in librados use the osd_pool_default_crush_replicated_ruleset
1534 Signed-off-by: Yuan Zhou <yuan.zhou@intel.com>
1535 (cherry picked from commit da96a89033590277460aef1c80f385bd93d625e1)
1537 commit a6517d286b59eac283d7bfd04b620f5b859657b6
1538 Merge: a89c173 35fa47a
1539 Author: Loic Dachary <loic@dachary.org>
1540 Date: Wed Sep 2 02:19:26 2015 +0200
1542 Merge pull request #5551 from ceph/wip-corpus-hammer
1544 ceph-object-corpus: add 0.94.2-207-g88e7ee7 hammer objects
1546 Reviewed-by: Loic Dachary <ldachary@redhat.com>
1548 commit 35fa47a117a3e8d185f7ebe02eea5671f8229427 (refs/remotes/me/wip-corpus-hammer, refs/remotes/gh/wip-corpus-hammer)
1549 Author: Sage Weil <sage@redhat.com>
1550 Date: Tue Sep 1 17:44:06 2015 -0400
1552 ceph-object-corpus: add 0.94.2-207-g88e7ee7 hammer objects
1554 Signed-off-by: Sage Weil <sage@redhat.com>
1556 commit b80859e2b2efc2402dc62aa319e40319863586f7 (refs/remotes/gh/wip-11455-hammer)
1557 Author: Yehuda Sadeh <yehuda@redhat.com>
1558 Date: Wed Aug 26 14:34:30 2015 -0700
1560 rgw: init some manifest fields when handling explicit objs
1563 When dealing with old manifest that has explicit objs, we also
1564 need to set the head size and head object correctly so that
1565 code that relies on this info doesn't break.
1567 Signed-off-by: Yehuda Sadeh <yehuda@redhat.com>
1568 (cherry picked from commit d7bf8cb594e5276d1c80544f5ec954d52b159750)
1570 commit f47ba4b1a1029a55f8bc4ab393a7fa3712cd4e00
1571 Author: Loic Dachary <ldachary@redhat.com>
1572 Date: Thu Jul 16 18:02:02 2015 +0200
1574 mon: test the crush ruleset when creating a pool
1576 We want to fix the following scenario:
1578 * an erasure code plugin (or another part of the code) creates a
1580 * the ruleset crashes during mapping (for whatever reason)
1581 * ceph osd pool create uses the bugous ruleset
1582 * the monitors try to do mapping a crash
1584 Having a bugous ruleset in the crush map is very difficult prevent. The
1585 catastrophic event of using it with a newly created pool can however be
1586 prevented by calling the CrushTester just before creating the pool and
1587 after all implicit or explicit crush ruleset creation happened.
1589 http://tracker.ceph.com/issues/11814 Fixes: #11814
1591 Signed-off-by: Loic Dachary <ldachary@redhat.com>
1592 (cherry picked from commit f1e86be589803596e86acc964ac5c5c03b4038d8)
1595 src/test/mon/osd-crush.sh
1596 removed the run_mon because in hammer it is shared
1598 src/mon/OSDMonitor.cc
1599 prepare_new_pool changed stringstream to *ostream
1601 commit b58cbbab4f74e352c3d4a61190cea2731057b3c9
1602 Author: Loic Dachary <ldachary@redhat.com>
1603 Date: Sat May 30 12:40:26 2015 +0200
1605 erasure-code: set max_size to chunk_count() instead of 20 for shec
1607 See 8b64fe93b088a3a33d357869c47e6bf928c3f0e4 for a detailed explanation.
1609 Signed-off-by: Loic Dachary <ldachary@redhat.com>
1610 (cherry picked from commit a44499f7e6609d9a32254dc6f4303122de74db31)
1612 commit 6f0af185ad7cf9640557efb7f61a7ea521871b5b
1613 Author: Mykola Golub <mgolub@mirantis.com>
1614 Date: Thu Feb 26 21:22:31 2015 +0200
1616 vstart.sh: set PATH to include pwd
1618 ceph-mon needs crushtool to be in PATH. Don't set if it is run
1619 from ceph_vstart_wrapper, which already sets it as it needs.
1621 Signed-off-by: Mykola Golub <mgolub@mirantis.com>
1622 (cherry picked from commit e08bccf377292d72463796ae1dff5d8be40d88e9)
1624 commit a89c1731e0e327acdb876d39df574e27379aaa1e
1625 Merge: c3a2bf3 0fde3a2
1626 Author: Gregory Farnum <greg@gregs42.com>
1627 Date: Mon Aug 31 09:41:33 2015 +0100
1629 Merge pull request #5365 from dachary/wip-12491-hammer
1631 buffer: critical bufferlist::zero bug
1633 Reviewed-by: Greg Farnum <gfarnum@redhat.com>
1635 commit c3a2bf3f6b1d1bf416fa04df1629630abe2fe563
1636 Merge: 0a44eb1 4457d3e
1637 Author: Loic Dachary <loic@dachary.org>
1638 Date: Sun Aug 30 19:58:25 2015 +0200
1640 Merge pull request #5369 from dachary/wip-12498-hammer
1642 get pools health'info have error
1644 Reviewed-by: Kefu Chai <kchai@redhat.com>
1646 commit 0a44eb12fc7d2ef902f1999e37abae0f11511d66
1647 Merge: dfcdba7 9916d37
1648 Author: Loic Dachary <loic@dachary.org>
1649 Date: Sun Aug 30 19:52:52 2015 +0200
1651 Merge pull request #5370 from dachary/wip-12496-hammer
1653 pgmonitor: wrong at/near target max“ reporting
1655 Reviewed-by: Kefu Chai <kchai@redhat.com>
1657 commit dfcdba75f310eaa2cfc5ae97ba3093afe1198737
1658 Merge: 69a9a4c b8176d0
1659 Author: Loic Dachary <loic@dachary.org>
1660 Date: Sun Aug 30 19:44:56 2015 +0200
1662 Merge pull request #5378 from dachary/wip-12394-hammer
1664 Memory leak in Mutex.cc, pthread_mutexattr_init without pthread_mutexattr_destroy
1666 Reviewed-by: Kefu Chai <kchai@redhat.com>
1668 commit 69a9a4cbf28c87fcd68f6c6c0c09f7a6d95ab003
1669 Merge: d3b160e 42bff0b
1670 Author: Loic Dachary <loic@dachary.org>
1671 Date: Sun Aug 30 19:44:26 2015 +0200
1673 Merge pull request #5372 from dachary/wip-12493-hammer
1675 the output is wrong when runing ceph osd reweight
1677 Reviewed-by: Kefu Chai <kchai@redhat.com>
1679 commit da00bedbc2f5d2facb9472078fd6cdd7260492ac
1680 Author: Radoslaw Zarzynski <rzarzynski@mirantis.com>
1681 Date: Thu Aug 6 15:52:58 2015 +0200
1683 rgw: rework X-Trans-Id header to be conform with Swift API.
1686 Signed-off-by: Radoslaw Zarzynski <rzarzynski@mirantis.com>
1687 (cherry picked from commit d1735a4e8ea2056d333a5001a7a410fc03fe1b9e)
1689 commit 9937c818f1073bceffcf5811c7cc543b9043e646
1690 Author: Abhishek Dixit <dixitabhi@gmail.com>
1691 Date: Mon Jun 8 22:59:54 2015 +0530
1693 Transaction Id added in response
1695 Signed-off-by: Abhishek Dixit dixitabhi@gmail.com
1697 (cherry picked from commit b711e3124f8f73c17ebd19b38807a1b77f201e44)
1701 the prototype of the end_header( function
1702 in the context of the diff changed
1704 commit f1c7c629f2a7baf87a3d25d68b7a40fb3b2f6c49
1705 Author: Yehuda Sadeh <yehuda@redhat.com>
1706 Date: Mon Jun 29 15:35:04 2015 -0700
1708 rgw: api adjustment following a rebase
1710 Signed-off-by: Yehuda Sadeh <yehuda@redhat.com>
1711 (cherry picked from commit 7cf1f376e7dbc98a44f4a8e78bfa11f84291a941)
1713 commit 85911df5226dccfd2f1806a204a23eae3966eee7
1714 Author: Yehuda Sadeh <yehuda@redhat.com>
1715 Date: Mon Jun 29 15:34:44 2015 -0700
1717 rgw: orphans, fix check on number of shards
1719 Signed-off-by: Yehuda Sadeh <yehuda@redhat.com>
1720 (cherry picked from commit 47edec359be2a4d230a4a54c0c670d0aa711fa1e)
1722 commit c1cf7dfb672032012889840fb4462b1a076f34c0
1723 Author: Yehuda Sadeh <yehuda@redhat.com>
1724 Date: Mon Jun 29 15:34:11 2015 -0700
1726 rgw: orphans, change default number of shards
1728 Signed-off-by: Yehuda Sadeh <yehuda@redhat.com>
1729 (cherry picked from commit 5528f21d0861d42378a8eb10759a1e55ddd66c09)
1731 commit bb1d4cc3a3e9af3bf8cd8fd80085fbbbaeb8b89f
1732 Author: Yehuda Sadeh <yehuda@redhat.com>
1733 Date: Tue May 5 14:43:05 2015 -0700
1735 rgw: change error output related to orphans
1737 Signed-off-by: Yehuda Sadeh <yehuda@redhat.com>
1738 (cherry picked from commit cac57caf639236adc9a641a85352eb729b58725a)
1740 commit 2e0f6fe98787e1f3ea7d107a814caadce7ee5f23
1741 Author: Yehuda Sadeh <yehuda@redhat.com>
1742 Date: Mon May 4 17:02:29 2015 -0700
1744 rgw: orphan, fix truncated detection
1746 Signed-off-by: Yehuda Sadeh <yehuda@redhat.com>
1747 (cherry picked from commit d8ce04db8542dae0027423f3c50355e5de74dc9a)
1749 commit 1bfebefba2e09448d28853b34415d3f65e323c76
1750 Author: Yehuda Sadeh <yehuda@redhat.com>
1751 Date: Mon May 4 16:32:57 2015 -0700
1753 radosgw-admin: simplify orphan command
1755 No need for --init-search
1757 Signed-off-by: Yehuda Sadeh <yehuda@redhat.com>
1758 (cherry picked from commit 4980cbd04ff650d79a033bc5298e6c1abe985133)
1760 commit f244b159ac8793c11a5707ea0b104b3180efb58f
1761 Author: Yehuda Sadeh <yehuda@redhat.com>
1762 Date: Mon May 4 15:24:00 2015 -0700
1764 radosgw-admin: stat orphan objects before reporting leakage
1766 We don't want to report new objects as leaked, because they might just
1767 be written, and waiting to be linked to a new object. The number of
1768 seconds prior to the job initialization can be cofigured through
1769 --orphan-stale-secs (by default it's 24h).
1771 Signed-off-by: Yehuda Sadeh <yehuda@redhat.com>
1772 (cherry picked from commit 66b0090023859d97da2bbaca698b6a50c225abd0)
1774 commit f80e2b2bc8879d1cc40338816196f51378aaa785
1775 Author: Yehuda Sadeh <yehuda@redhat.com>
1776 Date: Mon May 4 14:39:20 2015 -0700
1778 radosgw-admin: orphans finish command
1780 A command to remove orphans data
1782 Signed-off-by: Yehuda Sadeh <yehuda@redhat.com>
1783 (cherry picked from commit 55d6f5ecf18f532c7f056f8b60c101843594b00c)
1785 commit 88d32c641a06823df93f0e3fae640a880778e1e4
1786 Author: Yehuda Sadeh <yehuda@redhat.com>
1787 Date: Sat May 2 17:28:30 2015 -0700
1789 rgw: cannot re-init an orphan scan job
1791 Signed-off-by: Yehuda Sadeh <yehuda@redhat.com>
1792 (cherry picked from commit f8993102731518153bb18eb74c3e6e4943a5dbac)
1794 commit 80a40342b6a0d9e7a47ca4b672901315246385df
1795 Author: Yehuda Sadeh <yehuda@redhat.com>
1796 Date: Sat May 2 16:38:08 2015 -0700
1798 rgw: stat_async() sets the object locator appropriately
1800 Signed-off-by: Yehuda Sadeh <yehuda@redhat.com>
1801 (cherry picked from commit d7d117121858155acc8548908e9459a48c676dd1)
1803 commit 0082036348e0463e42520504466961eb2bd77602
1804 Author: Yehuda Sadeh <yehuda@redhat.com>
1805 Date: Sat May 2 16:34:09 2015 -0700
1807 rgw: list_objects() sets namespace appropriately
1809 list_objects() now uses parse_raw_oid(), so that it can set the correct
1810 namespace. It only affects users of the function that want to get all
1811 objects in bucket, regardless to the namespace associated with it. This
1812 makes it so that the orphan code actually works now with namespaced
1813 objects, and with special named objects (namely, start with underscore).
1815 Signed-off-by: Yehuda Sadeh <yehuda@redhat.com>
1816 (cherry picked from commit 20bd490bebf9be3553d4e25322c9544b0b090086)
1818 commit 1c3707291e83771ca1d5cd24ce445f723006b744
1819 Author: Yehuda Sadeh <yehuda@redhat.com>
1820 Date: Fri May 1 17:23:44 2015 -0700
1822 rgw: modify orphan search fingerprints
1824 Now works with multipart uploads.
1826 Signed-off-by: Yehuda Sadeh <yehuda@redhat.com>
1827 (cherry picked from commit fa61ea2ccc94f04afb39868ac34abc264806d8d4)
1829 commit ef81367a1eaeb7778dc60a668b9fe2f6a77e06ff
1830 Author: Yehuda Sadeh <yehuda@redhat.com>
1831 Date: Fri May 1 15:17:10 2015 -0700
1833 rgw: compare oids and dump leaked objects
1835 Signed-off-by: Yehuda Sadeh <yehuda@redhat.com>
1836 (cherry picked from commit 1bc63d98ff8a3aa180065153690b4f8a73658b79)
1838 commit f4d0544e63f68fa94b21fda71c0b78edd05038a1
1839 Author: Yehuda Sadeh <yehuda@redhat.com>
1840 Date: Thu Apr 30 16:17:54 2015 -0700
1842 rgw: keep accurate state for linked objects orphan scan
1844 Signed-off-by: Yehuda Sadeh <yehuda@redhat.com>
1845 (cherry picked from commit f19b2f087cfb5a89edf3b19487c0df3a214e350f)
1847 commit 748ea57cfeca7b295afca5b8126d454eb85550c7
1848 Author: Yehuda Sadeh <yehuda@redhat.com>
1849 Date: Wed Apr 29 17:12:34 2015 -0700
1851 rgw: iterate over linked objects, store them
1853 only keep part of the oid name if it is in a namespace.
1855 Signed-off-by: Yehuda Sadeh <yehuda@redhat.com>
1856 (cherry picked from commit 75902fdb12ce0413d3beb455df47134f3d7386e7)
1858 commit 6c6aa5ddf38102012e61f79f5234b9647d42b461
1859 Author: Yehuda Sadeh <yehuda@redhat.com>
1860 Date: Wed Apr 29 17:12:00 2015 -0700
1862 rgw: add rgw_obj::parse_raw_oid()
1864 Signed-off-by: Yehuda Sadeh <yehuda@redhat.com>
1865 (cherry picked from commit 7c1aa8373a090d5c1b2aa121582547723120a079)
1867 commit 62d562d76e0456cc3c735c6708531c7deb2874da
1868 Author: Yehuda Sadeh <yehuda@redhat.com>
1869 Date: Wed Apr 29 14:50:15 2015 -0700
1871 rgw: iterate asynchronously over linked objects
1873 Read objects manifest. So that we could keep the relevant info later.
1875 Signed-off-by: Yehuda Sadeh <yehuda@redhat.com>
1876 (cherry picked from commit 05a953d965a0a6a0e9feddaa62f7429e154e828e)
1878 commit 00ecf2d0a895407bc8ec0e55e1b1b0f47a1aa96f
1879 Author: Yehuda Sadeh <yehuda@redhat.com>
1880 Date: Wed Apr 29 14:15:33 2015 -0700
1882 rgw: async object stat functionality
1884 An async functionality that stats object.
1886 Signed-off-by: Yehuda Sadeh <yehuda@redhat.com>
1887 (cherry picked from commit 13adf3cf4f5fd8a0feb7cc1d7a4ee4ded9f573db)
1889 commit 7d1cc48ba32e8bc732c43ecb539312c6bfcc0f80
1890 Author: Yehuda Sadeh <yehuda@redhat.com>
1891 Date: Tue Apr 28 16:45:49 2015 -0700
1893 rgw-admin: build index of bucket indexes
1895 for the orphan search tool
1897 Signed-off-by: Yehuda Sadeh <yehuda@redhat.com>
1898 (cherry picked from commit 16a2dbd838b69fb445c06122a68fc65ec3a7b3de)
1900 commit c1b0e7a985cc9e742dcef4c8c427c57a2f58fa43
1901 Author: Yehuda Sadeh <yehuda@redhat.com>
1902 Date: Sat Apr 25 09:37:53 2015 -0700
1904 rgw: initial work of orphan detection tool implementation
1906 So far doesn't do much, iterate through all objects in a specific pool
1907 data, store it in a sharded index.
1909 Signed-off-by: Yehuda Sadeh <yehuda@redhat.com>
1910 (cherry picked from commit 767fc29aa62a0a4e4dd8a78785576e8f1b44ee5b)
1912 commit b16129c3393d6e301b4af5203ffa1c6d8dfdda38
1913 Author: Pavan Rallabhandi <pavan.rallabhandi@sandisk.com>
1914 Date: Wed Apr 29 13:35:29 2015 +0530
1916 Avoid an extra read on the atomic variable
1918 Signed-off-by: Pavan Rallabhandi <pavan.rallabhandi@sandisk.com>
1919 (cherry picked from commit 7476e944af049aefdb8871cce62883b86f6b6f8d)
1921 commit 1f6916d5a9b0ebe5181b801c4c381a699883bb09
1922 Author: Pavan Rallabhandi <pavan.rallabhandi@sandisk.com>
1923 Date: Wed Apr 8 18:53:14 2015 +0530
1925 RGW: Make RADOS handles in RGW to be a configurable option
1927 Having a configurable number of RADOS handles is resulting
1928 in significant performance boost for all types of workloads.
1929 Each RGW worker thread would now get to pick a RADOS handle
1930 for its lifetime, from the available bunch.
1932 Signed-off-by: Pavan Rallabhandi <pavan.rallabhandi@sandisk.com>
1933 (cherry picked from commit b3c3a24fe9c0fbb2bdbbccb08b40efc1dca207cc)
1935 commit d3b160ecbe181c2b969a14825cf314a056630188
1936 Merge: 7357659 695f782
1937 Author: Loic Dachary <loic@dachary.org>
1938 Date: Sun Aug 30 15:50:03 2015 +0200
1940 Merge pull request #5374 from dachary/wip-12487-hammer
1942 ceph osd crush reweight-subtree does not reweight parent node
1944 Reviewed-by: Sage Weil <sage@redhat.com>
1946 commit 7357659b59d9bf3522bf664f1a3da7475a7c8a49
1947 Merge: 8f559d5 8a2ad05
1948 Author: Loic Dachary <loic@dachary.org>
1949 Date: Sun Aug 30 15:49:41 2015 +0200
1951 Merge pull request #5380 from dachary/wip-12390-hammer
1953 PGLog::proc_replica_log: correctly handle case where entries between olog.head and log.tail were split out
1955 Reviewed-by: Sage Weil <sage@redhat.com>
1957 commit a13c7fd7c39e83072f2975477e1cdcfe079fb041
1958 Author: Ruifeng Yang <149233652@qq.com>
1959 Date: Sat Aug 1 17:39:17 2015 +0800
1961 rgw:the arguments 'domain' should not be assigned when return false
1966 Output: [B.A] it is wrong.
1969 Signed-off-by: Ruifeng Yang <149233652@qq.com>
1970 (cherry picked from commit 9420d24c21b4e06d62f356637599c6ec8e7e3e4d)
1972 commit 6acf36f8851b8a369949f0fad9d0156af7dde20f
1973 Author: Ruifeng Yang <149233652@qq.com>
1974 Date: Sun Aug 2 16:16:05 2015 +0800
1976 rgw:segmentation fault when rgw_gc_max_objs > HASH_PRIME
1978 obj_names[] is index out of bounds when rgw_gc_max_objs > HASH_PRIME
1981 Signed-off-by: Ruifeng Yang <149233652@qq.com>
1982 (cherry picked from commit e3147b8ddcaafafde9ce2029f7af5062e22076d3)
1984 commit 6b365144862cbd387d8b931939a3dc2cfe17bbbd
1985 Author: Yehuda Sadeh <yehuda@redhat.com>
1986 Date: Mon Jul 20 20:27:33 2015 -0700
1988 rgw: avoid using slashes for generated secret keys
1990 Just use plain alphanumeric characterset.
1994 Signed-off-by: Yehuda Sadeh <yehuda@redhat.com>
1995 (cherry picked from commit 3f04a6126fdbfb93304f798da3775c0eec9b7d44)
1997 commit 8f559d58b42b91f070bfa590d9c33e9a156e0066
1998 Merge: d94e764 931ffe3
1999 Author: Sage Weil <sage@redhat.com>
2000 Date: Sun Aug 30 08:36:37 2015 -0400
2002 Merge pull request #5366 from dachary/wip-12303-hammer
2004 arm: all programs that link to librados2 hang forever on startup
2006 commit 8ba6b2f628a0f80964afbf4a83b61c4eee9c78ce
2007 Author: Yehuda Sadeh <yehuda@redhat.com>
2008 Date: Thu Jul 30 14:47:15 2015 -0700
2010 rgw: url encode exposed bucket
2013 Don't send the bucket name back without url encoding it.
2015 Signed-off-by: Yehuda Sadeh <yehuda@redhat.com>
2016 (cherry picked from commit ad5507fe0bf72ed5bdf8353e315cc9092c740144)
2018 commit d94e7642d0b0213a4b650a5acffa050d042087eb
2019 Merge: 11a80d8 6417e8e
2020 Author: Loic Dachary <loic@dachary.org>
2021 Date: Sun Aug 30 13:41:01 2015 +0200
2023 Merge pull request #5202 from SUSE/wip-12293-hammer
2025 ceph.spec.in: rpm: cephfs_java not fully conditionalized
2027 Reviewed-by: Loic Dachary <ldachary@redhat.com>
2029 commit 11a80d8bec6ff704e794a2995d653cd5d5c1c979
2030 Merge: a792ae6 3728477
2031 Author: Loic Dachary <loic@dachary.org>
2032 Date: Sun Aug 30 13:40:48 2015 +0200
2034 Merge pull request #5203 from SUSE/wip-11997-hammer
2036 ceph.spec.in: rpm: not possible to turn off Java
2038 Reviewed-by: Loic Dachary <ldachary@redhat.com>
2040 commit a792ae65e2e03e1ca8804380cdbbfa1aec7a2123
2041 Merge: f7e76d2 8f78001
2042 Author: Loic Dachary <loic@dachary.org>
2043 Date: Sun Aug 30 13:40:30 2015 +0200
2045 Merge pull request #5204 from SUSE/wip-11629-hammer
2047 ceph.spec.in: SUSE/openSUSE builds need libbz2-devel
2049 Reviewed-by: Loic Dachary <ldachary@redhat.com>
2051 commit f7e76d2c9e681b3c5775f74c582830f546359703
2052 Merge: d1feb04 4eb58ad
2053 Author: Loic Dachary <loic@dachary.org>
2054 Date: Sun Aug 30 13:40:09 2015 +0200
2056 Merge pull request #5207 from SUSE/wip-12267-hammer
2058 ceph.spec.in: 50-rbd.rules conditional is wrong
2060 Reviewed-by: Loic Dachary <ldachary@redhat.com>
2062 commit d1feb0415b3b3dc9c5b003d8528079ccccd0729f
2063 Merge: 2dca2ec 8b576bd
2064 Author: Loic Dachary <loic@dachary.org>
2065 Date: Sun Aug 30 13:39:57 2015 +0200
2067 Merge pull request #5216 from SUSE/wip-12269-hammer
2069 ceph.spec.in: ceph-common needs python-argparse on older distros, but doesn't require it
2071 Reviewed-by: Loic Dachary <ldachary@redhat.com>
2073 commit 2dca2ece0478972e89ed6611a15892ad9e449cc8
2074 Merge: 64fafc9 0818e9f
2075 Author: Loic Dachary <loic@dachary.org>
2076 Date: Sun Aug 30 13:39:31 2015 +0200
2078 Merge pull request #5264 from SUSE/wip-12361-hammer
2080 ceph.spec.in: snappy-devel for all supported distros
2082 Reviewed-by: Loic Dachary <ldachary@redhat.com>
2084 commit 64fafc95785024421cbb7ccae41ed4276eeb8fe9
2085 Merge: 4675971 add0f1e
2086 Author: Loic Dachary <loic@dachary.org>
2087 Date: Sun Aug 30 13:39:14 2015 +0200
2089 Merge pull request #5368 from dachary/wip-12331-hammer
2091 ceph: cli throws exception on unrecognized errno
2093 Reviewed-by: Loic Dachary <ldachary@redhat.com>
2095 commit 4675971ce3331f1285398d1556a594f671e489c0
2096 Merge: 53cdcf2 e004941
2097 Author: Loic Dachary <loic@dachary.org>
2098 Date: Sun Aug 30 13:39:00 2015 +0200
2100 Merge pull request #5371 from dachary/wip-12494-hammer
2102 ceph tell: broken error message / misleading hinting
2104 Reviewed-by: Loic Dachary <ldachary@redhat.com>
2106 commit 53cdcf20482823cdb8682dde89706d6f21d8d0d2
2107 Merge: 362874c eccf369
2108 Author: Loic Dachary <loic@dachary.org>
2109 Date: Sun Aug 30 13:38:43 2015 +0200
2111 Merge pull request #5385 from dachary/wip-12501-hammer
2113 error in ext_mime_map_init() when /etc/mime.types is missing
2115 Reviewed-by: Loic Dachary <ldachary@redhat.com>
2117 commit 362874ca39e32c3bc5ee13b13f905fd7693d9449
2118 Merge: 2044f3c 8804b3f
2119 Author: Loic Dachary <loic@dachary.org>
2120 Date: Sun Aug 30 13:38:29 2015 +0200
2122 Merge pull request #5411 from SUSE/wip-12446-hammer
2124 ceph.spec.in: radosgw requires apache for SUSE only -- makes no sense
2126 Reviewed-by: Loic Dachary <ldachary@redhat.com>
2128 commit 2044f3cd7b220cc642245e7ca93c21a17cc8c536
2129 Merge: 260e24b a785193
2130 Author: Loic Dachary <loic@dachary.org>
2131 Date: Sun Aug 30 13:38:13 2015 +0200
2133 Merge pull request #5412 from SUSE/wip-12448-hammer
2135 ceph.spec.in: useless %py_requires breaks SLE11-SP3 build
2137 Reviewed-by: Loic Dachary <ldachary@redhat.com>
2139 commit 260e24bba0e83dfa61d25ac0489f656244211db4
2140 Merge: dc944fb b575ecc
2141 Author: Loic Dachary <loic@dachary.org>
2142 Date: Sun Aug 30 13:30:18 2015 +0200
2144 Merge pull request #5318 from ceph/wip-12236-hammer
2146 Possible crash while concurrently writing and shrinking an image
2148 Reviewed-by: Loic Dachary <ldachary@redhat.com>
2150 commit dc944fbbedcee56dfbd9d4f5ac3c38d15a6621ea
2151 Merge: a8c1b4e 92272dd
2152 Author: Loic Dachary <loic@dachary.org>
2153 Date: Sun Aug 30 13:30:01 2015 +0200
2155 Merge pull request #5319 from ceph/wip-12235-hammer
2157 librbd: deadlock during cooperative exclusive lock transition
2159 Reviewed-by: Loic Dachary <ldachary@redhat.com>
2161 commit a8c1b4e7436832ed3040d5a90df713ef6029cb69
2162 Merge: a39aeee e971820
2163 Author: Loic Dachary <loic@dachary.org>
2164 Date: Sun Aug 30 13:29:26 2015 +0200
2166 Merge pull request #5296 from ceph/wip-12345-hammer
2168 librbd: correct issues discovered via lockdep / helgrind
2170 Reviewed-by: Loic Dachary <ldachary@redhat.com>
2172 commit a39aeeedf07518d65c9944476b98a0788585f5c1
2173 Merge: 6a949e1 153744d
2174 Author: Loic Dachary <loic@dachary.org>
2175 Date: Sun Aug 30 12:14:22 2015 +0200
2177 Merge pull request #5559 from ceph/wip-12682-hammer
2179 object_map_update fails with -EINVAL return code
2181 Reviewed-by: Loic Dachary <ldachary@redhat.com>
2183 commit 6a949e10198a1787f2008b6c537b7060d191d236
2184 Merge: 1e423af 4faa8e0
2185 Author: Loic Dachary <loic@dachary.org>
2186 Date: Sun Aug 30 12:11:10 2015 +0200
2188 Merge pull request #5468 from ceph/wip-12577-hammer
2190 osd: fix repair when recorded digest is wrong
2192 Reviewed-by: Loic Dachary <ldachary@redhat.com>
2194 commit 1e423afd3c7e534e302dbea0798a47f82acec141
2195 Merge: 574932b 0ca93db
2196 Author: Loic Dachary <loic@dachary.org>
2197 Date: Sun Aug 30 12:06:59 2015 +0200
2199 Merge pull request #5376 from dachary/wip-12433-hammer
2201 Show osd as NONE in ceph osd map <pool> <object> output
2203 Reviewed-by: Loic Dachary <ldachary@redhat.com>
2205 commit 574932b2f438787f1a330a271644d9787eecbb63
2206 Merge: ed162d4 d08db7a
2207 Author: Loic Dachary <loic@dachary.org>
2208 Date: Sun Aug 30 11:54:55 2015 +0200
2210 Merge pull request #5359 from ceph/wip-11470.hammer
2212 mon: PaxosService: call post_refresh() instead of post_paxos_update()
2214 Reviewed-by: Loic Dachary <ldachary@redhat.com>
2216 commit ed162d498051ac237be33ac6a6764ae0c61d57a6
2217 Merge: 41a245a bee8666
2218 Author: Josh Durgin <jdurgin@redhat.com>
2219 Date: Fri Aug 28 14:43:31 2015 -0400
2221 Merge pull request #5691 from jdurgin/wip-10399-hammer
2223 is_new_interval() fixes
2225 Reviewed-by: Samuel Just <sjust@redhat.com>
2227 commit 0bc909e54b0edd8a32a3e7db16877f1d875c3186
2228 Author: Kefu Chai <kchai@redhat.com>
2229 Date: Mon Aug 10 04:25:03 2015 -0700
2231 mon: add a cache layer over MonitorDBStore
2233 the cache of of leveldb does not perform well under some condition,
2234 so we need a cache in our own stack.
2236 * add an option "mon_osd_cache_size" to control the size of cache size
2240 Signed-off-by: Kefu Chai <kchai@redhat.com>
2241 (cherry picked from commit 5e99a578050976ca22b549812ac80d494fe7041d)
2244 src/mon/OSDMonitor.h
2245 minor differences in OSDMonitor ctor
2247 commit bee86660377cfaa74f7ed668dd02492f25553ff9
2248 Author: Jason Dillaman <dillaman@redhat.com>
2249 Date: Fri Aug 21 11:32:39 2015 -0400
2251 Objecter: pg_interval_t::is_new_interval needs pgid from previous pool
2253 When increasing the pg_num of a pool, an assert would fail since the
2254 calculated pgid seed would be for the pool's new pg_num value instead
2255 of the previous pg_num value.
2258 Backport: infernalis, hammer, firefly
2259 Signed-off-by: Jason Dillaman <dillaman@redhat.com>
2260 (cherry picked from commit f20f7a23e913d09cc7fc22fb3df07f9938ddc144)
2262 Conflicts: (hobject_t sort order not backported, trivial resolution)
2263 src/osdc/Objecter.cc
2266 commit b5418b9a7138b45b5fe4f24cfb679c2abb30ab71
2267 Author: Samuel Just <sjust@redhat.com>
2268 Date: Tue May 26 12:35:11 2015 -0700
2270 osd_types::is_new_interval: size change triggers new interval
2273 Signed-off-by: Samuel Just <sjust@redhat.com>
2274 (cherry picked from commit ff79959c037a7145f7104b06d9e6a64492fdb95f)
2276 commit f0283890d22e8d5c2210d565077ad3af1921f098 (refs/remotes/gh/liewegas-wip-hammer-feature-hammer)
2277 Author: Sage Weil <sage@redhat.com>
2278 Date: Thu Aug 27 14:09:50 2015 -0400
2280 include/ceph_features: define HAMMER_0_94_4 feature
2282 This is to constrain upgrades past hammer to version that include
2283 the appropriate compatibility fixes (e.g., hobject_t encoding).
2285 Signed-off-by: Sage Weil <sage@redhat.com>
2286 (cherry picked from commit 2868b49c0e39fdc7ae72af81962370c4f95a859e)
2289 # src/include/ceph_features.h
2291 commit 41a245aff77dde8a2ab212f9c91a73a23a77c40d
2292 Merge: 95cefea 65b380f
2293 Author: Sage Weil <sage@redhat.com>
2294 Date: Thu Aug 27 13:07:29 2015 -0400
2296 Merge remote-tracking branch 'gh/wip-12536-hammer' into hammer
2298 This includes wip-osd-compat-hammer.
2300 commit 65b380f7f28a36bb51a6771a2fd9b657ea2848ba (refs/remotes/gh/wip-12536-hammer)
2301 Merge: 00d802d 9b91adc
2302 Author: Sage Weil <sage@redhat.com>
2303 Date: Wed Aug 19 09:52:02 2015 -0400
2305 Merge remote-tracking branch 'gh/wip-osd-compat-hammer' into wip-12536-hammer
2307 commit 153744d7c596705c4f92bee5e827846b46c80141
2308 Author: Jason Dillaman <dillaman@redhat.com>
2309 Date: Mon Aug 10 09:39:50 2015 -0400
2311 tests: increase test coverage for partial encodes/decodes
2313 Multiple combinations of offsets/lengths are now tested when
2314 performing partial encodes/decodes of the bit vector.
2316 Signed-off-by: Jason Dillaman <dillaman@redhat.com>
2317 (cherry picked from commit 3e145f714ac9b2d599b45a058c6b93595e38f424)
2319 commit fca78765366f3ac365cfa98224aca5fb79b2a7fe
2320 Author: Jason Dillaman <dillaman@redhat.com>
2321 Date: Mon Aug 10 09:34:42 2015 -0400
2323 common: bit_vector extent calculation incorrect for last page
2325 It's highly probable that the last page in the bit vector will not
2326 be a full page size. As a result, the computed extents will extend
2327 beyond the data portion of the bit vector, resulting in a end_of_buffer
2332 Signed-off-by: Jason Dillaman <dillaman@redhat.com>
2333 (cherry picked from commit c6d98992691683524d3d96def83a90a6f5fe7f93)
2335 commit 3396a969741f74e5b1f55d8d7af258493ca26478
2336 Author: John Spray <john.spray@redhat.com>
2337 Date: Thu Jul 30 14:20:56 2015 +0100
2339 osd/OSDMap: handle incrementals that modify+del pool
2341 Because new_pools was processed after old_pools, if something
2342 was modified and then deleted in the same incremental
2343 map, then the resulting state would be a pool in the map
2344 that had no entry in pool_name.
2347 Signed-off-by: John Spray <john.spray@redhat.com>
2348 (cherry picked from commit 8d4932e721e32a33c1bada54e11d4ecccd868618)
2350 commit 3ab5d82a1c09142267c27156b69a3117d30c5b7d (refs/remotes/gh/wip-12432-hammer)
2351 Author: Yehuda Sadeh <yehuda@redhat.com>
2352 Date: Wed Jul 22 10:01:00 2015 -0700
2354 rgw: set http status in civetweb
2356 Need to set the http status in civetweb so that we report it correctly.
2359 Signed-off-by: Yehuda Sadeh <yehuda@redhat.com>
2360 (cherry picked from commit b8e28ab9f914bf48c9ba4f0def9a0deb9dbb93bc)
2362 commit 10a03839853246c535e639b6dc6cea45b8673642
2363 Author: Yehuda Sadeh <yehuda@redhat.com>
2364 Date: Fri Jul 31 11:03:29 2015 -0700
2366 civetweb: update submodule to support setting of http status
2368 Signed-off-by: Yehuda Sadeh <yehuda@redhat.com>
2370 commit 00d802d97ba5e1a8bde2414f8710494f15807fcd
2371 Author: Sage Weil <sage@redhat.com>
2372 Date: Tue Dec 23 11:03:17 2014 -0800
2374 hobject_t: fix get_boundary to work with new sorting regime
2376 The hash is no longer the most significant field; set everything that is
2377 more significant, too.
2379 Signed-off-by: Sage Weil <sage@redhat.com>
2380 (cherry picked from commit 45281efad1ae6b2b5ddd5549c56dbdd46671e5c5)
2382 commit 9b91adc560c8dffc1f5f7470d28c25f448b2b619 (refs/remotes/gh/wip-osd-compat-hammer)
2383 Author: Sage Weil <sage@redhat.com>
2384 Date: Tue Aug 4 13:21:50 2015 -0400
2386 mon: disallow post-hammer OSDs if there are up pre-hammer OSDs
2388 Force *all* OSDs to upgrade to hammer before allowing post-hammer
2389 OSDs to join. This prevents any pre-hammer OSDs from running at
2390 the same time as a post-hammer OSD.
2392 This commit, as well as the definition of the sentinal post-hammer
2393 feature, should get backported to hammer stable series.
2396 Signed-off-by: Sage Weil <sage@redhat.com>
2397 (cherry picked from commit e38d60e94503c7e8511a60a8dceceb3cd39c2bd8)
2400 # src/mon/OSDMonitor.cc
2402 commit 8a559c1e80bb7c70e13c46f1900896c9b8fcd9a9
2403 Author: Sage Weil <sage@redhat.com>
2404 Date: Tue Aug 4 13:16:11 2015 -0400
2406 include/ceph_features: define MON_METADATA feature
2408 This was added in master in 40307748b9b8f84c98354a7141da645a3c556ba9. We
2409 define it but notably do not support it!
2411 Signed-off-by: Sage Weil <sage@redhat.com>
2413 commit 4faa8e04069417a6e2a6c4ea5c10b54d76aa00c4 (refs/remotes/gh/wip-12577-hammer)
2414 Author: Sage Weil <sage@redhat.com>
2415 Date: Fri Mar 27 10:17:12 2015 -0700
2417 osd: include newlines in scrub errors
2419 We may log more than one of these in a scrubmap; make sure they are
2422 Signed-off-by: Sage Weil <sage@redhat.com>
2423 (cherry picked from commit 2b3dd1bb7ab360e8f467a47b7ce3a0e4d3a979b2)
2425 commit 455eb2a89197e9609c5bb510112d661d388b4c85
2426 Author: Sage Weil <sage@redhat.com>
2427 Date: Fri Mar 27 10:04:00 2015 -0700
2429 osd: fix condition for loggin scrub errors
2431 We may have an error to log even though the authoritative map is empty:
2432 when we have a digest that doesn't match all replicas.
2434 Signed-off-by: Sage Weil <sage@redhat.com>
2435 (cherry picked from commit d23766be902080ffb29528acb5a00d70beb97a9f)
2437 commit 67e7946ce8e28f00db9588bed670ef5141268f41
2438 Author: Sage Weil <sage@redhat.com>
2439 Date: Fri Mar 27 10:03:22 2015 -0700
2441 osd: fix fallback logic; move into be_select_auth_object
2443 The fallback behavior was printing a promising message but then doing
2444 a 'continue' and not actually scrubbing the object. Instead, fall back to
2445 a less-bad auth inside be_select_auth_object.
2447 Signed-off-by: Sage Weil <sage@redhat.com>
2448 (cherry picked from commit b597db59fedc9a96dbe4b37f03b819dcf8fdb1bb)
2450 commit 0f57c7017a98f7e416a160375b8b80355b8e0fc3
2451 Author: Sage Weil <sage@redhat.com>
2452 Date: Fri Mar 27 09:34:44 2015 -0700
2454 osd: log a scrub error when we can't pick an auth object
2456 If we really can't find a suitable auth, log an error.
2458 Signed-off-by: Sage Weil <sage@redhat.com>
2459 (cherry picked from commit a476d8f3903ca78b5cf36d63059f13cb5268efe5)
2461 commit d4f4c5c4334add6acfc66407c2383d01094d384d
2462 Author: Sage Weil <sage@redhat.com>
2463 Date: Fri Mar 27 10:20:23 2015 -0700
2465 osd: repair record digest if all replicas match but do not match
2467 If the recorded digest does not match the replicas, and all replicas
2468 match, then (on repair) fix the recorded digest.
2470 Signed-off-by: Sage Weil <sage@redhat.com>
2471 (cherry picked from commit f581fecd8a16b9250fa96beec5aa26074f401028)
2473 commit acfed6b5eae764d381c88ed7d3f4942c61191127
2474 Author: Sage Weil <sage@redhat.com>
2475 Date: Thu Mar 26 13:52:20 2015 -0700
2477 osd: move recorded vs on disk digest warning into be_compare_scrubmaps
2479 This is a better place for it. While we are here, change the wording to
2480 clearly call out the 'recorded' digest (from object_info_t) vs the 'on
2481 disk' digest (what we observed during scrub).
2483 Signed-off-by: Sage Weil <sage@redhat.com>
2484 (cherry picked from commit cf349ff35e9111ca52db1b0aad3753d0680b486d)
2486 commit 674029bc6c519e82011d1429ab5eddff9720c2dc
2487 Author: Sage Weil <sage@redhat.com>
2488 Date: Thu Mar 26 13:46:22 2015 -0700
2490 osd: be slightly paranoid about value of okseed
2492 The scrubber.seed value is set based on the peer feature; check for that
2493 too explicitly, and assert the scrubber.seed value matches.
2495 No change in behavior here.
2497 Signed-off-by: Sage Weil <sage@redhat.com>
2498 (cherry picked from commit e34d31b7a9a71132a17b95cf5ad60255ece60243)
2500 commit f2002b7fa2af2db52037da4c1cf66d405a4941e7
2501 Author: Sage Weil <sage@redhat.com>
2502 Date: Thu Mar 26 13:44:25 2015 -0700
2504 osd: be precise about "known" vs "best guess"
2506 We cannot assume that the auth info has the "known" digest; all replicas
2507 may have a matching digest that does not match the oi, or we may choose
2508 a different auth info for some other reason. Verify that the digest
2509 matches the oi before calling it "known".
2511 Signed-off-by: Sage Weil <sage@redhat.com>
2512 (cherry picked from commit 546d1c7b8224d96d0cc93dc3c96f52787cd69413)
2514 commit 4e5d146ad36ef0380eb72e45dca651494658aeeb
2515 Author: Sage Weil <sage@redhat.com>
2516 Date: Thu Mar 26 13:24:11 2015 -0700
2518 osd: record digest if object is clean (vs entire scrub chunk)
2520 If an individual object is clean, record it in the missing_digest map.
2521 If not, leave it out. This lets us drop the old condition that only
2522 recorded any digests if the entire scrub chunk was clean.
2524 Signed-off-by: Sage Weil <sage@redhat.com>
2525 (cherry picked from commit c39e0e5f126028f552d410ab6924cac6e1f714fb)
2527 commit 1357ed1bd0e250b942bcba0346c97c24bb79a5d1
2528 Author: Sage Weil <sage@redhat.com>
2529 Date: Mon Aug 3 13:05:45 2015 -0400
2531 hobject_t: decode future hobject_t::get_min() properly
2533 The post-hammer wip-temp branch changed hobject_t::get_min() so that pool
2534 is INT64_MIN instead of -1 and neglected to deal with the encoding compat
2535 with older versions. Compensate on hammer by mapping INT64_MIN to -1
2536 locally. See commit ff99af38df830e215359bfb8837bf310a2023a4d.
2538 Note that this means upgrades from hammer to post-hammer *must* include
2539 this fix prior to the upgrade. This will need to be well-documented in the
2542 Master gets a similar fix so that they map our min value to the new
2543 INT64_MIN one on decode.
2545 Fixes: #12536 (for hammer)
2546 Signed-off-by: Sage Weil <sage@redhat.com>
2548 commit 6d01d6b700300841f49f288a89d514f26a4f94b1
2549 Author: Samuel Just <sjust@redhat.com>
2550 Date: Wed Jul 29 11:49:57 2015 -0700
2552 OSDMonitor::preprocess_get_osdmap: send the last map as well
2556 Signed-off-by: Samuel Just <sjust@redhat.com>
2557 (cherry picked from commit 1320e29dfaee9995409a6d99b9ccaa748dc67b90)
2559 commit a7851933033473d11e69fa1f237c0ad8bbbd4a13
2560 Author: Nathan Cutler <ncutler@suse.com>
2561 Date: Thu Jul 9 11:51:13 2015 +0200
2563 ceph.spec.in: drop SUSE-specific %py_requires macro
2565 %py_requires expands to
2567 BuildRequires: /usr/bin/python
2568 PreReq: python = 2.7
2570 The BuildRequires: is already provided, and the PreReq is wrong because
2571 e.g. SLE11-SP3 (a platform we are trying to support) has Python 2.6.
2573 http://tracker.ceph.com/issues/12351 Fixes: #12351
2575 Signed-off-by: Nathan Cutler <ncutler@suse.com>
2576 (cherry picked from commit cbf6c7b46b42e163950bf7956e5b7565fca961c9)
2578 commit 8804b3f1ab5180c8a85b25b9b46b986fe7022868
2579 Author: Nathan Cutler <ncutler@suse.com>
2580 Date: Thu Jul 16 15:16:22 2015 +0200
2582 ceph.spec.in: remove SUSE-specific apache2-mod_fcgid dependency
2584 This package is no longer required for RGW to work in SUSE.
2586 http://tracker.ceph.com/issues/12358 Fixes: #12358
2588 Signed-off-by: Nathan Cutler <ncutler@suse.com>
2589 (cherry picked from commit f3f8d3660d0ece3139e5f3dc7881ccceadcf648d)
2593 In master, the "%if %else %endif" has been changed
2594 into two "%if %endif"s
2596 commit b575ecca9743ab08439bfe0b5283e483e2a799ce
2597 Author: Jason Dillaman <dillaman@redhat.com>
2598 Date: Tue Jun 23 14:20:16 2015 -0400
2600 tests: verify that image shrink properly handles flush op
2602 Signed-off-by: Jason Dillaman <dillaman@redhat.com>
2603 (cherry picked from commit 8dfcc72071d63c7867a5a59abe871cf5db673a0a)
2605 commit d4eb7bd63a5e46b790bad7001d2873c5c238cc90
2606 Author: Jason Dillaman <dillaman@redhat.com>
2607 Date: Tue Jun 23 14:18:20 2015 -0400
2609 librbd: invalidate cache outside cache callback context
2611 When shrinking an image, it's possible that the op flush callback
2612 will be from within the cache callback context. This would result
2613 in a deadlock when attempting to re-lock the cache lock in order to
2614 invalidate the cache.
2618 Signed-off-by: Jason Dillaman <dillaman@redhat.com>
2619 (cherry picked from commit 726d699b7790c7e371279281ab32cd3aeb8ece7b)
2621 commit 92272dd676ebb67a3095ce3933cac24809b054da
2622 Author: Jason Dillaman <dillaman@redhat.com>
2623 Date: Fri May 15 11:18:29 2015 -0400
2625 librbd: don't cancel request lock early
2627 It's possible that a stale notice is received and will
2628 be discarded after the request lock has been canceled.
2629 This will stale the client.
2631 Signed-off-by: Jason Dillaman <dillaman@redhat.com>
2632 (cherry picked from commit d9dd5c5890029107df40ed438f23fb9865d7de29)
2634 commit 58ae92f1a716e1ffac6408793e19d47f99d63520
2635 Author: Jason Dillaman <dillaman@redhat.com>
2636 Date: Thu May 14 16:13:38 2015 -0400
2638 tests: new test for transitioning exclusive lock
2640 Signed-off-by: Jason Dillaman <dillaman@redhat.com>
2641 (cherry picked from commit f97ce46ea4c9c7cdc6e37e3759871d5b41b25769)
2643 commit 7b21ccba7c2885027b1d345baaad08a978fbc72f
2644 Author: Jason Dillaman <dillaman@redhat.com>
2645 Date: Thu May 7 13:31:50 2015 -0400
2647 tests: verify that librbd will periodically resend lock request
2649 Signed-off-by: Jason Dillaman <dillaman@redhat.com>
2650 (cherry picked from commit d2a1c226ab3f7e202f62896f0c80c4cf3607cdaf)
2652 commit c95b37f48c129ef6780f67b326e97957f3771472
2653 Author: Jason Dillaman <dillaman@redhat.com>
2654 Date: Fri May 15 10:49:36 2015 -0400
2656 common: Mutex shouldn't register w/ lockdep if disabled
2658 Signed-off-by: Jason Dillaman <dillaman@redhat.com>
2659 (cherry picked from commit 879b8a73e2452332b26b8f3428ff5e3e0af8ddad)
2661 commit 117205af89398457e197793505381622705488b2
2662 Author: Jason Dillaman <dillaman@redhat.com>
2663 Date: Fri May 15 10:47:04 2015 -0400
2665 librbd: improve debugging output for ImageWatcher
2667 Include the instance pointer so that different images
2668 can be differentiated in the logs.
2670 Signed-off-by: Jason Dillaman <dillaman@redhat.com>
2671 (cherry picked from commit b951a7398e3e749cb64a3f6a604212a5627e82a0)
2673 commit 08ae01231754d3010c1ede762579cf75c1c7460d
2674 Author: Jason Dillaman <dillaman@redhat.com>
2675 Date: Fri May 15 10:45:04 2015 -0400
2677 librados_test_stub: watcher id should be the instance id (gid)
2679 Signed-off-by: Jason Dillaman <dillaman@redhat.com>
2680 (cherry picked from commit 3e1e561cc29043eb64dff8afb3a2c1ed77788e93)
2682 commit 704c0e09479d435ecbb84693a786c60694732b19
2683 Author: Jason Dillaman <dillaman@redhat.com>
2684 Date: Thu May 7 12:51:49 2015 -0400
2686 librbd: retry lock requests periodically until acquired
2688 If the exclusive lock owner acks the lock release request but crashes
2689 before it actually releases the lock, the requestor will wait forever.
2690 Therefore, after a certain timeout, retry the request again until it
2695 Signed-off-by: Jason Dillaman <dillaman@redhat.com>
2696 (cherry picked from commit 37c74e6e5274208d1b6efaf315afec03ea7eaa82)
2698 commit dbaaed9cf99121d0c97c22a695a7a6e872a11f48
2699 Author: Jason Dillaman <dillaman@redhat.com>
2700 Date: Thu May 7 12:35:36 2015 -0400
2702 librbd: don't hold owner_lock for write during flush
2704 The various IO callback codepaths will attempt to take
2705 the lock, which will result in deadlock since the flush
2710 Signed-off-by: Jason Dillaman <dillaman@redhat.com>
2712 (cherry picked from commit 2b6d0633d5c89de3a557cdd72621dfc19f0540de)
2714 commit e9718201c6ed785c61077797a74ffda7e0438fb9
2715 Author: Jason Dillaman <dillaman@redhat.com>
2716 Date: Mon Jun 22 15:30:02 2015 -0400
2718 lockdep: do not automatically collect all backtraces
2720 It is expensive to collect backtraces every time a lock is
2721 checked in order to provide cycle backtraces. The backtraces
2722 can be forced on for specific locks or globally via the new
2723 config option "lockdep_force_backtrace".
2725 Signed-off-by: Jason Dillaman <dillaman@redhat.com>
2726 (cherry picked from commit 7354d25f56c5e004f288febdee2b6961c069163b)
2728 commit 27f7042e923366c31748c4cc9992c1a8cb37b457
2729 Author: Jason Dillaman <dillaman@redhat.com>
2730 Date: Tue Jun 9 13:20:54 2015 -0400
2732 librbd: flush operations need to acquire owner lock
2734 Cache writeback operations will expect the owner lock to be held.
2736 Signed-off-by: Jason Dillaman <dillaman@redhat.com>
2737 (cherry picked from commit adfa2e0070ccca7b6556d3bfc5fac7ce4d43a4d0)
2739 commit 5b39983025f510c3119ebb375870e7669be7ddd0
2740 Author: Jason Dillaman <dillaman@redhat.com>
2741 Date: Thu May 21 00:13:31 2015 -0400
2743 librbd: avoid infinite loop if copyup fails
2745 Signed-off-by: Jason Dillaman <dillaman@redhat.com>
2746 (cherry picked from commit 43e0e3cd63f0067217ed0811d73f6c546f3027be)
2748 commit 88b583b249a74a71bc87409015f7a8d85200e1bf
2749 Author: Jason Dillaman <dillaman@redhat.com>
2750 Date: Tue May 12 10:19:48 2015 -0400
2752 librbd: flush pending ops while not holding lock
2754 Signed-off-by: Jason Dillaman <dillaman@redhat.com>
2755 (cherry picked from commit 3d5cef38c37e5dda6b23751ad560851f1304d86d)
2757 commit a88b180571cb481a31cb94c249a3b486220232fa
2758 Author: Jason Dillaman <dillaman@redhat.com>
2759 Date: Tue May 12 10:07:21 2015 -0400
2761 tests: fix possible deadlock in librbd ImageWatcher tests
2763 Signed-off-by: Jason Dillaman <dillaman@redhat.com>
2764 (cherry picked from commit 742a85d667b26b3490d96270b5c500b08f2a5283)
2766 commit 321eb8d1c3f7d553addb780928e201acf5091132
2767 Author: Jason Dillaman <dillaman@redhat.com>
2768 Date: Mon May 11 13:59:49 2015 -0400
2770 tests: enable lockdep for librbd unit tests
2772 Signed-off-by: Jason Dillaman <dillaman@redhat.com>
2773 (cherry picked from commit 45cb9cb5980e90235b8244d0d61ece40bde4784a)
2775 commit bfe5b90921a3faaad0533388cb150cea20f47ae9
2776 Author: Jason Dillaman <dillaman@redhat.com>
2777 Date: Thu May 7 14:17:37 2015 -0400
2779 librbd: owner_lock should be held during flush request
2781 Flush might result in the cache writing out dirty objects, which
2782 would require that the owner_lock be held.
2784 Signed-off-by: Jason Dillaman <dillaman@redhat.com>
2785 (cherry picked from commit c9142fe35372cf69b7a56f334622a775a6b7c43f)
2787 commit 1e84fb05f48c146cc9d2d090be8c2d355326938b
2788 Author: Jason Dillaman <dillaman@redhat.com>
2789 Date: Thu May 7 14:06:16 2015 -0400
2791 osdc: ObjectCacher flusher might needs additional locks
2793 librbd requires the ObjectCacher flusher thread to acquire
2794 an additional lock in order to maintain lock ordering
2797 Signed-off-by: Jason Dillaman <dillaman@redhat.com>
2798 (cherry picked from commit a38f9e5104a6e08e130dc4f15ad19a06d9e63719)
2800 commit 506a45a906024d4bb5d3d4d6cc6cbb9eec39c5f2
2801 Author: Jason Dillaman <dillaman@redhat.com>
2802 Date: Thu Apr 30 16:11:03 2015 -0400
2804 librbd: fix recursive locking issues
2806 Signed-off-by: Jason Dillaman <dillaman@redhat.com>
2807 (cherry picked from commit 1b57cc1da7a51e6f8ffea689b94ef843732c20a4)
2809 commit acf5125fe27fb3b9de8b97c4e44fa1f5c61147fd
2810 Author: Jason Dillaman <dillaman@redhat.com>
2811 Date: Thu Apr 30 16:04:28 2015 -0400
2813 librbd: simplify state machine handling of exclusive lock
2815 It is expected that all IO is flushed and all async ops are cancelled
2816 prior to releasing the exclusive lock. Therefore, replace handling of
2817 lost exclusive locks in state machines with an assertion.
2819 Signed-off-by: Jason Dillaman <dillaman@redhat.com>
2820 (cherry picked from commit d6b733dbdd0aeb5d1e136dcbf30c58c80952651e)
2822 commit 9454f04e62ece39fdcdbb4eb5a83945f76bcc0a5
2823 Author: Jason Dillaman <dillaman@redhat.com>
2824 Date: Thu Apr 30 15:32:38 2015 -0400
2826 librbd: ObjectMap::aio_update can acquire snap_lock out-of-order
2828 Detected during an fsx run where a refresh and CoR were occurring
2829 concurrently. The refresh held the snap_lock and was waiting on
2830 the object_map_lock, while the CoR held object_map_lock and was
2831 waiting for snap_lock.
2834 Signed-off-by: Jason Dillaman <dillaman@redhat.com>
2835 (cherry picked from commit 8cbd92b1fe835b1eb3a898976f9507f51cc115b2)
2837 commit 3e0358e17e7ddb488109bce04bb08cd16fd5bb2f
2838 Author: Jason Dillaman <dillaman@redhat.com>
2839 Date: Thu Apr 16 14:15:10 2015 -0400
2841 librbd: move copyup class method call to CopyupRequest
2843 Move AbstractWrite's invocation of copyup to the CopyupRequest
2844 class. The AioRequest write path will now always create a
2845 CopyupRequest, which will now append the actual write ops to the
2848 Signed-off-by: Jason Dillaman <dillaman@redhat.com>
2849 (cherry picked from commit 7be3df67809925164237cc185f9f29e145f45768)
2851 commit 2ee64a80794297d4fc5aeafbb185818f7f6a77b8
2852 Author: Jason Dillaman <dillaman@redhat.com>
2853 Date: Tue Mar 31 12:28:11 2015 -0400
2855 librbd: simplify AioRequest constructor parameters
2857 Moved all parent overlap computation to within AioRequest so that
2858 callers don't need to independently compute the overlap. Also
2859 removed the need to pass the snap_id for write operations since
2860 it can only be CEPH_NOSNAP.
2862 Signed-off-by: Jason Dillaman <dillaman@redhat.com>
2863 (cherry picked from commit 46515971edab8888284b1a8307ffca88a9c75f5c)
2865 commit 3e71a7554da05aee06d5a2227808c321a3e3f0f1
2866 Author: Danny Al-Gaaf <danny.al-gaaf@bisect.de>
2867 Date: Sat Mar 14 01:16:31 2015 +0100
2869 librbd/AioRequest.h: fix UNINIT_CTOR
2873 CID 1274319: Uninitialized scalar field (UNINIT_CTOR)
2874 uninit_member: Non-static class member m_object_state is not
2875 initialized in this constructor nor in any functions that it calls.
2877 Signed-off-by: Danny Al-Gaaf <danny.al-gaaf@bisect.de>
2878 (cherry picked from commit 48f18ea0e1c4c6de1921ea2359448deb761461e7)
2880 commit cb57fe581df7c14d174e6eaff382d6a28ab658d6
2881 Author: Jason Dillaman <dillaman@redhat.com>
2882 Date: Fri Jul 17 15:04:10 2015 -0400
2884 librbd: add object state accessor to ObjectMap
2886 Signed-off-by: Jason Dillaman <dillaman@redhat.com>
2888 commit 9249ab7dacd43e7a546656b99013a595381fd5fd
2889 Author: Jason Dillaman <dillaman@redhat.com>
2890 Date: Thu Apr 30 15:41:59 2015 -0400
2892 librbd: AsyncObjectThrottle should always hold owner_lock
2894 Signed-off-by: Jason Dillaman <dillaman@redhat.com>
2895 (cherry picked from commit c352bcdc0f63eea55677fe3c3b5f0c61347c0db3)
2897 commit 26902b94f97edb189ae620c86a8dda8166df471a
2898 Author: Jason Dillaman <dillaman@redhat.com>
2899 Date: Thu Apr 30 15:34:43 2015 -0400
2901 librbd: execute flush completion outside of cache_lock
2903 Signed-off-by: Jason Dillaman <dillaman@redhat.com>
2904 (cherry picked from commit 5f157f20980de7e9a05fb933fb57efdc759da78b)
2906 commit 571220d6f4642dd3cd78988882645fdf647c150e
2907 Author: Jason Dillaman <dillaman@redhat.com>
2908 Date: Thu Apr 30 15:17:54 2015 -0400
2910 librbd: add AsyncRequest task enqueue helper method
2912 In order to support the invariant that all state machine
2913 callbacks occur without holding locks, transitions that
2914 don't always involve a librados call should queue their
2917 Signed-off-by: Jason Dillaman <dillaman@redhat.com>
2918 (cherry picked from commit 218bc2d0f8d90c9c64408cb22e26680e88138844)
2920 commit 8e280f4cfef0d2fdc706d43bbee0c377d288a457
2921 Author: Jason Dillaman <dillaman@redhat.com>
2922 Date: Thu May 7 15:32:27 2015 -0400
2924 librbd: disable lockdep on AioCompletion
2926 It is only used by clients and it causes a large slowdown
2927 in performance due to the rate at which the lock is constructed/
2928 destructed for each IO request.
2930 Signed-off-by: Jason Dillaman <dillaman@redhat.com>
2931 (cherry picked from commit 66e74641baeed9aadb7e474d6e6b142b7725722d)
2933 commit b38da48059792e9182fe2877786d9159007683d3
2934 Author: Jason Dillaman <dillaman@redhat.com>
2935 Date: Thu Apr 30 13:42:19 2015 -0400
2937 librbd: AioCompletion shouldn't hold its lock during callback
2939 The callback routine most likely will attempt to retrieve the result
2940 code, which will result in a recursive lock attempt.
2942 Signed-off-by: Jason Dillaman <dillaman@redhat.com>
2943 (cherry picked from commit 3ad19ae349ebb27ff6027e40883735240fa97a3c)
2945 commit 6fdd3f1ce69fa2e00c6f1bedd5f72352953e1e44
2946 Author: Jason Dillaman <dillaman@redhat.com>
2947 Date: Thu Apr 30 13:51:03 2015 -0400
2949 librbd: give locks unique names to prevent false lockdep failures
2951 Signed-off-by: Jason Dillaman <dillaman@redhat.com>
2952 (cherry picked from commit c474ee42b87975c04eeb2b40b976deb5a5e2d518)
2954 commit 70041499f99cad551b3aab16f086f725b33b25e2
2955 Author: Jason Dillaman <dillaman@redhat.com>
2956 Date: Thu Apr 30 13:40:16 2015 -0400
2958 librbd: complete cache read in a new thread context
2960 The ObjectCacher complete the read callback while still holding
2961 the cache lock. This introduces lock ordering issues which are
2962 resolved by queuing the completion to execute in a clean (unlocked)
2965 Signed-off-by: Jason Dillaman <dillaman@redhat.com>
2966 (cherry picked from commit 0024677dc8adfd610418ca47599dd95d3a5a6612)
2968 commit 65ef695eba0f5b5d87347ffb3407bb5d6d75b402
2969 Author: Jason Dillaman <dillaman@redhat.com>
2970 Date: Thu Mar 19 15:35:59 2015 -0400
2972 librbd: require callers to ObjectMap::aio_update to acquire lock
2974 This is needed to allow an atomic compare and update operation
2975 from the rebuild object map utility.
2977 Signed-off-by: Jason Dillaman <dillaman@redhat.com>
2978 (cherry picked from commit 2db758cb4cb6d88cbaf9842b1e5d0872a185d8df)
2980 commit 58b8fafb87b18f2a4d8c38f77ecf81ff58452409
2981 Author: Jason Dillaman <dillaman@redhat.com>
2982 Date: Thu Apr 30 13:38:29 2015 -0400
2984 log: fix helgrind warnings regarding possible data race
2986 Signed-off-by: Jason Dillaman <dillaman@redhat.com>
2987 (cherry picked from commit c1e14451775049bbd95bcb19a0b62ab5e2c0a7bb)
2989 commit a5203d32544d40cb51eb60fc906cc3ceaacbfe96
2990 Author: Jason Dillaman <dillaman@redhat.com>
2991 Date: Thu Apr 30 13:37:56 2015 -0400
2993 librados_test_stub: fix helgrind warnings
2995 Signed-off-by: Jason Dillaman <dillaman@redhat.com>
2996 (cherry picked from commit b65ae4bc26f2dcaaa9518d5cce0e8b83ea310de8)
2998 commit b73e87eb307137786e0f0d89362ef0e92e2670b8
2999 Author: Jason Dillaman <dillaman@redhat.com>
3000 Date: Thu Apr 30 13:36:26 2015 -0400
3002 librados_test_stub: add support for flushing watches
3004 Signed-off-by: Jason Dillaman <dillaman@redhat.com>
3005 (cherry picked from commit 6e400b9049ede5870e40e4dd2cb41874550eac25)
3007 commit 2fa35b1c5ca8e33959fff8c84eaa4feca0f67df3
3008 Author: Jason Dillaman <dillaman@redhat.com>
3009 Date: Thu Apr 30 13:29:12 2015 -0400
3011 common: lockdep now support unregistering once destructed
3013 librbd use of an image hierarchy resulted in lock names being
3014 re-used and incorrectly analyzed. librbd now uses unique lock
3015 names per instance, but to prevent an unbounded growth of
3016 tracked locks, we now remove lock tracking once a lock is
3019 Signed-off-by: Jason Dillaman <dillaman@redhat.com>
3020 (cherry picked from commit 7c7df2ce9f837628535d21df61ae4f13d809c4fa)
3022 commit 7b85c7bfe599644eb29e6b1e03733da4774c2eac
3023 Author: Jason Dillaman <dillaman@redhat.com>
3024 Date: Thu Apr 30 13:26:41 2015 -0400
3026 common: add valgrind.h convenience wrapper
3028 Conditionally support helgrind annotations if valgrind support is
3029 enabled during the build.
3031 Signed-off-by: Jason Dillaman <dillaman@redhat.com>
3032 (cherry picked from commit 650ad32d74a24d248dd8e19bff1bbd31f0bb224b)
3034 commit 6d3db5ffbef71c8927a9ffc68c5955dca0f5612c
3035 Author: Jason Dillaman <dillaman@redhat.com>
3036 Date: Wed Apr 8 17:24:08 2015 -0400
3038 librbd: add work queue for op completions
3040 Signed-off-by: Jason Dillaman <dillaman@redhat.com>
3041 (cherry picked from commit 21f990efe6c6d710376d5b1a49fc8cd53aa020e6)
3043 commit 64425e861400fde4703fb06529777c55650b6676
3044 Author: Jason Dillaman <dillaman@redhat.com>
3045 Date: Wed Apr 8 16:46:34 2015 -0400
3047 WorkQueue: ContextWQ can now accept a return code
3049 Signed-off-by: Jason Dillaman <dillaman@redhat.com>
3050 (cherry picked from commit e5ffae578e83c1e4fca3f328c937e7c9be8cc03c)
3052 commit eccf36980511de7ed122a38d426170496ffdea64
3053 Author: Ken Dreyer <kdreyer@redhat.com>
3054 Date: Tue Jun 23 13:41:53 2015 -0600
3056 packaging: RGW depends on /etc/mime.types
3058 If the mimecap RPM or mime-support DEB is not installed, then the
3059 /etc/mime.types file is not present on the system. RGW attempts to read
3060 this file during startup, and if the file is not present, RGW logs an
3063 ext_mime_map_init(): failed to open file=/etc/mime.types ret=-2
3065 Make the radosgw package depend on the mailcap/mime-support packages so
3066 that /etc/mime.types is always available on RGW systems.
3068 http://tracker.ceph.com/issues/11864 Fixes: #11864
3070 Signed-off-by: Ken Dreyer <kdreyer@redhat.com>
3071 (cherry picked from commit c74a2f86a1f9b15aae0e723b410bedfc1d01a79b)
3073 commit 7bddf5d4dad09b1f6d3b4e80e4bdd895191560b0
3074 Author: huangjun <hjwsm1989@gmail.com>
3075 Date: Thu Jul 16 09:42:55 2015 +0800
3077 rest_bench: bucketname is not mandatory as we have a default name
3079 Signed-off-by: huangjun <hjwsm1989@gmail.com>
3081 (cherry picked from commit bc56a87ea3b69b4856a2b9f9fb2a63dbfb1211c3)
3083 commit 6e7358b3958b7ee11ef612302b300d07f3982d54
3084 Author: huangjun <hjwsm1989@gmail.com>
3085 Date: Thu Jul 16 09:17:59 2015 +0800
3087 rest_bench: drain the work queue to fix a crash
3089 Signed-off-by: huangjun <hjwsm1989@gmail.com>
3091 (cherry picked from commit f3d34d8ff921dbd2ff21f6b72af7c73bb9c6940e)
3093 commit 1e055782f5ade94c19c2614be0d61d797490701e
3094 Author: John Spray <john.spray@redhat.com>
3095 Date: Tue Jul 21 16:09:32 2015 +0100
3097 auth: check return value of keyring->get_secret
3099 get_secret can fail to populate the passed CryptoKey, for
3100 example if the entity name is not found in the keyring. In
3101 this case, attempts to use the CryptoKey will lead to
3105 Signed-off-by: John Spray <john.spray@redhat.com>
3106 (cherry picked from commit 64e5041008744362fdbb16e16bc3e049a2d426aa)
3108 commit 256620e37fd94ee4b3af338ea6955be55529d0d6
3109 Author: Greg Farnum <gfarnum@redhat.com>
3110 Date: Thu Jul 16 04:45:05 2015 -0700
3112 Client: check dir is still complete after dropping locks in _readdir_cache_cb
3114 We drop the lock when invoking the callback, which means the directory
3115 we're looking at might get dentries trimmed out of memory. Make sure that
3116 hasn't happened after we get the lock back. If it *has* happened, fall back
3117 to requesting the directory contents from the MDS. Update the dirp location
3118 pointers after each entry to facilitate this.
3119 Because this requires we update the dirp->at_cache_name value on every loop,
3120 we rework the updating scheme a bit: to dereference the dn->name before
3121 unlocking, so we know it's filled in; and since we update it on every loop
3122 we don't need to refer to the previous dentry explicitly like we did before.
3124 This should also handle racing file deletes: we get back a trace on
3125 the removed dentry and that will clear the COMPLETE|ORDERED flags.
3129 Signed-off-by: Greg Farnum <gfarnum@redhat.com>
3130 (cherry picked from commit 62dd63761701a7e0f7ce39f4071dcabc19bb1cf4)
3132 commit 8a2ad0540994efe530a1775ebd949aeaf6609e57
3133 Author: Samuel Just <sjust@redhat.com>
3134 Date: Thu Apr 9 16:29:45 2015 -0700
3136 TestPGLog: fix invalid proc_replica_log test caes
3138 If 1'3 is non-divergent, no osd could correctly have created 2'3. Also,
3139 proc_replica_log does not add the extra entries from the auth log to the
3140 missing set, that happens later on in activate.
3142 Signed-off-by: Samuel Just <sjust@redhat.com>
3143 (cherry picked from commit 1333b676dba4ed4b5db854fe935bdadbf9f21d08)
3145 commit df71e6be7ccbe18d99200b2a5844a9f19ac70141
3146 Author: Samuel Just <sjust@redhat.com>
3147 Date: Thu Apr 9 16:28:23 2015 -0700
3149 TestPGLog: fix noop log proc_replica_log test case
3151 Need to fill in log.head, olog.head.
3153 Signed-off-by: Samuel Just <sjust@redhat.com>
3154 (cherry picked from commit 1e5b22053381580708af29a1536f1e106c9b479c)
3156 commit 549ff9a938c5126d4dcb15535845d5354cb44bb2
3157 Author: Samuel Just <sjust@redhat.com>
3158 Date: Thu Apr 9 15:01:41 2015 -0700
3160 TestPGLog: add test for 11358
3162 Signed-off-by: Samuel Just <sjust@redhat.com>
3163 (cherry picked from commit b61e5aef15d7448d24b713d13dce77697bc0af74)
3165 commit c224fc70e14bd76494bc437595c8b0b0d275694e
3166 Author: Samuel Just <sjust@redhat.com>
3167 Date: Thu Apr 9 15:01:29 2015 -0700
3169 PGLog::proc_replica_log: handle split out overlapping entries
3171 See the comment, if the overlapping entries between olog.head
3172 and log.tail were split out, last_update should be log.tail.
3175 Signed-off-by: Samuel Just <sjust@redhat.com>
3176 (cherry picked from commit 6561e0d955c4563a8505b717de41f5bd89119f63)
3178 commit b8176d0f21ce80d2e5a92057ab85f7618504cd9f
3179 Author: Ketor Meng <d.ketor@gmail.com>
3180 Date: Tue May 26 18:50:17 2015 +0800
3182 Mutex: fix leak of pthread_mutexattr
3184 Need pthread_mutexattr_destroy after pthread_mutexattr_init
3187 Signed-off-by: Ketor Meng <d.ketor@gmail.com>
3188 (cherry picked from commit 2b23327b3aa8d96341d501a5555195ca1bc0de8f)
3190 commit 43a72e47c4adae674e02bb262645d88ac528e2be
3191 Author: Xinze Chi <xmdxcxz@gmail.com>
3192 Date: Fri Jul 3 18:27:13 2015 +0800
3194 mon/PGMonitor: bug fix pg monitor get crush rule
3196 when some rules have been deleted before, the index in array of crush->rules
3197 is not always equals to crush_ruleset of pool.
3200 Reported-by: Ning Yao <zay11022@gmail.com>
3201 Signed-off-by: Xinze Chi <xmdxcxz@gmail.com>
3202 (cherry picked from commit 498793393c81c0a8e37911237969fba495a3a183)
3204 commit 0ca93db7d79f11e51af75e322941419b860bb2f8
3205 Author: Shylesh Kumar <shmohan@redhat.com>
3206 Date: Wed Jul 22 18:47:20 2015 +0530
3208 mon: ceph osd map shows NONE when an osd is missing
3210 Signed-off-by: Shylesh Kumar <shmohan@redhat.com>
3211 (cherry picked from commit 4d030302ea3709eb021710e873f1ebe62f951cec)
3213 commit 695f782e59baffbe8429f9daa47a4f1476a02976
3214 Author: Sage Weil <sage@redhat.com>
3215 Date: Sun Jun 7 20:05:58 2015 -0400
3217 crush/CrushWrapper: fix adjust_subtree_weight debug
3219 Signed-off-by: Sage Weil <sage@redhat.com>
3220 (cherry picked from commit 26be86f649a1e9846280204fe1e3f8344eb975ce)
3222 commit 0bd4c81ce41a22b22ca76e539de87260b140bbc4
3223 Author: Sage Weil <sage@redhat.com>
3224 Date: Fri Jun 5 12:15:45 2015 -0700
3226 crush/CrushWrapper: return changed from adjust_subtree_weight
3228 Signed-off-by: Sage Weil <sage@redhat.com>
3229 (cherry picked from commit 551ab2d69dd50a078d62340740d357d85ab8dcbb)
3231 commit 05fc59bee9138d04227255a2f31bd82efe94b1e9
3232 Author: Sage Weil <sage@redhat.com>
3233 Date: Fri Jun 5 12:15:22 2015 -0700
3235 crush/CrushWrapper: adjust subtree base in adjust_subtree_weight
3237 We were correctly adjusting all the children, but the subtree root was
3241 Signed-off-by: Sage Weil <sage@redhat.com>
3242 (cherry picked from commit 1da96ab8442522bda4fc39e05597f17b8ebddf89)
3244 commit d2f31adc49a7c6b8276dbf8773575e91ba5e18fa
3245 Author: Sage Weil <sage@redhat.com>
3246 Date: Fri Jun 5 12:12:41 2015 -0700
3248 unittest_crush_wrapper: test adjust_subtree_weight
3250 Signed-off-by: Sage Weil <sage@redhat.com>
3251 (cherry picked from commit b461dc2e970f139dcb25d93b32be6b4bb9f1c8eb)
3253 commit 0ccdf342e7cf0072d3c38c6e1a6cf3a87f154a28
3254 Author: Sage Weil <sage@redhat.com>
3255 Date: Fri Jun 5 12:17:26 2015 -0700
3257 unittest_crush_wrapper: attach buckets to root in adjust_item_weight test
3259 Signed-off-by: Sage Weil <sage@redhat.com>
3260 (cherry picked from commit a8351eb3de2b4069ffec9c4f561eb1519ea11aa2)
3262 commit 1e737532cfe1daf43bea3d9bd4e55faed560f0f2
3263 Author: Sage Weil <sage@redhat.com>
3264 Date: Fri Jun 5 12:12:32 2015 -0700
3266 unittest_crush_wrapper: parse env
3268 Signed-off-by: Sage Weil <sage@redhat.com>
3269 (cherry picked from commit 14c60b288b33e4cec9c9819d4668a5ff9855ccc1)
3271 commit cd11b887c6c586085a7014ba44123b115370a462
3272 Author: Guang G Yang <yguang@renownedground.corp.gq1.yahoo.com>
3273 Date: Thu Jul 2 05:29:47 2015 +0000
3275 osd: pg_interval_t::check_new_interval should not rely on pool.min_size to determine if the PG was active
3277 If the pool's min_size is set improperly, during peering, pg_interval_t::check_new_interval
3278 might wrongly determine the PG's state and cause the PG to stuck at down+peering forever
3281 Signed-off-by: Guang Yang yguang@yahoo-inc.com
3282 (cherry picked from commit 684927442d81ea08f95878a8af69d08d3a14d973)
3286 because PG::start_peering_interval has an assert
3287 that is not found in hammer in the context
3288 src/test/osd/types.cc
3289 because include/stringify.h is not included by
3292 commit c5f0e2222cc55a92ebc46d12b618fadb3d2fb577
3293 Author: Guang G Yang <yguang@renownedground.corp.gq1.yahoo.com>
3294 Date: Wed Jul 1 20:26:54 2015 +0000
3296 osd: Move IsRecoverablePredicate/IsReadablePredicate to osd_types.h
3298 Signed-off-by: Guang Yang <yguang@yahoo-inc.com>
3299 (cherry picked from commit 466b083166231ec7e4c069fef8c9e07d38accab9)
3301 commit 42bff0b9e2f80e32001082a608d63719332677e3
3302 Author: Joao Eduardo Luis <joao@suse.de>
3303 Date: Tue Jul 21 11:20:53 2015 +0100
3305 mon: OSDMonitor: fix hex output on 'osd reweight'
3309 Signed-off-by: Joao Eduardo Luis <joao@suse.de>
3310 (cherry picked from commit 787fa80c2746fde44ac0583ff7995ec5be9a672d)
3312 commit e00494118ee42322784988fe56623cbc7aac4cc9
3313 Author: Kefu Chai <kchai@redhat.com>
3314 Date: Wed Jul 15 19:03:33 2015 +0800
3316 ceph.in: print more detailed warning for 'ceph <type> tell'
3318 put the full signature of "tell <target> <command> [options...]"
3319 instead of "tell {0}.<id>", which could be misleading somehow.
3322 Signed-off-by: Kefu Chai <kchai@redhat.com>
3323 (cherry picked from commit aa1a5222eebd0256d4ecffb0eefeb0f78ecf0a1f)
3325 commit f18900f2eb724acd898e8b32fe75c0850d7cf94c
3326 Author: Kefu Chai <kchai@redhat.com>
3327 Date: Wed Jul 15 18:01:52 2015 +0800
3329 ceph.in: print more detailed error message for 'tell' command
3331 * we do not allow user specify a certain daemon when starting an
3332 interactive session. the existing error message could lead to
3333 some confusion. so put more details in it.
3336 Signed-off-by: Kefu Chai <kchai@redhat.com>
3337 (cherry picked from commit 972dc9145235ff0f60663369ebcab8c16628b89f)
3339 commit 9916d37ead705d467b1d730adb15bb6974b3bf1b
3340 Author: huangjun <hjwsm1989@gmail.com>
3341 Date: Mon Jul 20 10:50:20 2015 +0800
3343 mon/PGMonitor: avoid uint64_t overflow when checking pool 'target/max' status.
3346 Signed-off-by: huangjun <hjwsm1989@gmail.com>
3348 (cherry picked from commit f84e6b8157e78fd05cd1c8ac957c92df17e6ba06)
3350 commit 4457d3eb70712ec3b7d8e98af973e099c8c7d258
3351 Author: renhwztetecs <rhwlyw@163.com>
3352 Date: Tue Jul 21 18:55:00 2015 +0800
3354 Update OSDMonitor.cc
3356 OSDMonitor::get_pools_health(), s/objects/bytes/
3359 Signed-off-by: ren.huanwen@zte.com.cn
3360 (cherry picked from commit 7fc13c9d6b9a4962d7640240416105d8f558d600)
3362 commit add0f1e5e229c9ad66e8ef77ad59e0e390e20db6
3363 Author: Kefu Chai <kchai@redhat.com>
3364 Date: Thu Apr 16 18:29:00 2015 +0800
3366 ceph.in: do not throw on unknown errno
3368 some of the errnos are not listed in errno.errorcode. if we happen
3369 to run into them, print 'Unknown' instead.
3372 Signed-off-by: Kefu Chai <kchai@redhat.com>
3373 (cherry picked from commit 53514ad97f6f75746613f09d464f69a7c31bed55)
3375 commit fa194744fdaed13949bffb6f2b8a0fc420006e0b
3376 Author: Sage Weil <sage@redhat.com>
3377 Date: Wed Mar 18 13:49:20 2015 -0700
3379 os/chain_xattr: handle read on chnk-aligned xattr
3381 If we wrote an xattr that was a multiple of a chunk, we will try to read
3382 the next chunk and get ENODATA. If that happens bail out of the loop and
3383 assume we've read the whole thing.
3385 Backport: hammer, firefly
3386 Signed-off-by: Sage Weil <sage@redhat.com>
3387 (cherry picked from commit 8614dcebf384b74b283cd352001a65fa26ba925c)
3389 commit 931ffe3b3a43157798717e64de8f46fadbcd9e9b
3390 Author: Boris Ranto <branto@redhat.com>
3391 Date: Wed Jun 3 12:24:48 2015 +0200
3393 common/Cycles.cc: skip initialization if rdtsc is not implemented
3395 The Cycles initialization gets stuck in infinite loop if rdtsc is not
3396 implemented. This patch fixes the issue by quitting the initialization
3399 The patch was cherry-picked from ubuntu patch by James Page, see
3401 https://bugzilla.redhat.com/show_bug.cgi?id=1222286
3403 for more details on the patch.
3405 Signed-off-by: James Page <james.page@ubuntu.com>
3406 (cherry picked from commit 35c5fd0091fc4d63b21207fb94e46b343519fd56)
3408 commit 0fde3a2465e156ebf9a5cdc6adc45d66d6d647fc
3409 Author: Haomai Wang <haomaiwang@gmail.com>
3410 Date: Thu Jul 9 13:42:42 2015 +0800
3412 buffer: Fix bufferlist::zero bug with special case
3415 Signed-off-by: Haomai Wang <haomaiwang@gmail.com>
3416 (cherry picked from commit 43f583d160ccaf879eaf0f3020e77860cf8d1df0)
3418 commit dabc61110a5773262614fc77d3653674aac421f7
3419 Author: Haomai Wang <haomaiwang@gmail.com>
3420 Date: Thu Jul 9 13:32:03 2015 +0800
3422 UnittestBuffer: Add bufferlist zero test case
3424 Signed-off-by: Haomai Wang <haomaiwang@gmail.com>
3425 (cherry picked from commit 577acf665948e531e50d36f5780ea19351a5bf65)
3427 commit d08db7a0677412dbe590c850976fb4fe2f503a79 (refs/remotes/gh/wip-11470.hammer)
3428 Author: Joao Eduardo Luis <joao@suse.de>
3429 Date: Fri Jun 12 19:21:10 2015 +0100
3431 mon: PaxosService: call post_refresh() instead of post_paxos_update()
3433 Whenever the monitor finishes committing a proposal, we call
3434 Monitor::refresh_from_paxos() to nudge the services to refresh. Once
3435 all services have refreshed, we would then call each services
3436 post_paxos_update().
3438 However, due to an unfortunate, non-critical bug, some services (mainly
3439 the LogMonitor) could have messages pending in their
3440 'waiting_for_finished_proposal' callback queue [1], and we need to nudge
3443 This patch adds a new step during the refresh phase: instead of calling
3444 directly the service's post_paxos_update(), we introduce a
3445 PaxosService::post_refresh() which will call the services
3446 post_paxos_update() function first and then nudge those callbacks when
3449 [1] - Given the monitor will send MLog messages to itself, and given the
3450 service is not readable before its initial state is proposed and
3451 committed, some of the initial MLog's would be stuck waiting for the
3452 proposal to finish. However, by design, we only nudge those message's
3453 callbacks when an election finishes or, if the leader, when the proposal
3454 finishes. On peons, however, we would only nudge those callbacks if an
3455 election happened to be triggered, hence the need for an alternate path
3456 to retry any message waiting for the initial proposal to finish.
3460 Signed-off-by: Joao Eduardo Luis <joao@suse.de>
3461 (cherry picked from commit 1551ebb63238073d2fd30201e6b656a8988e958c)
3463 commit 0818e9f78b11801aa619b9ec04eeffb2795c3f62
3464 Author: Nathan Cutler <ncutler@suse.com>
3465 Date: Thu Jul 9 13:28:22 2015 +0200
3467 ceph.spec.in: snappy-devel for all supported distros
3469 Signed-off-by: Nathan Cutler <ncutler@suse.com>
3470 (cherry picked from commit e4634ddcc1c0a2abe51679425d15e6f693811e17)
3476 commit 8b576bd529d7c1604f0a4cdf97ece64fc5d69447
3477 Author: Nathan Cutler <ncutler@suse.com>
3478 Date: Tue Jun 16 18:27:20 2015 +0200
3480 ceph.spec.in: python-argparse only in Python 2.6
3482 argparse is a widely-used Python module for parsing command-line arguments.
3483 Ceph makes heavy use of Python scripts, both in the build environment and on
3484 cluster nodes and clients.
3486 Until Python 2.6, argparse was distributed separately from Python proper.
3487 As of 2.7 it is part of the Python standard library.
3489 Although the python package in a given distro may or may not Provide:
3490 python-argparse, this cannot be relied upon.
3492 Therefore, this commit puts appropriate conditionals around Requires:
3493 python-argparse and BuildRequires: python-argparse. It does so for Red
3494 Hat/CentOS and SUSE only, because the last Fedora version with Python 2.6
3495 was Fedora 13, which is EOL.
3497 argparse is required by both the ceph and ceph-common packages, but since ceph
3498 requires ceph-common, the argparse Requires and BuildRequires need only appear
3499 once, under ceph-common.
3501 http://tracker.ceph.com/issues/12034 Fixes: #12034
3503 Signed-off-by: Nathan Cutler <ncutler@suse.com>
3504 (cherry picked from commit 23171c952997d104cfa6b5700ec0bb658a8e0915)
3506 commit cdde626f6371e1da501306ed16bc450ee1ec0b91
3507 Author: Nathan Cutler <ncutler@suse.com>
3508 Date: Mon Jul 13 18:12:01 2015 +0200
3510 ceph.spec.in: do not run fdupes, even on SLE/openSUSE
3512 In openSUSE there is a policy to use %fdupes in the spec file if RPMLINT
3513 complains about duplicate files wasting space in the filesystem.
3515 However, RPMLINT is not so complaining, so drop fdupes.
3517 http://tracker.ceph.com/issues/12301 Fixes: #12301
3519 Signed-off-by: Nathan Cutler <ncutler@suse.com>
3520 (cherry picked from commit 53072b9019caf72e0313b2804ea174237ed7da33)
3522 commit 6417e8eff942dbbf6cd231ffb73136b7bdcd837f
3523 Author: Nathan Cutler <ncutler@suse.com>
3524 Date: Fri Jun 12 17:50:58 2015 +0200
3526 rpm: add missing Java conditionals
3528 http://tracker.ceph.com/issues/11991 Fixes: #11991
3530 Signed-off-by: Nathan Cutler <ncutler@suse.com>
3531 (cherry picked from commit 04e91bbd929801de7f1d6a05086d47a781c2e361)
3533 commit 37284773a754ef4b91d2b5436d46a5fd2be4f67e
3534 Author: Owen Synge <osynge@suse.com>
3535 Date: Fri May 29 13:28:24 2015 +0200
3537 Add rpm conditionals : cephfs_java
3539 Extracted conditionals from SUSE rpm spec file to forward port.
3540 Original work done by Thorsten Behrens <tbehrens@suse.com>
3542 Signed-off-by: Owen Synge <osynge@suse.com>
3543 (cherry picked from commit c65538f489897821877421043300d9c008224f02)
3545 commit 8f78001c2e09d483dcbcd33c24e2ef8043119773
3546 Author: Nathan Cutler <ncutler@suse.cz>
3547 Date: Thu May 14 20:07:37 2015 +0200
3549 ceph.spec.in: SUSE/openSUSE builds need libbz2-devel
3551 http://tracker.ceph.com/issues/11629 Fixes: #11629
3553 Signed-off-by: Nathan Cutler <ncutler@suse.cz>
3554 (cherry picked from commit 584ed2e28b0a31821f29a99cf6078cfd76ee31fc)
3556 commit 4eb58ad2027148561d94bb43346b464b55d041a6
3557 Author: Nathan Cutler <ncutler@suse.com>
3558 Date: Thu Jun 25 22:37:52 2015 +0200
3560 ceph.spec.in: use _udevrulesdir to eliminate conditionals
3562 The conditionals governing where 50-rbd.rules is installed were not doing the
3563 right thing on SUSE distros.
3565 Start using the %_udevrulesdir RPM macro, while taking care that it is defined
3566 and set to the right value. Use it to eliminate some conditionals around other
3567 udev rules files as well.
3569 http://tracker.ceph.com/issues/12166 Fixes: #12166
3571 Signed-off-by: Nathan Cutler <ncutler@suse.com>
3572 (cherry picked from commit 8aa758ee7a14e93ded1d55b4aca7a54aa47d7174)
3576 hammer does not yet have _with_systemd or ceph_daemon.py