1 commit 8424145d49264624a3b0a204aedb127835161070 (tag: refs/tags/v0.80.11, refs/remotes/gh/firefly)
2 Author: Jenkins Build Slave User <jenkins-build@jenkins-slave-wheezy.localdomain>
3 Date: Tue Nov 17 18:57:48 2015 +0000
7 commit 2a14d16712fbe2ee9d7d2aef389e18fcb2809a47
9 Author: Yuri Weinstein <yuri.weinstein@gmail.com>
10 Date: Mon Nov 2 14:34:28 2015 -0800
12 Merge pull request #6451 from jcsp/wip-firefly-urls
14 qa: http://ceph.com/qa -> http://download.ceph.com/qa
16 commit a06179acb268b073fcf67fbbf19152c1a5a275af
17 Author: Sage Weil <sage@redhat.com>
18 Date: Thu Sep 24 18:10:46 2015 -0400
20 qa: http://ceph.com/qa -> http://download.ceph.com/qa
22 Signed-off-by: Sage Weil <sage@redhat.com>
23 (cherry picked from commit d4d65fbd6ad15320339751c604613ac63511e645)
25 commit c551622ca21fe044bc1083614c45d888a2a34aeb
26 Merge: 241edfa bff2f47
27 Author: Loic Dachary <loic@dachary.org>
28 Date: Fri Oct 23 21:47:27 2015 +0200
30 Merge pull request #5823 from SUSE/wip-11700-firefly
32 make the all osd/filestore thread pool suicide timeouts separately configurable
34 Reviewed-by: Samuel Just <sjust@redhat.com>
36 commit 241edfabf6a6c767afc1e283dd7387377ece36c7
37 Merge: 07e90f5 f39c791
38 Author: Loic Dachary <loic@dachary.org>
39 Date: Fri Oct 23 21:01:54 2015 +0200
41 Merge pull request #5815 from SUSE/wip-12838-firefly
43 Mutex Assert from PipeConnection::try_get_pipe
45 Reviewed-by: David Zafman <dzafman@redhat.com>
46 Reviewed-by: Samuel Just <sjust@redhat.com>
48 commit 07e90f54318c0a6202add4de16f965c6e331d513
49 Merge: eaf7dc3 d0d6727
50 Author: Loic Dachary <loic@dachary.org>
51 Date: Thu Oct 22 18:19:39 2015 +0200
53 Merge pull request #5529 from SUSE/wip-12586-firefly
55 FileStore calls syncfs(2) even it is not supported
57 Reviewed-by: Kefu Chai <kchai@redhat.com>
59 commit eaf7dc343eb78f09d9a4b1815f6bd6ceb4217ad7
60 Merge: a960327 98769f7
61 Author: Yehuda Sadeh <yehuda@redhat.com>
62 Date: Thu Oct 22 07:44:12 2015 -0700
64 Merge pull request #5532 from SUSE/wip-12635-firefly
66 swift smoke test fails on TestAccountUTF8
68 commit a96032738ba1e1cc35f21607eba06e7e241abd97
69 Merge: 15f3628 86f9e55
70 Author: Yehuda Sadeh <yehuda@redhat.com>
71 Date: Thu Oct 22 07:33:02 2015 -0700
73 Merge pull request #5997 from SUSE/wip-13178-firefly
75 rgw: content length parsing calls strtol() instead of strtoll()
77 commit 15f362808b7100a5bade24cc78e718b04d391b54
78 Merge: e010899 ebba1d5
79 Author: Loic Dachary <loic@dachary.org>
80 Date: Thu Oct 22 10:45:48 2015 +0200
82 Merge pull request #5526 from SUSE/wip-12584-firefly
84 OSD crash creating/deleting pools
86 Reviewed-by: Samuel Just <sjust@redhat.com>
88 commit e010899ac9cd4060c09f0f1c754f0840b5149f6b
89 Merge: 24ffe4c 8c02376
90 Author: Loic Dachary <loic@dachary.org>
91 Date: Thu Oct 22 10:43:29 2015 +0200
93 Merge pull request #5287 from SUSE/wip-12389-firefly
95 PGLog::proc_replica_log: correctly handle case where entries between olog.head and log.tail were split out
97 Reviewed-by: Sage Weil <sage@redhat.com>
98 Reviewed-by: Samuel Just <sjust@redhat.com>
100 commit 24ffe4c5156843ce69cae30087fb767063fd489a
101 Merge: f48450b c155df2
102 Author: Loic Dachary <loic@dachary.org>
103 Date: Wed Oct 21 10:03:13 2015 +0200
105 Merge pull request #6328 from dachary/wip-13551-firefly
107 qemu workunit refers to apt-mirror.front.sepia.ceph.com
109 Reviewed-by: Loic Dachary <ldachary@redhat.com>
111 commit c155df217287b980e176c3c98ee49753d6378740
112 Author: Yuan Zhou <yuan.zhou@intel.com>
113 Date: Fri Jun 19 00:02:20 2015 +0800
115 qa: Use public qemu repo
117 This would allow some external tests outside of sepia lab
119 Signed-off-by: Yuan Zhou <yuan.zhou@intel.com>
120 (cherry picked from commit 4731c1e35539c9506ff3fe3141553fad099d0eee)
122 commit d9ae7d80850b9ea4a84549c3afb4603cf9585b6a
123 Author: Sage Weil <sage@redhat.com>
124 Date: Thu Apr 16 08:29:41 2015 -0700
126 use git://git.ceph.com
128 Signed-off-by: Sage Weil <sage@redhat.com>
129 (cherry picked from commit 34c467ba06345eacd2fd0e79c162dfd0b22f0486)
132 qa/run_xfstests-obsolete.sh : does not exist in firefly
133 qa/workunits/suites/fsx.sh : trivial resolution
135 commit f48450bdc70eef7cef8f76bc30f678a6d8d52b78
136 Merge: f275ad2 b894b36
137 Author: Loic Dachary <loic@dachary.org>
138 Date: Wed Oct 21 01:02:08 2015 +0200
140 Merge pull request #5822 from SUSE/wip-12837-firefly
142 WBThrottle::clear_object: signal on cond when we reduce throttle values
144 Reviewed-by: David Zafman <dzafman@redhat.com>
146 commit f275ad21a48cc4f2812f3ab9e3e9b8b05469c3c7
147 Merge: 143e267 81bb6dd
148 Author: Loic Dachary <loic@dachary.org>
149 Date: Tue Oct 20 21:48:39 2015 +0200
151 Merge pull request #6325 from dachary/wip-13041-firefly
153 common/Thread:pthread_attr_destroy(thread_attr) when done with it
155 Reviewed-by: Loic Dachary <ldachary@redhat.com>
157 commit 143e267a30ef7aac04fdfa530babab529cf30c1a
158 Merge: 725401e 6d396a3
159 Author: Loic Dachary <loic@dachary.org>
160 Date: Tue Oct 20 21:46:28 2015 +0200
162 Merge pull request #6091 from SUSE/wip-13252-firefly
164 mon: does not check for IO errors on every transaction
166 Reviewed-by: Sage Weil <sage@redhat.com>
168 commit 725401e05d0fbe41d93b200fe39d73d9dddae80c
169 Merge: 8fd4dd7 c85d063
170 Author: Loic Dachary <loic@dachary.org>
171 Date: Tue Oct 20 21:45:19 2015 +0200
173 Merge pull request #6087 from SUSE/wip-13244-firefly
175 client nonce collision due to unshared pid namespaces
177 Reviewed-by: Josh Durgin <jdurgin@redhat.com>
179 commit 8fd4dd7480148c664d2c99956a7c987b8707c3c3
180 Merge: 77a1b5f 6d82eb1
181 Author: Loic Dachary <loic@dachary.org>
182 Date: Tue Oct 20 21:44:32 2015 +0200
184 Merge pull request #6010 from liewegas/wip-globalid-firefly
186 increase globalid default for firefly
188 Reviewed-by: Kefu Chai <kchai@redhat.com>
190 commit 77a1b5f1292d8410a1c29cebb0da449b09a230d3
191 Merge: a3823fe 8c14cad
192 Author: Loic Dachary <loic@dachary.org>
193 Date: Tue Oct 20 21:43:33 2015 +0200
195 Merge pull request #5991 from SUSE/wip-13043-firefly
197 ThreadPool add/remove work queue methods not thread safe
199 Reviewed-by: Josh Durgin <jdurgin@redhat.com>
201 commit a3823fe76171479e4c8f071f6e8eee3f13a02441
202 Merge: e771054 48c929e
203 Author: Loic Dachary <loic@dachary.org>
204 Date: Tue Oct 20 21:42:28 2015 +0200
206 Merge pull request #5988 from SUSE/wip-13038-firefly
208 PG::handle_advance_map: on_pool_change after handling the map change
210 Reviewed-by: Sage Weil <sage@redhat.com>
212 commit e771054241fcba536a478f1d014feb20a434fa3f
213 Merge: e43d049 cd1396c
214 Author: Loic Dachary <loic@dachary.org>
215 Date: Tue Oct 20 21:41:19 2015 +0200
217 Merge pull request #5820 from SUSE/wip-12845-firefly
219 osd/PGLog.cc: 732: FAILED assert(log.log.size() == log_keys_debug.size())
221 Reviewed-by: Sage Weil <sage@redhat.com>
223 commit e43d0492a415be5a0feb361efbfbdda6e00a6234
224 Merge: 11ed9f9 2c2ffa1
225 Author: Loic Dachary <loic@dachary.org>
226 Date: Tue Oct 20 21:40:28 2015 +0200
228 Merge pull request #5814 from SUSE/wip-12840-firefly
230 recursive lock of md_config_t (0)
232 Reviewed-by: Sage Weil <sage@redhat.com>
234 commit 11ed9f9b263a1d3a8df7996d5644f0f6de7a9b1a
235 Merge: aead425 2c0d7fe
236 Author: Loic Dachary <loic@dachary.org>
237 Date: Tue Oct 20 21:39:32 2015 +0200
239 Merge pull request #5813 from SUSE/wip-12822-firefly
241 ceph-fuse crash in test_client_recovery
243 Reviewed-by: Loic Dachary <ldachary@redhat.com>
245 commit aead4255528e60a2ccc6f4d47a6cb482bfdccc82
246 Merge: 1c65d09 d3c9469
247 Author: Loic Dachary <loic@dachary.org>
248 Date: Tue Oct 20 21:37:58 2015 +0200
250 Merge pull request #5726 from ceph/wip-12752
252 is_new_interval() fixes
254 Reviewed-by: Sage Weil <sage@redhat.com>
256 commit 1c65d09ef8de2732cfa6f18d22de2c7b3fac0f0f
257 Merge: 1f0f4bb 411769c
258 Author: Loic Dachary <loic@dachary.org>
259 Date: Tue Oct 20 21:37:01 2015 +0200
261 Merge pull request #5698 from tchaikov/wip-12638-firefly
263 mon: add a cache layer over MonitorDBStore
265 Reviewed-by: Loic Dachary <ldachary@redhat.com>
267 commit 1f0f4bb5347cb935095dd374d6dcdec44ff175e2
268 Merge: 4a1e54f 278d732
269 Author: Loic Dachary <loic@dachary.org>
270 Date: Tue Oct 20 21:35:56 2015 +0200
272 Merge pull request #5619 from liewegas/wip-journal-firefly
274 os/FileJournal: Fix journal write fail, align for direct io
276 Reviewed-by: Loic Dachary <ldachary@redhat.com>
277 Reviewed-by: Nathan Cutler <ncutler@suse.cz>
279 commit 4a1e54fc88e43885c57049d1ad4c5641621b6c29
280 Merge: f272435 de53add
281 Author: Loic Dachary <loic@dachary.org>
282 Date: Tue Oct 20 21:31:52 2015 +0200
284 Merge pull request #5360 from ceph/wip-11786.firefly
286 mon: MonitorDBStore: get_next_key() only if prefix matches
288 Reviewed-by: Loic Dachary <ldachary@redhat.com>
290 commit f272435c1303327e7dae248179387d12e28ea6bc
291 Merge: 21452ed e7d08b1
292 Author: Loic Dachary <loic@dachary.org>
293 Date: Tue Oct 20 21:30:04 2015 +0200
295 Merge pull request #5236 from SUSE/wip-12312-firefly
297 mon: adding exsting pool as tier with --force-nonempty clobbers removed_snaps
299 Reviewed-by: Loic Dachary <ldachary@redhat.com>
301 commit 21452ed93b707984d5301a51040ba8c311678c9c
302 Merge: 14c8528 971eb3f
303 Author: Loic Dachary <loic@dachary.org>
304 Date: Tue Oct 20 21:28:31 2015 +0200
306 Merge pull request #5050 from SUSE/wip-12015-firefly
308 LibRadosTwoPools[EC]PP.PromoteSnap failure
310 Reviewed-by: Sage Weil <sage@redhat.com>
312 commit 14c85282399aca29a31566e0f068376b10cb95e6
313 Merge: b6d5768 7a8b6fb
314 Author: Loic Dachary <loic@dachary.org>
315 Date: Tue Oct 20 13:09:38 2015 +0200
317 Merge pull request #6207 from SUSE/wip-13417-firefly
319 Fix -Wno-format and -Werror=format-security options clash
321 Reviewed-by: Loic Dachary <ldachary@redhat.com>
323 commit b6d57686e1f482bf510e6c6c0ae56cff575f3eca
324 Merge: fca8a17 415335d
325 Author: Loic Dachary <loic@dachary.org>
326 Date: Tue Oct 20 13:08:01 2015 +0200
328 Merge pull request #6203 from SUSE/wip-13413-firefly
330 fedora21 has junit, not junit4
332 Reviewed-by: Loic Dachary <ldachary@redhat.com>
334 commit fca8a17bbb91ce5b3cb0c6a282cded03fe37956f
335 Merge: 6240b61 20ad17d
336 Author: Loic Dachary <loic@dachary.org>
337 Date: Tue Oct 20 13:06:14 2015 +0200
339 Merge pull request #5992 from SUSE/wip-13168-firefly
341 upstart: configuration is too generous on restarts
343 Reviewed-by: Loic Dachary <ldachary@redhat.com>
345 commit 6240b61fc48e4f7b306ec18eca51f40081b5160d
346 Merge: d2459d8 04b2a87
347 Author: Loic Dachary <loic@dachary.org>
348 Date: Tue Oct 20 13:04:40 2015 +0200
350 Merge pull request #5541 from dachary/wip-12418-firefly
352 ceph-disk: sometimes the journal symlink is not created
354 Reviewed-by: Loic Dachary <ldachary@redhat.com>
356 commit d2459d8b9c189e557f81bb7d10dd17cc53cf9af1
357 Merge: 3e0e3c5 644f213
358 Author: Loic Dachary <loic@dachary.org>
359 Date: Tue Oct 20 12:36:01 2015 +0200
361 Merge pull request #5729 from ceph/wip-11455-firefly
363 rgw: init some manifest fields when handling explicit objs
365 Reviewed-by: Loic Dachary <ldachary@redhat.com>
367 commit 3e0e3c518bf68e9d0ad7a79d804602634b9c6466
368 Merge: e44a777 e7931a7
369 Author: Loic Dachary <loic@dachary.org>
370 Date: Tue Oct 20 12:35:22 2015 +0200
372 Merge pull request #5730 from ceph/wip-12537-firefly
374 RGW returns requested bucket name raw in "Bucket" response header
376 Reviewed-by: Loic Dachary <ldachary@redhat.com>
378 commit e44a7775fb34ccdf7ff36e5e56aba9a309521be1
379 Merge: 028da25 3be204f
380 Author: Loic Dachary <loic@dachary.org>
381 Date: Tue Oct 20 12:34:01 2015 +0200
383 Merge pull request #5831 from gnuoy/wip-11140-firefly
385 /etc/init.d/radosgw restart does not work correctly
387 Reviewed-by: Loic Dachary <ldachary@redhat.com>
389 commit 81bb6dd7de8a191c11c16b3cdc07355ef3434435
390 Author: zqkkqz <zheng.qiankun@h3c.com>
391 Date: Fri Aug 7 10:49:45 2015 +0800
393 Common/Thread: pthread_attr_destroy(thread_attr) when done with it
394 When a thread attributes object is no longer required, it should be destroyed using the
395 pthread_attr_destroy() function. Destroying a thread attributes object has no effect on threads that were created using that object.
398 Signed-off-by: zqkqkz zheng.qiankun@h3c.com
399 (cherry picked from commit 9471bb838a420ef5f439191d87e5388fecccb9e6)
401 commit 77d6426a8323db91a60f246f385826c35f31521c
402 Author: Piotr Dałek <piotr.dalek@ts.fujitsu.com>
403 Date: Fri Jul 17 12:43:52 2015 +0200
405 Thread.cc: remove malloc/free pair
407 There's no need for mallocing pthread_attr_t in Thread::try_create(),
408 it can be located on stack as it is freed in same function. This reduces
409 pressure put on memory manager.
411 Signed-off-by: Piotr Dałek <piotr.dalek@ts.fujitsu.com>
412 (cherry picked from commit 614ac0f4e375f3dd008c8937d3f89f6005d6e0a0)
414 commit 7a8b6fb4c7b80c33dd3565fc4140d45fe70f0438
415 Author: Boris Ranto <branto@redhat.com>
416 Date: Fri Aug 15 19:34:27 2014 +0200
418 Fix -Wno-format and -Werror=format-security options clash
420 This causes build failure in latest fedora builds, ceph_test_librbd_fsx adds -Wno-format cflag but the default AM_CFLAGS already contain -Werror=format-security, in previous releases, this was tolerated but in the latest fedora rawhide it no longer is, ceph_test_librbd_fsx builds fine without -Wno-format on x86_64 so there is likely no need for the flag anymore
422 Signed-off-by: Boris Ranto <branto@redhat.com>
423 Reviewed-by: Sage Weil <sage@redhat.com>
424 (cherry picked from commit 7df67a544f6401fd31157b6e2a6ca590d87d7cbc)
429 commit 415335df53bf0d6b42f745b038c1c0c9c4aaa810
430 Author: Loic Dachary <ldachary@redhat.com>
431 Date: Tue Feb 3 16:14:23 2015 +0100
433 ceph.spec.in: junit always except for EPEL 6
435 The package was renamed a long time ago (around the Fedora 15
436 timeframe). The "junit4" name is only relevant for EPEL 6. For EPEL 7
437 and Fedora 20, the "junit" package has "Provides: junit4". And most
438 recently, in the junit package that ships in Fedora 21 and 22, the
439 package maintainer dropped the old Provides: line.
441 http://tracker.ceph.com/issues/10728 Fixes: #10728
443 Signed-off-by: Ken Dreyer <kdreyer@redhat.com>
444 Signed-off-by: Loic Dachary <ldachary@redhat.com>
445 (cherry picked from commit 4f36eaee9ea71ed48018ccb2d7eac3e809c72d82)
447 commit 98769f7ccf97d6a0c98438f1eab7a5705fb9920b
448 Author: Robin H. Johnson <robin.johnson@dreamhost.com>
449 Date: Fri Sep 4 01:07:48 2015 +0000
451 Fix casing of Content-Type header
453 It turns out, despite the HTTP spec declaring that header field names
454 should be case-insensitive, some clients treat them wrongly, and
455 consider "Content-type" to not match "Content-Type".
457 CyberDuck was one of those clients, now fixed upstream in
458 https://trac.cyberduck.io/ticket/8999
460 To reduce future occurances of this bug, fix the casing of the
461 Content-Type header, to strictly comply with the HTTP specification (be
462 strict about what you send, and generous about what you receive).
465 Backport: hammer, firefly
466 Signed-off-by: Robin H. Johnson <robin.johnson@dreamhost.com>
467 (cherry picked from commit 1b9fbffdc24160251b96cec820d62fb2a12b6eab)
469 commit ccf268502b0592f5ff58cd1a5920ec02a117acfb
470 Author: Orit Wasserman <owasserm@redhat.com>
471 Date: Thu Aug 13 15:02:03 2015 +0200
473 rgw: we should not overide Swift sent content type
478 Signed-off-by: Orit Wasserman <owasserm@redhat.com>
479 (cherry picked from commit 423cf136f15df3099c9266f55932542d303c2713)
481 commit 6829ee29a484a3b7055122addcda06a233d3ce53
482 Author: Radoslaw Zarzynski <rzarzynski@mirantis.com>
483 Date: Thu Jun 25 18:28:50 2015 +0200
485 rgw: send Content-Length in response for GET on Swift account.
488 Signed-off-by: Radoslaw Zarzynski <rzarzynski@mirantis.com>
489 (cherry picked from commit 5f47b11791cd46ce2771046e4888575869cc9685)
492 src/rgw/rgw_rest_swift.cc
493 omit dump_account_metadata() call
495 commit c9226afd7b9e54a8998bf88dfbd6753b262fdf9d
496 Author: Radoslaw Zarzynski <rzarzynski@mirantis.com>
497 Date: Tue Jun 23 11:54:09 2015 +0200
499 rgw: enforce Content-Type in Swift responses.
501 Swift sends Content-Type HTTP header even if the response
502 doesn't contain body. We have this behaviour implemented
503 until applying some changes in end_header() function.
504 Unfortunately, lack of Content-Type causes early exits in
505 many Tempest's tests for Swift API verification.
508 Signed-off-by: Radoslaw Zarzynski <rzarzynski@mirantis.com>
509 (cherry picked from commit 106aeba206736d4080326f9bc191876bed63370b)
511 commit 9139fbbc1ab3ae455d2e36b3a768601ccb216801
512 Author: Orit Wasserman <owasserm@redhat.com>
513 Date: Mon Jun 22 13:51:49 2015 +0200
515 rgw: force content_type for swift bucket stats request
519 Signed-off-by: Orit Wasserman <owasserm@redhat.com>
520 (cherry picked from commit da259e4da1329e1e448d75c48c6ee3d1318e19b2)
522 commit 767f1f74c9f015d6bb3376266ec77a10a2d2c59a
523 Author: Orit Wasserman <owasserm@redhat.com>
524 Date: Tue Apr 21 12:31:19 2015 +0200
526 rgw: force content-type header for swift account responses without body
528 This regression was introduced in commit 4cfeca03ade21861ab70ca759f94eb244c16cb39
532 Signed-off-by: Orit Wasserman <owasserm@redhat.com>
533 (cherry picked from commit 009664e560ee37ef6dc170a0155d6c749e151b06)
535 commit e3da03dd6be04e782c3e5786767bba4b4964fbec
536 Author: Orit Wasserman <owasserm@redhat.com>
537 Date: Thu Apr 9 20:04:59 2015 +0200
539 rgw: shouldn't return content-type: application/xml if content length is 0
543 Signed-off-by: Orit Wasserman <owasserm@redhat.com>
544 (cherry picked from commit 5cc92bb27cfbaee4af24dc312c9b98331f899794)
546 commit e7d08b1bca951f9a963eb0aeb4cd88abbe5723ba
547 Author: Samuel Just <sjust@redhat.com>
548 Date: Fri May 8 10:26:48 2015 -0700
550 test/librados/tier.cc: destroy and recreate cache pool on every test
552 Namespaces are not sufficient with the checks for 11493 in the mon.
554 Signed-off-by: Samuel Just <sjust@redhat.com>
555 (cherry picked from commit bef09e0cdb274cb1c87335a2af9ee532d14a4596)
557 commit d7f53442260dcfc92e7afa8bba2dd116c15e030d
558 Author: Sage Weil <sage@redhat.com>
559 Date: Wed Apr 29 12:34:25 2015 -0700
561 mon: prevent pool with snapshot state from being used as a tier
563 If we add a pool with snap state as a tier the snap state gets clobbered
564 by OSDMap::Incremental::propogate_snaps_to_tiers(), and may prevent OSDs
565 from starting. Disallow this.
570 Backport: hammer, giant, firefly
571 Signed-off-by: Sage Weil <sage@redhat.com>
572 (cherry picked from commit bbec53edf9e585af4e20bbc9ba9057d6fdfda342)
575 qa/workunits/cephtool/test.sh
576 properly co-exist with "# make sure we can't create an ec pool tier"
577 src/mon/OSDMonitor.cc
578 properly co-exist with preceding "if (tp->ec_pool())"
579 (The changes to both files would have applied cleanly if
580 https://github.com/ceph/ceph/pull/5389 had not been merged first.)
582 commit 028da252d85733fcc7d8495750b9ae97043cb0dd
583 Merge: ecabc67 f71a6eb
584 Author: Loic Dachary <loic@dachary.org>
585 Date: Sat Oct 3 09:40:33 2015 +0200
587 Merge pull request #5993 from SUSE/wip-13169-firefly
589 update docs to point to download.ceph.com
591 Reviewed-by: Ken Dreyer <kdreyer@redhat.com>
593 commit 971eb3f6dddb509d15e5e4bcfd5e34d42ae2e75b
594 Author: Sage Weil <sage@redhat.com>
595 Date: Thu Nov 13 22:33:10 2014 -0800
597 ceph_test_rados_api_tier: fix cleanup of whiteouts
599 Doing a normal DELETE on a whiteout returns ENOENT. Use the IGNORE_CACHE
600 flag to make sure it removes the whiteout, too.
603 Signed-off-by: Sage Weil <sage@redhat.com>
604 (cherry picked from commit 8dde6a61921798aa188bd69406c8fa7afeb96b45)
607 src/test/librados/TestCase.cc
608 for it of type ObjectIterator:
609 - use it->first instead of it->get_oid()
610 - use it->second instead of it->get_locator()
612 commit b134a629863b4601892db03d7c4377040d556bdd
613 Author: Sage Weil <sage@redhat.com>
614 Date: Thu Nov 13 22:32:20 2014 -0800
616 osd/ReplicatedPG: allow whiteout deletion with IGNORE_CACHE flag
618 If the client specifies IGNORE_CACHE, allow a regular DELETE to zap a
619 whiteout. Expand test case to verify this works.
621 Signed-off-by: Sage Weil <sage@redhat.com>
622 (cherry picked from commit 34e4d24)
625 src/test/librados/tier.cc
626 replaced NObjectIterator -> ObjectIterator
627 replaced cache_ioctx.nobjects_begin -> cache_ioctx.objects_begin
628 replaced cache_ioctx.nobjects_end -> cache_ioctx.objects_end
629 replace it->get_oid() with it->first for it of type ObjectIterator
631 commit 6d396a322e5ad61a3ca262b1c78a670481ff740c
632 Author: Sage Weil <sage@redhat.com>
633 Date: Wed Sep 23 10:58:01 2015 -0400
635 mon/Elector: do a trivial write on every election cycle
637 Currently we already do a small write when the *first* election in
638 a round happens (to update the election epoch). If the backend
639 happens to fail while we are already in the midst of elections,
640 however, we may continue to call elections without verifying we
643 Signed-off-by: Sage Weil <sage@redhat.com>
644 (cherry picked from commit ef909ccbdc303cce8a39edef255325127832ff16)
648 no MonitorDBStore::TransactionRef in firefly
650 commit d2bfa19077422eea37205fee9ed6fd5f2692f222
651 Author: Sage Weil <sage@redhat.com>
652 Date: Wed Sep 23 10:58:13 2015 -0400
654 mon/MonitorDBStore: assert/crash if there is a write error
656 Do this globally intead of relying on teh zillion mon callers to
657 check the error code. There are no cases where we want to
658 tolerate a commit failure.
661 Signed-off-by: Sage Weil <sage@redhat.com>
662 (cherry picked from commit 2fb7b1f0e33ada7c9a1be3de2f7708eb0760fcef)
664 commit c85d06384e795ab2717868ce45f01ae93181a6be
665 Author: Josh Durgin <jdurgin@redhat.com>
666 Date: Mon Sep 21 15:46:42 2015 -0700
668 use simplifed messenger constructor for clients
670 This is all mechanical except the calculation of the nonce, which is
671 now always randomized for clients.
674 Signed-off-by: Josh Durgin <jdurgin@redhat.com>
675 (cherry picked from commit a3a8c85b79afef67681c32c57b591c0e0a87a349)
681 src/librados/RadosClient.cc
682 src/mds/MDSUtility.cc
684 src/test/mon/test_mon_workloadgen.cc
685 - different arguments to Messenger::create() in firefly
687 commit ef24b4ec7245959b1960a415617439bc3baac9a3
688 Author: Sage Weil <sage@redhat.com>
689 Date: Mon Oct 6 21:20:08 2014 -0700
691 librados: use Messenger factory method
693 Signed-off-by: Sage Weil <sage@redhat.com>
694 (cherry picked from commit 5a860bec48348ed4c09256ccd394b33ace6b4f52)
696 commit 3a52acde3133aab2a717fefa967b0d404e50e2d8
697 Author: Sage Weil <sage@redhat.com>
698 Date: Mon Oct 6 08:50:48 2014 -0700
700 mon/MonClient: use generic Messenger factory
702 Signed-off-by: Sage Weil <sage@redhat.com>
703 (cherry picked from commit e55cb1f68ae1a04cc5d281bea57ee79ef4a8af38)
705 commit de8da652b6c969aa0446f0d3351f63b012063c71
706 Author: Josh Durgin <jdurgin@redhat.com>
707 Date: Mon Sep 21 15:20:53 2015 -0700
709 msg: add simplified messenger constructor
711 Several callers create messengers using exactly the same parameters:
713 - reading the ms type from cct that is also passed in
714 - a default entity_name_t::CLIENT
715 - the default features
717 Additionally, the nonce should be randomized and not depend on
718 e.g. pid, as it does in several callers now. Clients running in
719 containers can easily have pid collisions, leading to hangs, so
720 randomize the nonce in this simplified constructor rather than
721 duplicating that logic in every caller.
723 Daemons have meaningful entity_name_ts, and monitors currently depend
724 on using 0 as a nonce, so make this simple constructer
728 Signed-off-by: Josh Durgin <jdurgin@redhat.com>
729 (cherry picked from commit e3785b0b5fbff870adbd5f121021b671912386cf)
733 - different #includes in firefly
734 - different Messenger::create() arguments in firefly
736 commit ecabc6796f19c03947bb5b14da4e1b761ff8847f
737 Merge: 741dd5b 2e09a2c
738 Author: Sage Weil <sage@redhat.com>
739 Date: Wed Sep 23 17:09:18 2015 -0400
741 Merge pull request #6038 from liewegas/wip-13088
743 common/HeartBeat: include types
745 Reviewed-by: Kefu Chai <kchai@redhat.com>
747 commit 2e09a2c22ab885f8ec81dbc55f2c8fc0f2984543
748 Author: Sage Weil <sage@redhat.com>
749 Date: Wed Sep 23 17:06:52 2015 -0400
751 common/RWLock: include assert
753 Fixes atomic_t build issue without libatomic-ops.
756 Signed-off-by: Sage Weil <sage@redhat.com>
758 commit 741dd5b9433d1592ce91aea70ebec106a1354973
759 Author: Sage Weil <sage@redhat.com>
760 Date: Tue Sep 22 13:03:49 2015 -0400
762 .gitmodules: fix ceph-object-corpus source url
765 Signed-off-by: Sage Weil <sage@redhat.com>
767 commit a1997879aa26299519ed157017f8f5715c626dcc
768 Merge: ee81812 01dc221
769 Author: Sage Weil <sage@redhat.com>
770 Date: Tue Sep 22 12:29:08 2015 -0400
772 Merge pull request #6000 from SUSE/wip-12097-firefly
774 kernel_untar_build fails on EL7
776 commit ee8181206726712729235e4a5f8d2094a6e72f71
777 Merge: bb8b422 4e7de5b
778 Author: Yehuda Sadeh <yehuda@redhat.com>
779 Date: Mon Sep 21 13:07:49 2015 -0700
781 Merge pull request #5234 from SUSE/wip-12244-firefly
783 Civetweb RGW appears to report full size of object as downloaded when only partially downloaded
785 commit bb8b422beac22bd8c2e9c1ecd83e265a3de1a3e6
786 Merge: 4b08895 f602263
787 Author: Yehuda Sadeh <yehuda@redhat.com>
788 Date: Mon Sep 21 12:35:52 2015 -0700
790 Merge pull request #4535 from xinxinsh/wip-11416-firefly
792 rgw: use correct objv_tracker for bucket instance
794 commit 6d82eb165fdc91851f702a463022b26c50f5094b (refs/remotes/me/wip-globalid-firefly)
795 Author: Sage Weil <sage@redhat.com>
796 Date: Sun Jan 18 16:49:20 2015 -0800
798 mon: handle case where mon_globalid_prealloc > max_global_id
800 This triggers with the new larger mon_globalid_prealloc value. It didn't
801 trigger on the existing cluster I tested on because it already had a very
804 Signed-off-by: Sage Weil <sage@redhat.com>
805 (cherry picked from commit bbaf582e5bb54ae35327ffd234e4ccf0f07317db)
807 commit f545a0f430bf6f1e26983fc0ff20a645697f017c
808 Author: Sage Weil <sage@redhat.com>
809 Date: Sun Jan 18 10:39:25 2015 -0800
811 mon: change mon_globalid_prealloc to 10000 (from 100)
813 100 ids (session 100 authentications) can be consumed quite quickly if
814 the monitor is being queried by the CLI via scripts or on a large cluster,
815 especially if the propose interval is long (many seconds). These live in
816 a 64-bit value and are only "lost" if we have a mon election before they
817 are consumed, so there's no real risk here.
819 Backport: giant, firefly
820 Reviewed-by: Joao Eduardo Luis <joao@redhat.com>
821 Signed-off-by: Sage Weil <sage@redhat.com>
822 (cherry picked from commit 1d1215fe5f95c2bafee5b670cdae1353104636a0)
824 commit 01dc221f280ad25aa556f246414e3c3c42762956
825 Author: Greg Farnum <gfarnum@redhat.com>
826 Date: Tue Jun 16 08:13:41 2015 -0700
828 qa: update to newer Linux tarball
830 This should make newer gcc releases happier in their default configuration.
831 kernel.org is now distributing tarballs as .xz files so we change to that
832 as well when decompressing (it is supported by Ubuntu Precise so we should
837 Signed-off-by: Greg Farnum <gfarnum@redhat.com>
838 (cherry picked from commit 1ea3f47ab806d48ca7b045c2731d344eae3900e1)
841 qa/workunits/kernel_untar_build.sh
844 commit 86f9e55f0c151c0b9a289b475f87b6a11329e6e1
845 Author: Yehuda Sadeh <yehuda@redhat.com>
846 Date: Fri Jan 30 10:51:52 2015 -0800
848 rgw: use strict_strtoll() for content length
850 instead of strict_strtol().
852 Backport: giant, firefly
855 Reported-by: Axel Dunkel <ad@dunkel.de>
856 Signed-off-by: Yehuda Sadeh <yehuda@redhat.com>
857 (cherry picked from commit 1c25dbafb45caf1b261cfcec15b868a2ba6b5fef)
859 commit f71a6ebf1b371f9389865a0a33652841726ff77b
860 Author: Alfredo Deza <adeza@redhat.com>
861 Date: Fri Sep 18 14:13:02 2015 -0400
863 doc: correct links to download.ceph.com
865 Signed-off-by: Alfredo Deza <adeza@redhat.com>
866 (cherry picked from commit 387d7800359154950431d0984c756f43f21dd9b4)
869 doc/install/get-packages.rst
870 preferred the new version in all cases
872 commit 20ad17d271fb443f6c40591e205e880b5014a4f3
873 Author: Sage Weil <sage@redhat.com>
874 Date: Wed Jun 3 14:57:34 2015 -0400
876 upstart: limit respawn to 3 in 30 mins (instead of 5 in 30s)
878 It may take tens of seconds to restart each time, so 5 in 30s does not stop
879 the crash on startup respawn loop in many cases. In particular, we'd like
880 to catch the case where the internal heartbeats fail.
882 This should be enough for all but the most sluggish of OSDs and capture
883 many cases of failure shortly after startup.
886 Signed-off-by: Sage Weil <sage@redhat.com>
887 (cherry picked from commit eaff6cb24ef052c54dfa2131811758e335f19939)
889 commit 8c14cad0895590f19a6640c945b52213f30a9671
890 Author: Jason Dillaman <dillaman@redhat.com>
891 Date: Mon Aug 10 19:10:19 2015 -0400
893 WorkQueue: add/remove_work_queue methods now thread safe
895 These methods were not acquiring the ThreadPool lock when
896 manipulating the work_queue collection. This was causing
897 occasional crashes within librbd when opening and closing
901 Backport: hammer, firefly
902 Signed-off-by: Jason Dillaman <dillaman@redhat.com>
903 (cherry picked from commit 3e18449b01c1ab78d1bbfc1cf111aa9bdbef7b1f)
905 commit 48c929e689b0fa5138922fcb959be5d05296e59a
906 Author: Samuel Just <sjust@redhat.com>
907 Date: Thu Aug 27 11:08:33 2015 -0700
909 PG::handle_advance_map: on_pool_change after handling the map change
911 Otherwise, the is_active() checks in the hitset code can erroneously
912 return true firing off repops stamped with the new epoch which then get
913 cleared in the map change code. The filestore callbacks then pass the
914 interval check and call into a destroyed repop structure.
917 Backport: hammer,firefly
918 Signed-off-by: Samuel Just <sjust@redhat.com>
919 (cherry picked from commit 14e02bc90a463805f4c3e2de210892067a52514b)
921 commit 4b0889567d7fe28ef6976f8dbb0d329b03200986
922 Merge: 0a740d7 5426421
923 Author: Nathan Cutler <presnypreklad@gmail.com>
924 Date: Thu Sep 10 11:46:41 2015 +0200
926 Merge pull request #5200 from SUSE/wip-12289-firefly
928 mon: scrub error (osdmap encoding mismatch?) upgrading from 0.80 to ~0.80.2
930 Reviewed-by: Kefu Chai <kchai@redhat.com>
932 commit de53addac8234037a66cdd45cf8007deba7a0530 (refs/remotes/gh/wip-11786.firefly)
933 Author: Joao Eduardo Luis <joao@suse.de>
934 Date: Mon Sep 7 14:12:19 2015 +0100
936 mon: MonitorDBStore: make get_next_key() work properly
938 We introduced a significant bug with 2cc7aee, when we fixed issue #11786.
939 Although that patch would fix the problem described in #11786, we
940 managed to not increment the iterator upon returning the current key.
941 This would have the iterator iterating over the same key, forever and
944 Signed-off-by: Joao Eduardo Luis <joao@suse.de>
945 (cherry picked from commit 70d31082fd3dc8c7857994104577f1a3631c678c)
947 commit 0a740d7b2b62e55056fb96d6423203c47e0f3cf6
948 Merge: aeea956 37d16a9
949 Author: Nathan Cutler <presnypreklad@gmail.com>
950 Date: Wed Sep 9 15:45:37 2015 +0200
952 Merge pull request #5388 from SUSE/wip-12490-firefly
954 buffer: critical bufferlist::zero bug
956 Reviewed-by: Gregory Farnum <gfarnum@redhat.com>
958 commit aeea9567e8b047b3a67badfcc4e43c880f0efa46
959 Merge: 973b74d 6b42b06
960 Author: Nathan Cutler <presnypreklad@gmail.com>
961 Date: Wed Sep 9 06:10:09 2015 +0200
963 Merge pull request #5408 from SUSE/wip-12492-firefly
965 the output is wrong when runing ceph osd reweight
967 Reviewed-by: Kefu Chai <kchai@redhat.com>
969 commit 973b74d5b266f4e9e7f992a85b22f42c9fb543b1
970 Merge: 79403ba 5445461
971 Author: Nathan Cutler <presnypreklad@gmail.com>
972 Date: Wed Sep 9 06:07:51 2015 +0200
974 Merge pull request #5404 from SUSE/wip-12395-firefly
976 register_new_pgs() should check ruleno instead of its index
978 Reviewed-by: Kefu Chai <kchai@redhat.com>
980 commit 79403ba8ce526d65521e6cd2ca2bc06f5fe261b9
981 Merge: db44bb2 dc12875
982 Author: Nathan Cutler <presnypreklad@gmail.com>
983 Date: Wed Sep 9 06:05:39 2015 +0200
985 Merge pull request #5199 from SUSE/wip-11980-firefly
987 MDSMonitor: handle MDSBeacon messages properly
989 Reviewed-by: Kefu Chai <kchai@redhat.com>
991 commit db44bb29b8b036642c50f5d322b5edd9d51699bb
992 Merge: fc699f4 31ff7d1
993 Author: Nathan Cutler <presnypreklad@gmail.com>
994 Date: Wed Sep 9 06:04:15 2015 +0200
996 Merge pull request #5410 from SUSE/wip-12497-firefly
998 get pools health'info have error
1000 Reviewed-by: Kefu Chai <kchai@redhat.com>
1002 commit fc699f42459ab3723f3e63b28a4aa5f481c78fc4
1003 Merge: 6762295 ea0d67c
1004 Author: Nathan Cutler <presnypreklad@gmail.com>
1005 Date: Wed Sep 9 06:01:50 2015 +0200
1007 Merge pull request #5409 from SUSE/wip-12495-firefly
1009 pgmonitor: wrong at/near target max“ reporting
1011 Reviewed-by: Kefu Chai <kchai@redhat.com>
1013 commit 67622955b1cfd0b97ac8d3d96b223995440af0b1
1014 Merge: c4fa5ef 2f35a41
1015 Author: Nathan Cutler <presnypreklad@gmail.com>
1016 Date: Mon Sep 7 14:22:52 2015 +0200
1018 Merge pull request #5358 from ceph/wip-11470.firefly
1020 mon: PaxosService: call post_refresh() instead of post_paxos_update()
1022 Reviewed-by: Joao Eduardo Luis <joao@suse.de>
1024 commit 3be204f6a22be109d2aa8cfd5cee09ec3381d9b2
1025 Author: Dmitry Yatsushkevich <dyatsushkevich@mirantis.com>
1026 Date: Tue Mar 17 15:44:52 2015 -0700
1028 rgw: init script waits until the radosgw stops
1031 Init script waits in stop action until the radowgw daemons stop.
1033 Signed-off-by: Dmitry Yatsushkevich <dyatsushkevich@mirantis.com>
1034 (cherry picked from commit 1cca0c1e91a1e18f82f4d22855e96b2fc947f5ea)
1036 commit c4fa5ef989247404ded7adf5dad4e5316b8af04a
1037 Merge: beff99b c6f3780
1038 Author: Nathan Cutler <presnypreklad@gmail.com>
1039 Date: Mon Sep 7 10:20:27 2015 +0200
1041 Merge pull request #5403 from SUSE/wip-12393-firefly
1043 Memory leak in Mutex.cc, pthread_mutexattr_init without pthread_mutexattr_destroy
1045 Reviewed-by: Kefu Chai <kchai@redhat.com>
1046 Reviewed-by: Nathan Cutler <ncutler@suse.com>
1048 commit beff99bc7358ee0625079ca60adbf109986f2285
1049 Merge: 4a3c178 68508e2
1050 Author: Nathan Cutler <presnypreklad@gmail.com>
1051 Date: Sun Sep 6 21:19:55 2015 +0200
1053 Merge pull request #5225 from SUSE/wip-12266-firefly
1055 ceph.spec.in: 50-rbd.rules conditional is wrong
1057 Reviewed-by: Loic Dachary <ldachary@redhat.com>
1059 commit 4a3c17834890b69e0467b65f83b1979328f3d27c
1060 Merge: 6571cd9 2544a6f
1061 Author: Nathan Cutler <presnypreklad@gmail.com>
1062 Date: Sun Sep 6 21:19:42 2015 +0200
1064 Merge pull request #5217 from SUSE/wip-12268-firefly
1066 ceph-common needs python-argparse on older distros, but doesn't require it
1068 Reviewed-by: Loic Dachary <ldachary@redhat.com>
1070 commit bff2f477c4ad86b4bd6e3ca3e637a6168c5c8053
1071 Author: Nathan Cutler <ncutler@suse.com>
1072 Date: Sun Sep 6 16:30:35 2015 +0200
1074 OSD: add scrub_finalize_wq suicide timeout
1076 Signed-off-by: Nathan Cutler <ncutler@suse.com>
1078 commit 91d4c217e32b8b76fcac49f37879a3f78088694d
1079 Author: Samuel Just <sjust@redhat.com>
1080 Date: Wed May 6 10:52:40 2015 -0700
1082 OSD: add scrub_wq suicide timeout
1084 Signed-off-by: Samuel Just <sjust@redhat.com>
1085 (cherry picked from commit 547a7041edc833f3cc8e04d388574809e30a8af6)
1089 firefly still has scrub_finalize_wq
1091 commit 7f6ec65b7c2ca0174142c1c48f18998d8c586b02
1092 Author: Samuel Just <sjust@redhat.com>
1093 Date: Wed May 6 11:02:19 2015 -0700
1095 OSD: add op_wq suicide timeout
1097 Signed-off-by: Samuel Just <sjust@redhat.com>
1098 (cherry picked from commit 3f2946aa9f9b4b432e8c018283503153a2b1eddc)
1102 adapt op_wq() and peering_wq() calls in OSD::OSD
1104 adapt OpWQ instead of ShardedOpWQ
1106 commit 9ce8ce689009cf8ef749edd320d1c2a73ecc2f90
1107 Author: Samuel Just <sjust@redhat.com>
1108 Date: Wed May 6 10:54:31 2015 -0700
1110 OSD: add remove_wq suicide timeout
1112 Signed-off-by: Samuel Just <sjust@redhat.com>
1113 (cherry picked from commit f2fbfa32a16666be46359f0eab7b04ca80a753f5)
1117 No next_removal_seq(0) in master
1119 commit 6926a64fbd4718b8a5df8e04545bc93c4981e413
1120 Author: Samuel Just <sjust@redhat.com>
1121 Date: Wed May 6 10:51:28 2015 -0700
1123 OSD: add snap_trim_wq suicide timeout
1125 Signed-off-by: Samuel Just <sjust@redhat.com>
1126 (cherry picked from commit e1073a4a577211672148a4112bd633831552d66f)
1128 commit d31d1f6f5b08362779fa6af72690e898d2407b90
1129 Author: Samuel Just <sjust@redhat.com>
1130 Date: Wed May 6 10:50:19 2015 -0700
1132 OSD: add recovery_wq suicide timeout
1134 Signed-off-by: Samuel Just <sjust@redhat.com>
1135 (cherry picked from commit 85311b656852af75bfbbc6699f92fc6aa233c316)
1138 src/common/config_opts.h
1139 place OPTION(osd_command_thread_suicide_timeout, ...)
1140 directly after OPTION(osd_command_thread_timeout, ...)
1142 commit f85ec2a52e969f9a7927d0cfacda6a1cc6f2898c
1143 Author: Samuel Just <sjust@redhat.com>
1144 Date: Wed May 6 10:49:00 2015 -0700
1146 OSD: add command_wq suicide timeout
1148 Signed-off-by: Samuel Just <sjust@redhat.com>
1149 (cherry picked from commit df4e5de819c30003cfbe50a071c49039cf534419)
1152 src/common/config_opts.h
1153 place OPTION(osd_command_thread_suicide_timeout, ...)
1154 directly after OPTION(osd_command_thread_timeout, ...)
1156 commit b894b368790de3383295372250888ba674502fb1
1157 Author: Samuel Just <sjust@redhat.com>
1158 Date: Tue Jul 7 08:49:54 2015 -0700
1160 WBThrottle::clear_object: signal if we cleared an object
1163 Signed-off-by: Samuel Just <sjust@redhat.com>
1164 (cherry picked from commit 8447b08d60d5cf7dce20f8b839cc2cb5a93782de)
1166 commit cd1396cd62c79b177e46cfb57ab6b3b6fdcd227b
1167 Author: Sage Weil <sage@redhat.com>
1168 Date: Sun Aug 9 10:46:10 2015 -0400
1170 osd/PGLog: dirty_to is inclusive
1172 There are only two callers of mark_dirty_to who do not pass max,
1173 and they are both in the merge_log extending tail path. In that
1174 case, we want to include the last version specified in the log
1175 writeout. Fix the tail extending code to always specify the
1176 last entry added, inclusive.
1179 Signed-off-by: Sage Weil <sage@redhat.com>
1180 (cherry picked from commit f0ca14df0641daa04eee39d98d8bd0faf46e4e6d)
1182 commit f39c7917a39e445efa8d73178657fc5960772275
1183 Author: Nathan Cutler <ncutler@suse.com>
1184 Date: Fri Sep 4 17:38:32 2015 +0200
1186 osd: Keep a reference count on Connection while calling send_message()
1190 Signed-off-by: David Zafman <dzafman@redhat.com>
1191 (manual backport of commit a140085)
1193 Conflicts: src/osd/OSD.cc
1194 master has share_map_peer; firefly has osd->_share_map_outgoing
1196 commit 6571cd9a9616637bf32f9ac97a9bd864c0bff856
1197 Merge: 8edc091 bd25136
1198 Author: Ken Dreyer <kdreyer@redhat.com>
1199 Date: Fri Sep 4 11:50:22 2015 -0600
1201 Merge pull request #5224 from SUSE/wip-12304-firefly
1203 ceph.spec.in running fdupes unnecessarily
1205 Reviewed-by: Ken Dreyer <kdreyer@redhat.com>
1207 commit 8edc091df90be6935fcfe59c18d7060c88501556
1208 Merge: 6053bff 601eadb
1209 Author: Ken Dreyer <kdreyer@redhat.com>
1210 Date: Fri Sep 4 11:47:49 2015 -0600
1212 Merge pull request #5394 from SUSE/wip-12447-firefly
1214 ceph.spec.in: useless %py_requires breaks SLE11-SP3 build
1216 Reviewed-by: Ken Dreyer <kdreyer@redhat.com>
1218 commit 2c2ffa1d6d1112dbf52cbbe36f4a5376e17da56a
1219 Author: Josh Durgin <jdurgin@redhat.com>
1220 Date: Mon Aug 24 15:40:39 2015 -0700
1222 config: skip lockdep for intentionally recursive md_config_t lock
1224 lockdep can't handle recursive locks, resulting in false positive
1225 reports for certain set_val_or_die() calls, like via
1226 md_config_t::parse_argv() passed "-m".
1229 Signed-off-by: Josh Durgin <jdurgin@redhat.com>
1230 (cherry picked from commit 415226995ebe841cc9e8639a3233150a75aecaed)
1232 commit 6053bffa8955953cf40a05f2b22684e77b6003a3
1233 Merge: 49d416e 5559a5f
1234 Author: Loic Dachary <loic@dachary.org>
1235 Date: Fri Sep 4 17:18:30 2015 +0200
1237 Merge pull request #5043 from SUSE/wip-12007-firefly
1239 Objecter: pg listing can deadlock when throttling is in use
1241 Reviewed-by: Yehuda Sadeh <ysadehwe@redhat.com>
1242 Reviewed-by: Loic Dachary <ldachary@redhat.com>
1244 commit 2c0d7feeb1c7592887e0408fe4fadaa9b4f659e9
1245 Author: Loic Dachary <ldachary@redhat.com>
1246 Date: Thu Aug 13 19:41:47 2015 +0200
1248 tests: tiering agent and proxy read
1250 Verify that an object promoted to a cache tier because of a proxy read
1251 is evicted as expected.
1253 http://tracker.ceph.com/issues/12673 Refs: #12673
1255 Signed-off-by: Loic Dachary <ldachary@redhat.com>
1256 (cherry picked from commit 7924231930732bd297d3bd034c8295e96cb81088)
1259 qa/workunits/cephtool/test.sh
1260 resolved by manually adding the new test to TESTS
1262 commit aa911767d9326c8aa37671883892b7d383596960
1263 Author: Loic Dachary <ldachary@redhat.com>
1264 Date: Thu Aug 13 13:47:24 2015 +0200
1266 osd: trigger the cache agent after a promotion
1268 When a proxy read happens, the object promotion is done in parallel. The
1269 agent_choose_mode function must be called to reconsider the situation
1270 to protect against the following scenario:
1273 * agent_choose_mode finds no object exists and the agent
1275 * object promotion happens
1276 * the agent does not reconsider and eviction does not happen
1279 http://tracker.ceph.com/issues/12673 Fixes: #12673
1281 Signed-off-by: Loic Dachary <ldachary@redhat.com>
1282 (cherry picked from commit e1f58feb9b1d20b72f2eb2eefdea5982e0cddccd)
1284 commit 49d416e6e2e2f3e1f317978371089f9d92358642
1285 Merge: 7744518 f93eadd
1286 Author: Loic Dachary <loic@dachary.org>
1287 Date: Fri Sep 4 17:08:41 2015 +0200
1289 Merge pull request #4769 from SUSE/wip-11741-firefly
1291 crush: take crashes due to invalid arg
1293 Reviewed-by: Loic Dachary <ldachary@redhat.com>
1295 commit 774451810ff055984b79e1d1875bd6a3468dfb7a
1296 Merge: fc27b0d f122788
1297 Author: Nathan Cutler <presnypreklad@gmail.com>
1298 Date: Fri Sep 4 16:50:45 2015 +0200
1300 Merge pull request #4788 from SUSE/wip-11801-firefly
1302 Make it more difficult to delete pools in firefly
1304 Reviewed-by: Sage Weil <sage@redhat.com>
1306 commit fc27b0dbe074dc15450d7669ea722aa3b605b3b2
1307 Merge: 055624b da78889
1308 Author: Loic Dachary <loic@dachary.org>
1309 Date: Fri Sep 4 16:44:57 2015 +0200
1311 Merge pull request #5235 from SUSE/wip-12310-firefly
1313 read on chunk-aligned xattr not handled
1315 Reviewed-by: Loic Dachary <ldachary@redhat.com>
1317 commit 055624bf9a6480fffefb3c7cf5c3a4d4d73985c1
1318 Merge: 045c056 46e5dea
1319 Author: Loic Dachary <loic@dachary.org>
1320 Date: Fri Sep 4 16:42:28 2015 +0200
1322 Merge pull request #5389 from SUSE/wip-12391-firefly
1324 EC pools are not allowed as cache pools, disallow in the mon
1326 Reviewed-by: Loic Dachary <ldachary@redhat.com>
1328 commit 045c0562f97e7d2c68278fbddb9ca1707bb8d5fd
1329 Merge: 99612e7 b8e3f6e
1330 Author: Sage Weil <sage@redhat.com>
1331 Date: Fri Sep 4 09:42:03 2015 -0400
1333 Merge pull request #5406 from ceph/wip-12465-firefly
1335 log: take mutex while opening fd
1337 Reviewed-by: Sage Weil <sage@redhat.com>
1339 commit 99612e7ed9573f2af36604f7d258535f6feadff3
1340 Merge: 7f8a397 4a709a4
1341 Author: Nathan Cutler <presnypreklad@gmail.com>
1342 Date: Thu Sep 3 12:34:32 2015 +0200
1344 Merge pull request #4854 from ceph/wip-11769-firefly
1346 librbd: aio calls may block
1348 Reviewed-by: Josh Durgin <jdurgin@redhat.com>
1350 commit 7f8a397e38101017ef9783d8ed08bb393a8d8825
1351 Merge: ea601ef f0d2429
1352 Author: Loic Dachary <loic@dachary.org>
1353 Date: Wed Sep 2 07:55:15 2015 +0200
1355 Merge pull request #5233 from SUSE/wip-12074-firefly
1357 RGW logfile does not get chowned properly
1359 Reviewed-by: Loic Dachary <ldachary@redhat.com>
1361 commit 644f213c672f6fe2786e041043fdd55f8399871e (refs/remotes/gh/wip-11455-firefly)
1362 Author: Yehuda Sadeh <yehuda@redhat.com>
1363 Date: Wed Aug 26 14:34:30 2015 -0700
1365 rgw: init some manifest fields when handling explicit objs
1368 When dealing with old manifest that has explicit objs, we also
1369 need to set the head size and head object correctly so that
1370 code that relies on this info doesn't break.
1372 Signed-off-by: Yehuda Sadeh <yehuda@redhat.com>
1373 (cherry picked from commit d7bf8cb594e5276d1c80544f5ec954d52b159750)
1375 commit e7931a73df1ab77feb1c2ece13e3de3989ef7a0e
1376 Author: Yehuda Sadeh <yehuda@redhat.com>
1377 Date: Thu Jul 30 14:47:15 2015 -0700
1379 rgw: url encode exposed bucket
1382 Don't send the bucket name back without url encoding it.
1384 Signed-off-by: Yehuda Sadeh <yehuda@redhat.com>
1385 (cherry picked from commit ad5507fe0bf72ed5bdf8353e315cc9092c740144)
1387 commit 558c52955d464827630e0aa2fed970df987bb036
1388 Author: Wido den Hollander <wido@42on.com>
1389 Date: Wed Jun 3 13:13:33 2015 +0200
1391 rgw: Do not enclose the Bucket header in quotes
1394 Signed-off-by: Wido den Hollander <wido@42on.com>
1395 (cherry picked from commit 8af25faed93fe02d3dad585b8579ce8b041cc4e6)
1397 commit d3c94698e4e852bef3e65fbf439f5f209fbc0b25 (refs/remotes/gh/wip-12752)
1398 Author: Jason Dillaman <dillaman@redhat.com>
1399 Date: Fri Aug 21 11:32:39 2015 -0400
1401 Objecter: pg_interval_t::is_new_interval needs pgid from previous pool
1403 When increasing the pg_num of a pool, an assert would fail since the
1404 calculated pgid seed would be for the pool's new pg_num value instead
1405 of the previous pg_num value.
1408 Backport: infernalis, hammer, firefly
1409 Signed-off-by: Jason Dillaman <dillaman@redhat.com>
1410 (cherry picked from commit f20f7a23e913d09cc7fc22fb3df07f9938ddc144)
1412 Conflicts: (hobject_t sort order not backported, trivial resolution)
1413 src/osdc/Objecter.cc
1416 commit 56d267b7ae02070a7d7ed247990b84124fd62411
1417 Author: Samuel Just <sjust@redhat.com>
1418 Date: Tue May 26 12:35:11 2015 -0700
1420 osd_types::is_new_interval: size change triggers new interval
1423 Signed-off-by: Samuel Just <sjust@redhat.com>
1424 (cherry picked from commit ff79959c037a7145f7104b06d9e6a64492fdb95f)
1426 commit 411769c1461c11611b479bd826c72c56b3ce47c5
1427 Author: Kefu Chai <kchai@redhat.com>
1428 Date: Mon Aug 10 04:25:03 2015 -0700
1430 mon: add a cache layer over MonitorDBStore
1432 the cache of of leveldb does not perform well under some condition,
1433 so we need a cache in our own stack.
1435 * add an option "mon_osd_cache_size" to control the size of cache size
1439 Signed-off-by: Kefu Chai <kchai@redhat.com>
1440 (cherry picked from commit 5e99a578050976ca22b549812ac80d494fe7041d)
1443 src/mon/OSDMonitor.h
1444 minor differences in OSDMonitor ctor
1446 commit 278d732ecd3594cd76e172d78ce3ec84e58e178b (refs/remotes/me/wip-journal-firefly)
1447 Author: Sage Weil <sage@redhat.com>
1448 Date: Thu Jan 15 11:20:18 2015 -0800
1450 os/FileJournal: Fix journal write fail, align for direct io
1452 when config journal_zero_on_create true, osd mkfs will fail when zeroing journal.
1453 journal open with O_DIRECT, buf should align with blocksize.
1455 Backport: giant, firefly, dumpling
1456 Signed-off-by: Xie Rui <875016668@qq.com>
1457 Reviewed-by: Sage Weil <sage@redhat.com>
1458 (cherry picked from commit 80473f63853593e53d72f17c098152caf17f5e9e)
1460 commit d0d6727762ebda858065101635935df3d44a18ad
1461 Author: Kefu Chai <kchai@redhat.com>
1462 Date: Wed Jul 29 15:52:19 2015 +0800
1464 common/syncfs: fall back to sync(2) if syncfs(2) not available
1467 Signed-off-by: Kefu Chai <kchai@redhat.com>
1468 (cherry picked from commit 08210d6a85527b0759f6dc4411f742bc13a4939f)
1470 commit 40494c6e479c2ec4dfe5f6c2d6aef3b6fa841620
1471 Author: Danny Al-Gaaf <danny.al-gaaf@bisect.de>
1472 Date: Thu Mar 12 17:38:27 2015 +0100
1474 sync_filesystem.h: fix unreachable code
1478 CID 1264460 (#1 of 1): Structurally dead code (UNREACHABLE)
1479 unreachable: This code cannot be reached: sync();
1481 Signed-off-by: Danny Al-Gaaf <danny.al-gaaf@bisect.de>
1482 (cherry picked from commit 9921836e982818b92e20280eecb74e701523045c)
1484 commit 3dbbc86ad6d1e7131bbe49a4eff1557d7da9822f
1485 Author: Jianpeng Ma <jianpeng.ma@intel.com>
1486 Date: Thu Jan 8 10:29:37 2015 +0800
1488 mon, os: check the result of sync_filesystem.
1490 Signed-off-by: Jianpeng Ma <jianpeng.ma@intel.com>
1491 (cherry picked from commit 27cb78bd3b2dde40c70de245769d1c0ec252533b)
1493 commit 7fa6fdc6c5e52f11456e4bea4ae32fd62248c80b
1494 Author: Jianpeng Ma <jianpeng.ma@intel.com>
1495 Date: Wed Jan 7 15:28:37 2015 +0800
1497 common: Don't call ioctl(BTRFS_IOC_SYNC) in sync_filesystem.
1499 From the manual of syncfs, it first appeared in Linux 2.6.39. At this
1500 point, btrfs didn't as a stable production. So remove this.
1502 Signed-off-by: Jianpeng Ma <jianpeng.ma@intel.com>
1503 (cherry picked from commit 397b261630dccc656f6eca509f4c70173c8d8d8f)
1505 commit 9f15ed5bb5a837727cb3bef70508e056c125a518
1506 Author: Jianpeng Ma <jianpeng.ma@intel.com>
1507 Date: Wed Jan 7 15:26:08 2015 +0800
1509 common: Directly return the result of syncfs().
1511 In commit 808c644248e4, it will try sync() if syncfs() return error.
1512 No evidence prove this way can work. And sync() don't return result
1513 so make this function always return zero which cause filestore omit the
1516 Signed-off-by: Jianpeng Ma <jianpeng.ma@intel.com>
1517 (cherry picked from commit 3f7faa4d8d3af3d2e3889c23290b016f27d36386)
1519 commit ea601efc6c693ba8041509c98756fa8c49b9c13a
1520 Merge: 50fa963 a128ce3
1521 Author: Nathan Cutler <presnypreklad@gmail.com>
1522 Date: Wed Aug 12 09:43:19 2015 +0200
1524 Merge pull request #5390 from SUSE/wip-12503-firefly
1526 rgw/logrotate.conf calls service with wrong init script name
1528 Reviewed-by: Ken Dreyer <kdreyer@redhat.com>
1530 commit 50fa9636d0c5e8946fa410891623e2c8363e2067
1531 Merge: abe6c03 d4ef556
1532 Author: Yehuda Sadeh <yehuda@redhat.com>
1533 Date: Tue Aug 11 11:27:16 2015 -0700
1535 Merge pull request #4765 from SUSE/wip-11721-firefly
1537 rgw: Keystone PKI token expiration is not enforced
1539 commit 04b2a878b1329202758465cf8e9b0f874cbeeef5
1540 Author: Dan van der Ster <daniel.vanderster@cern.ch>
1541 Date: Tue Nov 18 15:51:46 2014 +0100
1543 ceph-disk: don't change the journal partition uuid
1545 We observe that the new /dev/disk/by-partuuid/<journal_uuid>
1546 symlink is not always created by udev when reusing a journal
1547 partition. Fix by not changing the uuid of a journal partition
1548 in this case -- instead we can reuse the existing uuid (and
1549 journal_symlink) instead. We also now assert that the symlink
1550 exists before further preparing the OSD.
1553 Signed-off-by: Dan van der Ster <daniel.vanderster@cern.ch>
1554 Tested-by: Dan van der Ster <daniel.vanderster@cern.ch>
1555 (cherry picked from commit 29eb1350b4acaeabfe1d2b19efedbce22641d8cc)
1557 commit e57e6f5da10a62f2f4d7b1a6a734a095ed494ebe
1558 Author: Dan van der Ster <daniel.vanderster@cern.ch>
1559 Date: Mon Sep 29 13:20:10 2014 +0200
1561 ceph-disk: set guid if reusing a journal partition
1563 When reusing a journal partition (e.g. /dev/sda2) we should set a
1564 new partition guid and link it correctly with the OSD. This way
1565 the journal is symlinked by its persistent name and ceph-disk list
1568 Signed-off-by: Dan van der Ster <daniel.vanderster@cern.ch>
1569 (cherry picked from commit beff616f506b96eb52285f0d2e268e10f8edaa2c)
1571 commit ebba1d59b5e4bc11cbdfcda4e480639f7d9e1498
1572 Author: John Spray <john.spray@redhat.com>
1573 Date: Thu Jul 30 14:20:56 2015 +0100
1575 osd/OSDMap: handle incrementals that modify+del pool
1577 Because new_pools was processed after old_pools, if something
1578 was modified and then deleted in the same incremental
1579 map, then the resulting state would be a pool in the map
1580 that had no entry in pool_name.
1583 Signed-off-by: John Spray <john.spray@redhat.com>
1584 (cherry picked from commit 8d4932e721e32a33c1bada54e11d4ecccd868618)
1586 commit 31ff7d14e801ef7b7dff3b050094f1eceacf75dd
1587 Author: renhwztetecs <rhwlyw@163.com>
1588 Date: Tue Jul 21 18:55:00 2015 +0800
1590 Update OSDMonitor.cc
1592 OSDMonitor::get_pools_health(), s/objects/bytes/
1595 Signed-off-by: ren.huanwen@zte.com.cn
1596 (cherry picked from commit 7fc13c9d6b9a4962d7640240416105d8f558d600)
1598 commit ea0d67c482659dfb8c08055a77f400d0689866e0
1599 Author: huangjun <hjwsm1989@gmail.com>
1600 Date: Mon Jul 20 10:50:20 2015 +0800
1602 mon/PGMonitor: avoid uint64_t overflow when checking pool 'target/max' status.
1605 Signed-off-by: huangjun <hjwsm1989@gmail.com>
1607 (cherry picked from commit f84e6b8157e78fd05cd1c8ac957c92df17e6ba06)
1609 commit 6b42b06239426e5ec54dfaa82f48abca56f824ca
1610 Author: Joao Eduardo Luis <joao@suse.de>
1611 Date: Tue Jul 21 11:20:53 2015 +0100
1613 mon: OSDMonitor: fix hex output on 'osd reweight'
1617 Signed-off-by: Joao Eduardo Luis <joao@suse.de>
1618 (cherry picked from commit 787fa80c2746fde44ac0583ff7995ec5be9a672d)
1620 commit 5445461ba08fa5b2e12a951a1432916224d3f668
1621 Author: Xinze Chi <xmdxcxz@gmail.com>
1622 Date: Fri Jul 3 18:27:13 2015 +0800
1624 mon/PGMonitor: bug fix pg monitor get crush rule
1626 when some rules have been deleted before, the index in array of crush->rules
1627 is not always equals to crush_ruleset of pool.
1630 Reported-by: Ning Yao <zay11022@gmail.com>
1631 Signed-off-by: Xinze Chi <xmdxcxz@gmail.com>
1632 (cherry picked from commit 498793393c81c0a8e37911237969fba495a3a183)
1634 commit c6f37807d04af66fbed24e75776aed1b44241150
1635 Author: Ketor Meng <d.ketor@gmail.com>
1636 Date: Tue May 26 18:50:17 2015 +0800
1638 Mutex: fix leak of pthread_mutexattr
1640 Need pthread_mutexattr_destroy after pthread_mutexattr_init
1643 Signed-off-by: Ketor Meng <d.ketor@gmail.com>
1644 (cherry picked from commit 2b23327b3aa8d96341d501a5555195ca1bc0de8f)
1646 commit 601eadbf707597d9a1572efbdf86b22ff6813602
1647 Author: Nathan Cutler <ncutler@suse.com>
1648 Date: Thu Jul 9 11:51:13 2015 +0200
1650 ceph.spec.in: drop SUSE-specific %py_requires macro
1652 %py_requires expands to
1654 BuildRequires: /usr/bin/python
1655 PreReq: python = 2.7
1657 The BuildRequires: is already provided, and the PreReq is wrong because
1658 e.g. SLE11-SP3 (a platform we are trying to support) has Python 2.6.
1660 http://tracker.ceph.com/issues/12351 Fixes: #12351
1662 Signed-off-by: Nathan Cutler <ncutler@suse.com>
1663 (cherry picked from commit cbf6c7b46b42e163950bf7956e5b7565fca961c9)
1667 in master, the %py_requires is in the python-rados package,
1668 while in firefly it is in the python-ceph package
1670 commit a128ce37038ef6e44c8d7bd082ba196583974ef7
1671 Author: wuxingyi <wuxingyi2015@outlook.com>
1672 Date: Wed Mar 11 17:34:40 2015 +0800
1674 rgw/logrotate.conf: Rename service name
1676 The service name for ceph rados gateway was changed to "ceph-radosgw",
1677 the previous version of service name "radosgw" would cause a failed reload,
1678 and finally make it impossible to write any log data to the log file.
1680 Signed-off-by: wuxingyi <wuxingyi2015@outlook.com>
1681 (cherry picked from commit 9df3f798179481fe8ae6ae873dcb793de7d8f367)
1683 commit 46e5deabebeeb047783cfc980aa28a3df46eeeca
1684 Author: Samuel Just <sjust@redhat.com>
1685 Date: Fri May 15 13:05:40 2015 -0700
1687 OSDMonitor: disallow ec pools as tiers
1690 Signed-off-by: Samuel Just <sjust@redhat.com>
1691 (cherry picked from commit 11b7801bb57cb25cd2d26d58722d49691747725b)
1694 qa/workunits/cephtool/test.sh
1695 no "# make sure we can't clobber snapshot state" tests in firefly
1696 src/mon/OSDMonitor.cc
1697 no tp->removed_snaps.empty() in firefly
1699 commit 37d16a9e572580eeae86a2bae6d4ddd0299fb833
1700 Author: Haomai Wang <haomaiwang@gmail.com>
1701 Date: Thu Jul 9 13:32:03 2015 +0800
1703 UnittestBuffer: Add bufferlist zero test case
1705 Signed-off-by: Haomai Wang <haomaiwang@gmail.com>
1706 (cherry picked from commit 577acf665948e531e50d36f5780ea19351a5bf65)
1708 commit 4443acdbef1148e0261bce25f7d7a3433e09cecc
1709 Author: Haomai Wang <haomaiwang@gmail.com>
1710 Date: Thu Jul 9 13:42:42 2015 +0800
1712 buffer: Fix bufferlist::zero bug with special case
1715 Signed-off-by: Haomai Wang <haomaiwang@gmail.com>
1716 (cherry picked from commit 43f583d160ccaf879eaf0f3020e77860cf8d1df0)
1718 commit d52187019d321fe8a2dc54fe8a67a5139c310db1
1719 Author: Joao Eduardo Luis <joao@suse.de>
1720 Date: Wed May 27 22:28:49 2015 +0100
1722 mon: MonitorDBStore: get_next_key() only if prefix matches
1724 get_next_key() had a bug in which we would always return the first key
1725 from the iterator, regardless of whether its prefix had been specified
1730 Signed-off-by: Joao Eduardo Luis <joao@suse.de>
1731 (cherry picked from commit 2cc7aee1abe40453093306c8fef2312b650dff5d)
1733 commit 2f35a415dcd31683151defb0ddb1a6d5696c7051
1734 Author: Joao Eduardo Luis <joao@suse.de>
1735 Date: Fri Jun 12 19:21:10 2015 +0100
1737 mon: PaxosService: call post_refresh() instead of post_paxos_update()
1739 Whenever the monitor finishes committing a proposal, we call
1740 Monitor::refresh_from_paxos() to nudge the services to refresh. Once
1741 all services have refreshed, we would then call each services
1742 post_paxos_update().
1744 However, due to an unfortunate, non-critical bug, some services (mainly
1745 the LogMonitor) could have messages pending in their
1746 'waiting_for_finished_proposal' callback queue [1], and we need to nudge
1749 This patch adds a new step during the refresh phase: instead of calling
1750 directly the service's post_paxos_update(), we introduce a
1751 PaxosService::post_refresh() which will call the services
1752 post_paxos_update() function first and then nudge those callbacks when
1755 [1] - Given the monitor will send MLog messages to itself, and given the
1756 service is not readable before its initial state is proposed and
1757 committed, some of the initial MLog's would be stuck waiting for the
1758 proposal to finish. However, by design, we only nudge those message's
1759 callbacks when an election finishes or, if the leader, when the proposal
1760 finishes. On peons, however, we would only nudge those callbacks if an
1761 election happened to be triggered, hence the need for an alternate path
1762 to retry any message waiting for the initial proposal to finish.
1766 Signed-off-by: Joao Eduardo Luis <joao@suse.de>
1767 (cherry picked from commit 1551ebb63238073d2fd30201e6b656a8988e958c)
1769 commit b8e3f6e190809febf80af66415862e7c7e415214
1770 Author: Samuel Just <sjust@redhat.com>
1771 Date: Fri Jul 24 15:38:18 2015 -0700
1773 Log::reopen_log_file: take m_flush_mutex
1775 Otherwise, _flush() might continue to write to m_fd after it's closed.
1776 This might cause log data to go to a data object if the filestore then
1777 reuses the fd during that time.
1780 Backport: firefly, hammer
1781 Signed-off-by: Samuel Just <sjust@redhat.com>
1782 (cherry picked from commit 8778ab3a1ced7fab07662248af0c773df759653d)
1784 - modified to drop m_flush_mutex_holder, which isn't present in firefly
1786 commit abe6c03d411f5f1d277f910ae87123309fa95d4a
1787 Merge: 5df7273 9614e4d
1788 Author: Nathan Cutler <presnypreklad@gmail.com>
1789 Date: Sat Jul 25 13:51:43 2015 +0200
1791 Merge pull request #4583 from SUSE/wip-11542-firefly
1793 rgw: Change variable length array of std::strings (not legal in C++) to std::vector<std::string>
1795 Reviewed-by: Sage Weil <sage@redhat.com>
1797 commit 5df727379b63abb25421a32c95fdda9a0b62eeac
1798 Merge: 2a77154 8c573c8
1799 Author: Nathan Cutler <presnypreklad@gmail.com>
1800 Date: Sat Jul 25 13:50:06 2015 +0200
1802 Merge pull request #4584 from SUSE/wip-8911-firefly
1804 rgw: doesn't return 'x-timestamp' in header which is used by 'View Details' of OpenStack
1806 Reviewed-by: Sage Weil <sage@redhat.com>
1808 commit 2a77154b4abcd4eebc8d1a315f7beeeabaf2213b
1809 Merge: 64c3a64 d78e31f
1810 Author: Nathan Cutler <presnypreklad@gmail.com>
1811 Date: Sat Jul 25 13:48:31 2015 +0200
1813 Merge pull request #4642 from SUSE/wip-11323-firefly
1815 rgw: quota not respected in POST object
1817 Reviewed-by: Yehuda Sadeh <yehuda@redhat.com>
1819 commit 64c3a645c4dba1846532997758c5d7c3a42760b2
1820 Merge: f11df3c 177ef55
1821 Author: Nathan Cutler <presnypreklad@gmail.com>
1822 Date: Sat Jul 25 13:24:39 2015 +0200
1824 Merge pull request #4641 from SUSE/wip-11149-firefly
1826 rgw: delete bucket does not remove .bucket.meta file
1828 Reviewed-by: Yahuda Sadeh <yehuda@redhat.com>
1830 commit f11df3c7819a5f8dd5336be7dbe43e0f21dbb394
1831 Merge: 090bd12 2ee6e9a
1832 Author: Nathan Cutler <presnypreklad@gmail.com>
1833 Date: Sat Jul 25 13:21:31 2015 +0200
1835 Merge pull request #5170 from SUSE/wip-12039-firefly
1837 rgw: Swift API: response for PUT on /container does not contain the mandatory Content-Length header when FCGI is used
1839 Reviewed-by: Yahuda Sadeh <yehuda@redhat.com>
1841 commit 090bd126e15acacc46bce3271d2aa0cfa889b499
1842 Merge: 94060b7 2d5d8da
1843 Author: Nathan Cutler <presnypreklad@gmail.com>
1844 Date: Sat Jul 25 13:20:15 2015 +0200
1846 Merge pull request #4762 from SUSE/wip-11754-firefly
1848 rgw: Object copy bug
1850 Reviewed-by: Yahuda Sadeh <yehuda@redhat.com>
1852 commit 94060b78a2e9b38422a951664dd4f37a6f52316f
1853 Merge: b9f45fe 3223bd6
1854 Author: Josh Durgin <jdurgin@redhat.com>
1855 Date: Fri Jul 24 11:42:58 2015 -0700
1857 Merge pull request #5171 from SUSE/wip-12240-firefly
1859 librbd: internal.cc: 1967: FAILED assert(watchers.size() == 1)
1861 Reviewed-by: Josh Durgin <jdurgin@redhat.com>
1863 commit b9f45fe7816da1ce1fcbfd2a60fe370abe0f1848
1864 Merge: 57bbc8e 9845258
1865 Author: Josh Durgin <jdurgin@redhat.com>
1866 Date: Fri Jul 24 11:26:23 2015 -0700
1868 Merge pull request #4639 from SUSE/wip-7385-firefly
1870 librbd: Objectcacher setting max object counts too low
1872 Reviewed-by: Josh Durgin <jdurgin@redhat.com>
1874 commit 8c02376bd58d463f742966b67fa075a59b5f4269
1875 Author: Nathan Cutler <ncutler@suse.com>
1876 Date: Fri Jul 24 18:50:03 2015 +0200
1878 TestPGLog: fix invalid proc_replica_log test case
1880 If 1'3 is non-divergent, no osd could correctly have created 2'3. Also,
1881 proc_replica_log does not add the extra entries from the auth log to the
1882 missing set, that happens later on in activate.
1884 Signed-off-by: Samuel Just <sjust@redhat.com>
1885 (cherry picked from commit 1333b676dba4ed4b5db854fe935bdadbf9f21d08)
1888 src/test/osd/TestPGLog.cc
1889 "sjustm: set_hash() method just does hash = and then rebuilds a cache,
1890 but the cache does not exist in firefly so hash = will do fine"
1892 Signed-off-by: Nathan Cutler <ncutler@suse.com>
1894 commit 4a709a450987cc90126b11a20c1c762d4fbfcea9
1895 Author: Jason Dillaman <dillaman@redhat.com>
1896 Date: Tue Jun 2 10:33:35 2015 -0400
1898 tests: verify librbd blocking aio code path
1900 Signed-off-by: Jason Dillaman <dillaman@redhat.com>
1901 (cherry picked from commit 4cf41486e9c9e1efcb863960a8f3e0326ffca7e5)
1903 commit 4501e90d9b667310a222c3616a7b1f1fb382f997
1904 Author: Jason Dillaman <dillaman@redhat.com>
1905 Date: Mon Jun 1 22:56:11 2015 -0400
1907 librbd: new rbd_non_blocking_aio config option
1909 Setting this option to false reverts librbd to legacy behavior
1910 where AIO operations could potentially block.
1912 Signed-off-by: Jason Dillaman <dillaman@redhat.com>
1913 (cherry picked from commit 769cad12716b85d87eacc1069dd9f5c21cad3915)
1916 src/common/config_opts.h: trivial resolution
1917 src/librbd/librbd.cc: trivial resolution
1919 commit 68863968dd734fc45557dca100c8a0a5060cbd8b
1920 Author: Jason Dillaman <dillaman@redhat.com>
1921 Date: Thu Apr 9 20:34:28 2015 -0400
1923 PendingReleaseNotes: document changes to librbd's aio_read methods
1925 Signed-off-by: Jason Dillaman <dillaman@redhat.com>
1927 commit 861c20bb2cc05a13a45b0e2368c591212e4eda72
1928 Author: Jason Dillaman <dillaman@redhat.com>
1929 Date: Wed Apr 8 21:55:36 2015 -0400
1931 tests: update librbd AIO tests to remove result code
1933 Signed-off-by: Jason Dillaman <dillaman@redhat.com>
1936 src/test/librbd/test_librbd.cc: trivial resolution
1938 commit 4e7cf6115798dd339efe41a528aac148be640b29
1939 Author: Jason Dillaman <dillaman@redhat.com>
1940 Date: Thu Apr 9 13:33:09 2015 -0400
1942 librbd: AioRequest::send no longer returns a result
1944 The librados calls used by AioRequest::send should always return
1945 zero unless there is a bug.
1947 Signed-off-by: Jason Dillaman <dillaman@redhat.com>
1948 (cherry picked from commit c77bce3311ab62892eb8c1d883263ba7ed663b20)
1951 src/librbd/AioRequest.cc: trivial resolution
1952 src/librbd/AioRequest.h: trivial resolution
1953 src/librbd/internal.cc: trivial resolution
1955 commit 0ccf47bf9f070ff7fe4e1dd3d3c3e3daa1f621c4
1956 Author: Jason Dillaman <dillaman@redhat.com>
1957 Date: Wed Apr 8 21:37:50 2015 -0400
1959 librbd: internal AIO methods no longer return result
1961 All failures should be returned via the AioCompletion.
1963 Signed-off-by: Jason Dillaman <dillaman@redhat.com>
1964 (cherry picked from commit 9ab42d613128ab08c688ddbea93df4c95068b9cd)
1967 src/librbd/AioRequest.cc: trivial resolution
1968 src/librbd/internal.cc: trivial resolution
1969 src/librbd/internal.h: trivial resolution
1971 commit e48040413849c5d3a577addff644502858ac5a66
1972 Author: Jason Dillaman <dillaman@redhat.com>
1973 Date: Wed Apr 8 21:48:21 2015 -0400
1975 Throttle: added pending_error method to SimpleThrottle
1977 Allow the client of SimpleThrottle to detect an async error
1978 so that it can exit early.
1980 Signed-off-by: Jason Dillaman <dillaman@redhat.com>
1981 (cherry picked from commit b88b88c5df91325fb713c2031a56bffe421268e0)
1983 commit 8ec341ee8f256e8ab91a931f6005e1ebd4b1db60
1984 Author: Jason Dillaman <dillaman@redhat.com>
1985 Date: Wed Apr 8 20:18:50 2015 -0400
1987 librbd: add new fail method to AioCompletion
1989 Helper method to handle passing fatal errors generated within
1990 librbd (not from the OSDs) back to the client.
1992 Signed-off-by: Jason Dillaman <dillaman@redhat.com>
1993 (cherry picked from commit 6d1d0c867855a96bee4c13a0c0a39a0e002ccd12)
1996 src/librbd/AioCompletion.cc: trivial resolution
1997 src/librbd/AioCompletion.h: trivial resolution
1999 commit eed8ce258d116e837d8777f246f31c21a03ceb74
2000 Author: Jason Dillaman <dillaman@redhat.com>
2001 Date: Wed Apr 8 19:06:52 2015 -0400
2003 librbd: avoid blocking AIO API methods
2005 Enqueue all AIO API methods within the new librbd thread pool to
2006 reduce the possibility of any blocking operations. To maintain
2007 backwards compatibility with the legacy return codes of the API's
2008 AIO methods, it's still possible to block attempting to acquire
2012 Signed-off-by: Jason Dillaman <dillaman@redhat.com>
2013 (cherry picked from commit 3a7b5e30efdb21aa1a0aeb68a5d02a1ac2a5faf3)
2016 src/librbd/librbd.cc: trivial resolution
2018 commit 220a23e8d99882838626d5647f9c055eefeb28c6
2019 Author: Jason Dillaman <dillaman@redhat.com>
2020 Date: Wed Apr 8 17:24:08 2015 -0400
2022 librbd: add task pool / work queue for AIO requests
2024 Signed-off-by: Jason Dillaman <dillaman@redhat.com>
2025 (cherry picked from commit afb896d91f886b647baf38f7ec94cc3739f6d2a9)
2028 src/common/config_opts.h: trivial resolution
2029 src/librbd/ImageCtx.cc: trivial resolution
2030 src/librbd/ImageCtx.h: trivial resolution
2031 src/librbd/internal.cc: trivial resolution
2033 commit 3292bd26127f5beb69ac9c5fbcdea7efe002fd40
2034 Author: Jason Dillaman <dillaman@redhat.com>
2035 Date: Mon May 11 17:05:49 2015 -0400
2037 WorkQueue: added virtual destructor
2039 Signed-off-by: Jason Dillaman <dillaman@redhat.com>
2040 (cherry picked from commit b3f5a75332c058816dc39b71e9d2b36e752159f4)
2042 commit 9bf970c1d956c94bc420d874eb83db1f16cf5d44
2043 Author: Jason Dillaman <dillaman@redhat.com>
2044 Date: Wed Apr 8 16:46:34 2015 -0400
2046 WorkQueue: add new ContextWQ work queue
2048 The queue holds a collection of Context pointers that will
2049 be completed by the thread pool.
2051 Signed-off-by: Jason Dillaman <dillaman@redhat.com>
2052 (cherry picked from commit 24a33e977f7b71962adeeb48f75d488a76e70fa9)
2055 src/common/WorkQueue.h: trivial resolution
2057 commit 94aa947b33083bcfe3cb61f41bfdaea7df605cbb
2058 Author: Sage Weil <sage@redhat.com>
2059 Date: Fri Dec 5 14:21:08 2014 -0800
2061 common/ceph_context: don't import std namespace
2063 This was broken by 7fed5dee4f96a83d1d6914f6fc0895bba2d15b99
2065 Signed-off-by: Sage Weil <sage@redhat.com>
2066 (cherry picked from commit 9029813b2784fe71879b1b7684855415b4c7566b)
2068 commit b91e594401ab75987a370f6dee0ee8d212189c9f
2069 Author: Haomai Wang <haomaiwang@gmail.com>
2070 Date: Mon Dec 1 23:54:16 2014 +0800
2072 CephContext: Add AssociatedSingletonObject to allow CephContext's singleton
2074 If some objects associated to CephContext want to create a singleton object,
2075 it can inherit AssociatedSingletonObject and implement destruction to get notified.
2077 Signed-off-by: Haomai Wang <haomaiwang@gmail.com>
2078 (cherry picked from commit 7fed5dee4f96a83d1d6914f6fc0895bba2d15b99)
2080 commit f122788b3592da516cdb9f6abdc3383a2e7424a2
2081 Author: Sage Weil <sage@redhat.com>
2082 Date: Fri Jan 16 07:54:22 2015 -0800
2084 mon/OSDMonitor: require mon_allow_pool_delete = true to remove pools
2086 This is a simple safety check. Since we default to true it is currently
2089 Backport: giant, firefly
2090 Signed-off-by: Sage Weil <sage@redhat.com>
2091 (cherry picked from commit 6b982e4cc00f9f201d7fbffa0282f8f3295f2309)
2094 src/test/mon/osd-pool-create.sh
2096 Changed $CEPH_MON to 127.0.0.1 -- the CEPH_MON was introduced after
2097 firefly to allow tests to run in parallel. Back in firefly all tests
2098 use the same port because 127.0.0.1 was hardcoded. We can't
2099 conveniently backport all that's necessary for tests to run in
2100 parallel, therefore we keep the 127.0.0.1 hardcoded.
2103 src/test/mon/osd-pool-create.sh
2104 TEST_no_pool_delete() follows a different test than in master
2106 commit 57bbc8e630e6611d1ab74b9756c96daf0e91be6b
2107 Merge: c0a601f 3f2fb8a
2108 Author: Loic Dachary <loic@dachary.org>
2109 Date: Fri Jul 24 10:49:50 2015 +0200
2111 Merge pull request #5051 from SUSE/wip-12016-firefly
2113 osd: cache full mode still skips young objects
2115 Reviewed-by: Sage Weil <sage@redhat.com>
2117 commit c0a601ff5be45c0630a8659994107672558a982a
2118 Merge: 71174a5 ef648e5
2119 Author: Sage Weil <sage@redhat.com>
2120 Date: Wed Jul 22 11:36:47 2015 -0400
2122 Merge pull request #5307 from dachary/wip-release-notes-firefly
2124 backport v0.80.10 release notes to firefly
2126 commit 5559a5fe0f593c0e7af47c6d8a8359dbcbfd1f08
2127 Author: Guang Yang <yguang@yahoo-inc.com>
2128 Date: Mon Sep 15 11:41:06 2014 +0000
2130 For pgls OP, get/put budget on per list session basis, instead of per OP basis, which could lead to deadlock.
2132 Signed-off-by: Guang Yang (yguang@yahoo-inc.com)
2133 (cherry picked from commit 0f884fdb31a26f241401ce2a9329dc0f2c4eccc7)
2136 src/librados/IoCtxImpl.cc
2137 In firefly, return value of objecter->pg_read() is not assigned to c->tid.
2138 src/osdc/Objecter.cc
2140 There is no _op_submit_with_budget() function in firefly.
2141 There is no Objecter::_finish_op() function in firefly.
2142 In firefly, _take_op_budget() is called take_op_budget().
2144 commit 71174a53c9a3a2b54b947da7e1425ff36fd44e37
2145 Merge: f7a085f 989e047
2146 Author: Loic Dachary <loic@dachary.org>
2147 Date: Wed Jul 22 10:20:32 2015 +0200
2149 Merge pull request #4867 from SUSE/wip-11878-firefly
2151 mon: Clock skew causes missing summary and confuses Calamari
2153 Reviewed-by: Joao Eduardo Luis <joao@suse.de>
2155 commit 989e04799be567ee93369a87b90405833c3fe8d2
2156 Author: Thorsten Behrens <tbehrens@suse.com>
2157 Date: Fri Feb 6 01:26:40 2015 +0100
2159 Always provide summary for non-healthy cluster.
2161 This fixes a problem, wherein calamari does not provide
2162 popup drill-downs for warnings or errors, should the summary
2165 Calamari gets health info from /api/v1/cluster/$FSID/health.
2166 If the data here has a summary field, this summary is provided
2169 /api/v1/cluster/$FSID/health is populated (ultimately) with
2170 status obtained via librados python bindings from the ceph
2171 cluster. In the case where there's clock skew, the summary
2172 field supplied by the ceph cluster is empty.
2174 No summary field, no popup window with more health details.
2176 Signed-off-by: Thorsten Behrens <tbehrens@suse.com>
2177 (cherry picked from commit eaf6e0cf48488fe604d0ef0db164d44948d4e8d4)
2183 commit f7a085f04c544895bab3d79569276574a96636eb
2184 Merge: 2f1dd10 d296120
2185 Author: Josh Durgin <jdurgin@redhat.com>
2186 Date: Tue Jul 21 13:31:33 2015 -0700
2188 Merge pull request #5062 from SUSE/wip-11699-firefly
2190 objecter: resend linger ops on split
2192 Reviewed-by: Josh Durgin <jdurgin@redhat.com>
2194 commit ef648e592404e8677df91f218c4400caad4034b6
2195 Author: Sage Weil <sage@redhat.com>
2196 Date: Tue Jul 21 12:54:19 2015 -0400
2198 doc: v0.80.10 changelog
2200 Signed-off-by: Sage Weil <sage@redhat.com>
2201 (cherry picked from commit b3a3e0e509eb9530994723ed9d68f85b8434a22d)
2203 commit df837e5cc30cf08ade26df6e42ca402fc628a9bb
2204 Author: Sage Weil <sage@redhat.com>
2205 Date: Tue Jul 21 10:55:18 2015 -0400
2207 doc/release-notes: v0.80.10
2209 Signed-off-by: Sage Weil <sage@redhat.com>
2210 (cherry picked from commit 2a61b611987d169d5a501d77fbd545068550c28c)
2213 doc/release-notes.rst
2215 commit 90cddd425ca8f1731fff0b8fb4b7e02a96048f92
2216 Author: Loic Dachary <ldachary@redhat.com>
2217 Date: Tue Jul 14 16:15:44 2015 +0200
2219 doc/release-notes: update notes for v0.80.10
2221 Signed-off-by: Loic Dachary <loic@dachary.org>
2222 (cherry picked from commit 1231ae09d675f2ee9e76a04b91f771d73a2fb346)
2225 doc/release-notes.rst
2227 commit 2f1dd10b71d29a647aa2bab4a51fecc0501d6741
2228 Merge: 0caf6b3 1d2af55
2229 Author: Loic Dachary <loic@dachary.org>
2230 Date: Tue Jul 21 13:03:07 2015 +0200
2232 Merge pull request #5129 from SUSE/wip-12202-firefly
2234 mon: ceph fails to compile with boost 1.58
2236 Reviewed-by: Loic Dachary <ldachary@redhat.com>
2238 commit 0caf6b32849bd673d83e96f0eae19bb6d7e423ae
2239 Merge: 3010f23 6793d50
2240 Author: Loic Dachary <loic@dachary.org>
2241 Date: Tue Jul 21 12:51:10 2015 +0200
2243 Merge pull request #5044 from SUSE/wip-12011-firefly
2245 osd: PG stuck with remapped
2247 Reviewed-by: Loic Dachary <ldachary@redhat.com>
2249 commit 3010f23fae10d37ff78b8d7d5049ccba798bbe6b
2250 Merge: ad58a32 fcac7ab
2251 Author: Loic Dachary <loic@dachary.org>
2252 Date: Tue Jul 21 12:45:22 2015 +0200
2254 Merge pull request #5039 from SUSE/wip-11961-firefly
2256 osd: Cleanup boost optionals for boost 1.56
2258 Reviewed-by: Loic Dachary <ldachary@redhat.com>
2260 commit ad58a32292eee39905ae7bacc8fc52a52c2d08fc
2261 Merge: 13946e9 a62c720
2262 Author: Loic Dachary <loic@dachary.org>
2263 Date: Tue Jul 21 12:39:34 2015 +0200
2265 Merge pull request #5037 from SUSE/wip-11875-firefly
2267 tools: ceph-post-file fails on rhel7
2269 Reviewed-by: Loic Dachary <ldachary@redhat.com>
2271 commit 13946e9c18bbeb209458a3faac6e687ff380ac6e
2272 Merge: bde35c1 90a1a75
2273 Author: Loic Dachary <loic@dachary.org>
2274 Date: Tue Jul 21 12:34:51 2015 +0200
2276 Merge pull request #4771 from SUSE/wip-11759-firefly
2278 tools: ceph-disk: support NVMe device partitions
2280 Reviewed-by: Loic Dachary <ldachary@redhat.com>
2282 commit bde35c11acd34c76d704b41292d167c3fc4cea53
2283 Merge: 9e11564 10d85e1
2284 Author: Loic Dachary <loic@dachary.org>
2285 Date: Tue Jul 21 12:22:52 2015 +0200
2287 Merge pull request #4636 from SUSE/wip-11143-firefly
2289 tools: src/ceph-disk : disk zap sgdisk invocation
2291 Reviewed-by: Loic Dachary <ldachary@redhat.com>
2293 commit 9e11564b36a42b605f256943c7236b91a736fceb
2294 Merge: 8638443 424289d
2295 Author: Loic Dachary <loic@dachary.org>
2296 Date: Tue Jul 21 12:20:50 2015 +0200
2298 Merge pull request #4635 from SUSE/wip-7387-firefly
2300 common: Malformed JSON command output when non-ASCII strings are present
2302 Reviewed-by: Loic Dachary <ldachary@redhat.com>
2304 commit 8638443eac92ac80b4c57271f7350d94e445abf5
2305 Merge: 4c37d3d 8372f1d
2306 Author: Loic Dachary <loic@dachary.org>
2307 Date: Tue Jul 21 12:17:43 2015 +0200
2309 Merge pull request #4632 from SUSE/wip-11507-firefly
2311 osd: object creation by write cannot use an offset on an erasure coded pool
2313 Reviewed-by: Loic Dachary <ldachary@redhat.com>
2315 commit 4c37d3d99f660c835d2f73e8f08b6ef170c0c7b7
2316 Merge: 8f11ab7 9090967
2317 Author: Loic Dachary <loic@dachary.org>
2318 Date: Tue Jul 21 12:12:36 2015 +0200
2320 Merge pull request #4630 from SUSE/wip-10983-firefly
2322 tools: use a new disk as journal disk,ceph-disk prepare fail
2324 Reviewed-by: Loic Dachary <ldachary@redhat.com>
2326 commit 8f11ab7ce27cab43145a2821f3996d3103cf7870
2327 Merge: 64a21b7 b1ff739
2328 Author: Loic Dachary <loic@dachary.org>
2329 Date: Tue Jul 21 12:08:26 2015 +0200
2331 Merge pull request #4597 from xinxinsh/wip-11026-firefly
2333 objecter: can get stuck in redirect loop if osdmap epoch == last_force_op_resend
2335 Reviewed-by: Loic Dachary <ldachary@redhat.com>
2337 commit 64a21b71526c51f8e787b365039de2b0cb0a285b
2338 Merge: 82e8359 2cb2333
2339 Author: Loic Dachary <loic@dachary.org>
2340 Date: Tue Jul 21 12:04:39 2015 +0200
2342 Merge pull request #4582 from SUSE/wip-11543-firefly
2344 tools: ceph_argparse_flag has no regular 3rd parameter
2346 Reviewed-by: Loic Dachary <ldachary@redhat.com>
2348 commit bba50ce8f227af29d559b486274871bb3999fb24
2349 Author: Samuel Just <sjust@redhat.com>
2350 Date: Thu Apr 9 16:28:23 2015 -0700
2352 TestPGLog: fix noop log proc_replica_log test case
2354 Need to fill in log.head, olog.head.
2356 Signed-off-by: Samuel Just <sjust@redhat.com>
2357 (cherry picked from commit 1e5b22053381580708af29a1536f1e106c9b479c)
2359 commit fdff8ce6c996cda7b3966d20c24b20ff545e468a
2360 Author: Samuel Just <sjust@redhat.com>
2361 Date: Thu Apr 9 15:01:41 2015 -0700
2363 TestPGLog: add test for 11358
2365 Signed-off-by: Samuel Just <sjust@redhat.com>
2366 (cherry picked from commit b61e5aef15d7448d24b713d13dce77697bc0af74)
2369 src/test/osd/TestPGLog.cc
2370 leave out all the other tests that have been
2371 added to master since firefly
2373 commit 65028b6304235ba5fa54d14805028db1a032e5a0
2374 Author: Samuel Just <sjust@redhat.com>
2375 Date: Thu Apr 9 15:01:29 2015 -0700
2377 PGLog::proc_replica_log: handle split out overlapping entries
2379 See the comment, if the overlapping entries between olog.head
2380 and log.tail were split out, last_update should be log.tail.
2383 Signed-off-by: Samuel Just <sjust@redhat.com>
2384 (cherry picked from commit 6561e0d955c4563a8505b717de41f5bd89119f63)
2386 commit 82e8359783f056cce5f7ede2dabd0cc61d278eab
2387 Merge: 741f0c2 b061140
2388 Author: Josh Durgin <jdurgin@redhat.com>
2389 Date: Fri Jul 17 16:10:43 2015 -0700
2391 Merge pull request #4633 from SUSE/wip-11535-firefly
2393 Client admin socket leaks file descriptors
2395 Reviewed-by: Josh Durgin <jdurgin@redhat.com>
2397 commit da78889d76f641dfcbb80164be32b85e880bb97f
2398 Author: Sage Weil <sage@redhat.com>
2399 Date: Wed Mar 18 13:49:20 2015 -0700
2401 os/chain_xattr: handle read on chnk-aligned xattr
2403 If we wrote an xattr that was a multiple of a chunk, we will try to read
2404 the next chunk and get ENODATA. If that happens bail out of the loop and
2405 assume we've read the whole thing.
2407 Backport: hammer, firefly
2408 Signed-off-by: Sage Weil <sage@redhat.com>
2409 (cherry picked from commit 8614dcebf384b74b283cd352001a65fa26ba925c)
2412 src/test/objectstore/chain_xattr.cc
2413 added #define FILENAME
2415 commit 4e7de5b5f0e32d1183e2a0490d65e4e01490d942
2416 Author: Yehuda Sadeh <yehuda@redhat.com>
2417 Date: Thu Jun 25 14:31:03 2015 -0700
2419 rgw: error out if frontend did not send all data
2422 The civetweb mg_write() doesn't return error when it can't flush all data
2423 to the user, it just sends the total number of bytes written. Modified the
2424 client io to return total number of bytes and return an error if didn't
2427 Signed-off-by: Yehuda Sadeh <yehuda@redhat.com>
2428 (cherry picked from commit daa679c3dd3770a6d5421e2cc9a36924f4550439)
2430 commit f0d2429f764aaa756fc8002309545b02dc45c199
2431 Author: Thorsten Behrens <tbehrens@suse.com>
2432 Date: Wed Dec 10 11:53:43 2014 +0100
2434 Unconditionally chown rados log file.
2436 This fixes bnc#905047 (in a somewhat ad-hoc way). Sadly the log
2437 file gets created from several places, so its existence does not
2438 mean init-radosgw had actually run.
2440 Signed-off-by: Thorsten Behrens <tbehrens@suse.com>
2441 (cherry picked from commit 41611ea27a42d6221c418613d24dce34d988f2e6)
2443 commit 68508e28adffae64a52ea4c0046fbe8537e0d093
2444 Author: Nathan Cutler <ncutler@suse.com>
2445 Date: Thu Jun 25 22:37:52 2015 +0200
2447 ceph.spec.in: use _udevrulesdir to eliminate conditionals
2449 The conditionals governing where 50-rbd.rules is installed were not doing the
2450 right thing on SUSE distros.
2452 Start using the %_udevrulesdir RPM macro, while taking care that it is defined
2453 and set to the right value. Use it to eliminate some conditionals around other
2454 udev rules files as well.
2456 http://tracker.ceph.com/issues/12166 Fixes: #12166
2458 Signed-off-by: Nathan Cutler <ncutler@suse.com>
2459 (cherry picked from commit 8aa758ee7a14e93ded1d55b4aca7a54aa47d7174)
2463 master spec file has diverged significantly from firefly
2465 commit bd25136d45d32b8d419fbe60e39572038e33cc59
2466 Author: Nathan Cutler <ncutler@suse.com>
2467 Date: Mon Jul 13 18:12:01 2015 +0200
2469 ceph.spec.in: do not run fdupes, even on SLE/openSUSE
2471 In openSUSE there is a policy to use %fdupes in the spec file if RPMLINT
2472 complains about duplicate files wasting space in the filesystem.
2474 However, RPMLINT is not so complaining, so drop fdupes.
2476 http://tracker.ceph.com/issues/12301 Fixes: #12301
2478 Signed-off-by: Nathan Cutler <ncutler@suse.com>
2479 (cherry picked from commit 53072b9019caf72e0313b2804ea174237ed7da33)
2481 commit 2544a6fc59a86d3f02dcd67d3c84a72d7a1100a0
2482 Author: Nathan Cutler <ncutler@suse.com>
2483 Date: Tue Jun 16 18:27:20 2015 +0200
2485 ceph.spec.in: python-argparse only in Python 2.6
2487 argparse is a widely-used Python module for parsing command-line arguments.
2488 Ceph makes heavy use of Python scripts, both in the build environment and on
2489 cluster nodes and clients.
2491 Until Python 2.6, argparse was distributed separately from Python proper.
2492 As of 2.7 it is part of the Python standard library.
2494 Although the python package in a given distro may or may not Provide:
2495 python-argparse, this cannot be relied upon.
2497 Therefore, this commit puts appropriate conditionals around Requires:
2498 python-argparse and BuildRequires: python-argparse. It does so for Red
2499 Hat/CentOS and SUSE only, because the last Fedora version with Python 2.6
2500 was Fedora 13, which is EOL.
2502 argparse is required by both the ceph and ceph-common packages, but since ceph
2503 requires ceph-common, the argparse Requires and BuildRequires need only appear
2504 once, under ceph-common.
2506 http://tracker.ceph.com/issues/12034 Fixes: #12034
2508 Signed-off-by: Nathan Cutler <ncutler@suse.com>
2509 (cherry picked from commit 23171c952997d104cfa6b5700ec0bb658a8e0915)
2514 commit 54264210f4ebec23b08dd6712e09aea49543b52b
2515 Author: Samuel Just <sam.just@inktank.com>
2516 Date: Tue Aug 12 15:24:26 2014 -0700
2518 ReplicatedPG::maybe_handle_cache: do not skip promote for write_ordered
2520 We cannot redirect a RW ordered read.
2523 Introduced: 0ed3adc1e0a74bf9548d1d956aece11f019afee0
2524 Signed-off-by: Samuel Just <sam.just@inktank.com>
2525 (cherry picked from commit 24aeca9c439474b079d23600d8d42c0af5d9ce5c)
2527 commit 7e2526784203b0f1bce08869aa7b1fda9c5eedd9
2528 Author: Zhiqiang Wang <wonzhq@hotmail.com>
2529 Date: Mon Jul 28 14:06:06 2014 +0800
2531 osd: promotion on 2nd read for cache tiering
2533 http://tracker.ceph.com/issues/8674 Fixes: #8674
2535 Signed-off-by: Zhiqiang Wang <wonzhq@hotmail.com>
2536 (cherry picked from commit 0ed3adc1e0a74bf9548d1d956aece11f019afee0)
2538 commit 66f61cd9ae105948f653fd888812df270ff1e832
2539 Author: Signed-off-by: Zhiqiang Wang <wonzhq@hotmail.com>
2540 Date: Thu Jul 31 15:49:44 2014 -0700
2542 ceph_test_rados_api_tier: test promote-on-second-read behavior
2544 http://tracker.ceph.com/issues/8674 References: #8674
2546 Signed-off-by: Zhiqiang Wang <wonzhq@hotmail.com>
2547 Signed-off-by: Sage Weil <sage@redhat.com>
2548 (cherry picked from commit 1417eded65f40bdb2a49c8252fcfffa383a7b965)
2550 commit a8f3d6e1f1f186cbe2299566a575bf5a40500227
2551 Author: Sage Weil <sage@redhat.com>
2552 Date: Fri Jul 11 11:31:22 2014 -0700
2554 osd/osd_types: be pedantic about encoding last_force_op_resend without feature bit
2556 The addition of the value is completely backward compatible, but if the
2557 mon feature bits don't match it can cause monitor scrub noice (due to the
2558 parallel OSDMap encoding). Avoid that by only adding the new field if the
2559 feature (which was added 2 patches after the encoding, see
2560 3152faf79f498a723ae0fe44301ccb21b15a96ab and
2561 45e79a17a932192995f8328ae9f6e8a2a6348d10.
2565 Signed-off-by: Sage Weil <sage@redhat.com>
2566 (cherry picked from commit c93da05bd3dd6b73e8cfa182c6349359b042717c)
2569 src/osd/osd_types.cc
2570 drop encode_compat variable which is otherwise unused in this context
2572 commit dc128758bb68a50992819f759b5684ff656406ba
2573 Author: Kefu Chai <kchai@redhat.com>
2574 Date: Fri May 15 22:50:36 2015 +0800
2576 mon: always reply mdsbeacon
2578 the MDS (Beacon) is always expecting the reply for the mdsbeacon messages from
2579 the lead mon, and it uses the delay as a metric for the laggy-ness of the
2580 Beacon. when it comes to the MDSMonitor on a peon, it will remove the route
2581 session at seeing a reply (route message) from leader, so a reply to
2582 mdsbeacon will stop the peon from resending the mdsbeacon request to the
2585 if the MDSMonitor re-forwards the unreplied requests after they are
2586 outdated, there are chances that the requests reflecting old and even wrong
2587 state of the MDSs mislead the lead monitor. for example, the MDSs which sent
2588 the outdated messages could be dead.
2591 Signed-off-by: Kefu Chai <kchai@redhat.com>
2592 (cherry picked from commit b3555e9c328633c9e1fbc27d652c004b30535e5b)
2594 commit da720d7d870a53faeacc169796083b4eba4b9406
2595 Author: Kefu Chai <kchai@redhat.com>
2596 Date: Tue Jun 2 23:20:21 2015 -0700
2598 mon/MDSMonitor: rename labels to a better name
2603 Signed-off-by: Kefu Chai <kchai@redhat.com>
2604 (cherry picked from commit f00ecb8b3df73ce6337985bc6d43bce5143ee537)
2607 src/mon/MDSMonitor.cc
2608 Do not compare known daemon health with m->get_health()
2610 commit 329da091d6e8444722b312ff94c7093f81939f13
2611 Author: Kefu Chai <kchai@redhat.com>
2612 Date: Tue Jun 2 12:55:06 2015 +0800
2614 mon: send no_reply() to peon to drop ignored mdsbeacon
2616 so the peon can remove the ignored mdsbeacon request from the
2617 routed_requets at seeing this reply, and hence no longer resend the
2621 Signed-off-by: Kefu Chai <kchai@redhat.com>
2622 (cherry picked from commit 72a37b3a8e145d8522ea67fc14ce2c5510b6852b)
2624 commit 508332f53075a99295b9edb683ed343d8f59ca4f
2625 Author: Kefu Chai <kchai@redhat.com>
2626 Date: Tue Jun 2 12:22:26 2015 +0800
2628 mon: remove unnecessary error handling
2630 msg.get_session() should always return a non-zero pointer in
2633 Signed-off-by: Kefu Chai <kchai@redhat.com>
2634 (cherry picked from commit 16e8e2cc82a90c49cd8aa3d0e3acc4694ba659a0)
2638 Monitor::_ms_dispatch(Message *m) is bool in firefly
2640 commit 741f0c235228b371486574ba0c3fbf55729e03bd
2641 Merge: 1623e6c 0a7f2b2
2642 Author: Samuel Just <sam.just@inktank.com>
2643 Date: Thu Jul 9 17:24:04 2015 -0700
2645 Merge pull request #4631 from SUSE/wip-11069-firefly
2647 PGLog: split divergent priors as well
2649 Reviewed-by: Samuel Just
2651 commit 3223bd604117b0b302b1e3e235ff9bced8e986bc
2652 Author: Jason Dillaman <dillaman@redhat.com>
2653 Date: Fri Jun 26 09:59:36 2015 -0400
2655 librbd: assertion failure race condition if watch disconnected
2657 It's possible for librbd's watch of the header object to be reset by
2658 connection issues just prior to the image being removed. This will
2659 causes an assertion failure which assumes at least one watcher on the
2663 Backport: hammer, firefly
2664 Signed-off-by: Jason Dillaman <dillaman@redhat.com>
2665 (cherry picked from commit af276de4f299960e43761904c043924cec5fef11)
2667 commit 2ee6e9a0cc6513097d6c9f9e40787a1c60c09796
2668 Author: Radoslaw Zarzynski <rzarzynski@mirantis.com>
2669 Date: Sun Mar 8 22:45:34 2015 +0100
2671 rgw: send Content-Length in response for HEAD on Swift account.
2674 Signed-off-by: Radoslaw Zarzynski <rzarzynski@mirantis.com>
2675 (cherry picked from commit 29073d84640b61d0221b2d2ab73c271d60fd13ba)
2677 commit 1b4536aa47fd6d2440fbd30b64b5b46eaa21215a
2678 Author: Radoslaw Zarzynski <rzarzynski@mirantis.com>
2679 Date: Fri Mar 6 13:18:19 2015 +0100
2681 rgw: send Content-Length in response for DELETE on Swift container.
2684 Signed-off-by: Radoslaw Zarzynski <rzarzynski@mirantis.com>
2685 (cherry picked from commit d260a93405a7a34a899f716bd9c4e731baa9ffd0)
2687 commit 177324ced968b254458a450f8e348c96cbd10931
2688 Author: Radoslaw Zarzynski <rzarzynski@mirantis.com>
2689 Date: Fri Mar 6 13:14:26 2015 +0100
2691 rgw: send Content-Length in response for PUT on Swift container.
2695 Signed-off-by: Radoslaw Zarzynski <rzarzynski@mirantis.com>
2696 (cherry picked from commit 10c1f1aa1e09df5edf5d2f9f9b0273ddbcb384fa)
2698 commit 0a580c4a23b985ede28cdfc751eed2a368473e09
2699 Author: Radoslaw Zarzynski <rzarzynski@mirantis.com>
2700 Date: Fri Feb 27 15:23:16 2015 +0100
2702 rgw: send Content-Length in response for GET on Swift container.
2706 Signed-off-by: Radoslaw Zarzynski <rzarzynski@mirantis.com>
2707 (cherry picked from commit 5a64fb5430b6b42174bf53a5910d9435043c1380)
2709 commit 84525806a057054dca295f860e0a31d6936d4985
2710 Author: Radoslaw Zarzynski <rzarzynski@mirantis.com>
2711 Date: Fri Mar 6 12:42:52 2015 +0100
2713 rgw: enable end_header() to handle proposal of Content-Length.
2716 Signed-off-by: Radoslaw Zarzynski <rzarzynski@mirantis.com>
2717 (cherry picked from commit f98fd4dfda7a105d621f99c2b48eb9ab9b45d979)
2719 commit 1d2af55f8c69b4a382eb516d8f90a8316fd3bc02
2720 Author: Kefu Chai <kchai@redhat.com>
2721 Date: Fri May 8 15:21:20 2015 +0800
2723 mon: remove unused variable
2725 * as a side effect, this change silences
2726 http://tracker.ceph.com/issues/11576
2729 Signed-off-by: Kefu Chai <kchai@redhat.com>
2730 (cherry picked from commit e7b196a4a091c0ea258866559ba06e7ed0cc4247)
2732 commit d296120c144a62794483338fd696ca8089778e84
2733 Author: Josh Durgin <josh.durgin@inktank.com>
2734 Date: Mon Oct 20 13:29:13 2014 -0700
2736 Objecter: resend linger ops on any interval change
2738 Watch/notify ops need to be resent after a pg split occurs, as well as
2739 a few other circumstances that the existing objecter checks did not
2742 Refactor the check the OSD uses for this to add a version taking the
2743 more basic types instead of the whole OSD map, and stash the needed
2744 info when an op is sent.
2747 Backport: giant, firefly, dumpling
2748 Signed-off-by: Josh Durgin <josh.durgin@inktank.com>
2749 (cherry picked from commit cb9262abd7fd5f0a9f583bd34e4c425a049e56ce)
2752 src/osd/osd_types.cc
2753 src/osdc/Objecter.cc
2756 commit a44b7b4bd870cc72e50a49900cd81fde6d40845b
2757 Author: Samuel Just <sam.just@inktank.com>
2758 Date: Wed May 7 11:02:16 2014 -0700
2760 osd_types: factor out is_new_interval from check_new_interval
2762 Signed-off-by: Samuel Just <sam.just@inktank.com>
2763 (cherry picked from commit 2ee355113ed4374c664909a8cd7dc5dfdc4c9ed7)
2765 commit 3f2fb8a996820843d0287a9337d38d80c7cb616b
2766 Author: Xinze Chi <xmdxcxz@gmail.com>
2767 Date: Wed Nov 5 01:00:01 2014 +0000
2769 osd: cache pool: flush object ignoring cache min flush age when cache pool is full
2770 Signed-off-by: Xinze Chi <xmdxcxz@gmail.com>
2772 (cherry picked from commit f4ee9496ed15e7734abd659f5a353a3cde0dcb23)
2774 commit 9969ae5710bc66e1e96125f22593ea1e0ba381af
2775 Author: Zhiqiang Wang <wonzhq@hotmail.com>
2776 Date: Fri Aug 1 16:09:50 2014 +0800
2778 osd: add local_mtime to struct object_info_t
2780 This fixes a bug when the time of the OSDs and clients are not
2781 synchronized (especially when client is ahead of OSD), and the cache
2782 tier dirty ratio reaches the threshold, the agent skips the flush work
2783 because it thinks the object is too young.
2785 Signed-off-by: Zhiqiang Wang <wonzhq@hotmail.com>
2786 (cherry picked from commit 13b9dc708426c35436ba58388fc54f04f066f03a)
2788 commit 6793d504b5b622a7a6f69ffb30d6822167b557a3
2789 Author: Guang Yang <yguang@yahoo-inc.com>
2790 Date: Mon Oct 13 04:18:45 2014 +0000
2792 The fix for issue 9614 was not completed, as a result, for those erasure coded PGs with one OSD down, the state was wrongly marked as active+clean+degraded. This patch makes sure the clean flag is not set for such PG.
2793 Signed-off-by: Guang Yang <yguang@yahoo-inc.com>
2795 (cherry picked from commit 35bbb2d6002b1b435ff0b1df653cc7ab271f9258)
2797 commit fcac7ab6305dc3215fdcb46951fa9ee4cbf68ce3
2798 Author: William A. Kennington III <william@wkennington.com>
2799 Date: Sat Sep 20 22:52:31 2014 -0700
2801 osd: Cleanup boost optionals
2803 Signed-off-by: William A. Kennington III <william@wkennington.com>
2804 (cherry picked from commit a53ead14c113047567177630b4906136a2109b65)
2806 commit a62c720dc783b40377dcd34d4e53aa778ac1a84b
2807 Author: Sage Weil <sage@redhat.com>
2808 Date: Mon Jun 1 15:10:14 2015 -0700
2810 Makefile: install ceph-post-file keys with mode 600
2812 Otherwise ssh (may) prompt for a password.
2814 Signed-off-by: Sage Weil <sage@redhat.com>
2815 (cherry picked from commit 106a1c3081d02446aa3d8e13865da0c3393bae90)
2817 commit fc59ca391f639b2ef1b5c9dda202a3d7d718cf25
2818 Author: Joseph McDonald <joseph.mcdonald@alcatel-lucent.com>
2819 Date: Mon Jun 1 15:05:32 2015 -0700
2821 ceph-post-file: improve check for a source install
2823 Signed-off-by: Joseph McDonald <joseph.mcdonald@alcatel-lucent.com>
2824 (cherry picked from commit ee170eadcdcb4b54d36a7d474558484de9d917eb)
2826 commit 50bfb354d0648162f94c2659e8b915d69f27bf73
2827 Author: Joseph McDonald <joseph.mcdonald@alcatel-lucent.com>
2828 Date: Mon Jun 1 15:00:39 2015 -0700
2830 ceph-post-file: behave when sftp doesn't take -i
2833 Signed-off-by: Joseph McDonald <joseph.mcdonald@alcatel-lucent.com>
2834 (cherry picked from commit b84031ed5eaace1222e14d3c4076a3ab1155da96)
2836 commit 1623e6c6517ff6958b343493fce6a8e50a52a042
2837 Merge: ea6c958 254676b
2838 Author: David Zafman <dzafman@redhat.com>
2839 Date: Fri Jun 19 15:57:46 2015 -0700
2841 Merge pull request #4960 from ceph/wip-10911-firefly
2843 osd,mon: explicitly specify OSD features in MOSDBoot
2845 Reviewed-by: David Zafman <dzafman@redhat.com>
2847 commit 254676b63b559fa0117520edb5dcbfe067a76747 (refs/remotes/jashan/wip-10911-firefly)
2848 Author: Sage Weil <sage@redhat.com>
2849 Date: Wed Feb 18 14:53:04 2015 -0800
2851 osd,mon: explicitly specify OSD features in MOSDBoot
2853 We are using the connection features to populate the features field in the
2854 OSDMap, but this is the *intersection* of mon and osd features, not the
2855 osd features. Fix this by explicitly specifying the features in
2859 Backport: giant, firefly
2860 Signed-off-by: Sage Weil <sage@redhat.com>
2861 (cherry picked from commit bee91548b07c6a28314dddc50fef8b6a2677e774)
2865 Minor difference in the MOSDBoot constructor
2867 commit b1ff7396c8e9668c0be1a4023e0ffa292067cf64
2868 Author: xinxin shu <xinxin.shu@intel.com>
2869 Date: Thu May 7 05:49:18 2015 +0800
2871 osdc: add epoch_t last_force_resend in Op/LingerOp.
2873 Using this field record the pg_poo_t::last_force_op_resend to avoid op
2874 endless when osd reply with redirect.
2877 Signed-off-by: Jianpeng Ma <jianpeng.ma@intel.com>
2878 Signed-off-by: Sage Weil <sage@redhat.com>
2879 (cherry picked from commit def4fc4a)
2881 commit 90a1a7587fcb567e2cc6e9188fdd56ec1255a881
2882 Author: islepnev <islepnev@gmail.com>
2883 Date: Fri Apr 17 22:33:01 2015 +0300
2885 ceph-disk: support NVMe device partitions
2887 Linux nvme kernel module v0.9 enumerate devices as following:
2889 /dev/nvme0 - characted revice
2890 /dev/nvme0n1 - whole block device
2891 /dev/nvme0n1p1 - first partition
2892 /dev/nvme0n1p2 - second partition
2894 http://tracker.ceph.com/issues/11612 Fixes: #11612
2896 Signed-off-by: Ilja Slepnev <islepnev@gmail.com>
2897 (cherry picked from commit 9b62cf254d02d30609793be8b1cb8a94f38891f1)
2902 commit f93eadd793f9f4fded30df5589f98ccfc0e1839f
2903 Author: Sage Weil <sage@redhat.com>
2904 Date: Tue May 12 16:37:56 2015 -0700
2906 mon: prevent bucket deletion when referenced by a rule
2908 If a rule references a bucket with 'take', prevent deletion.
2911 Signed-off-by: Sage Weil <sage@redhat.com>
2912 (cherry picked from commit 3d591afef90b0601572c748f13faac029d05f5a0)
2914 commit d4ef5566e6b1b2ac73925660a1d627dbb6cc7632
2915 Author: Anton Aksola <anton.aksola@nebula.fi>
2916 Date: Fri Apr 10 13:25:21 2015 +0300
2918 rgw: always check if token is expired
2922 Currently token expiration is only checked by the token cache. With PKI
2923 tokens no expiration check is done after decoding the token. This causes
2924 PKI tokens to be valid indefinitely. UUID tokens are validated by
2925 keystone after cache miss so they are not affected by this bug.
2927 This commit adds explicit token expiration check to
2928 RGWSwift::validate_keystone_token()
2930 Signed-off-by: Anton Aksola <anton.aksola@nebula.fi>
2931 Reported-by: Riku Lehto <riku.lehto@nexetic.com>
2932 (cherry picked from commit 2df069390ea3bbcfbab5022750e89f51d197cc11)
2934 commit 2d5d8da35f627840ee57a8fa071147943b890d0c
2935 Author: Javier M. Mellid <jmunhoz@igalia.com>
2936 Date: Fri May 15 14:22:29 2015 +0200
2938 rgw: Use attrs from source bucket on copy
2940 On copy objects, when bucket source is the same as the destination, use attrs
2945 Signed-off-by: Javier M. Mellid <jmunhoz@igalia.com>
2946 (cherry picked from commit 1dac80df1d4a2364154ed8b404d13609936c257b)
2948 commit d78e31f005756700e15d26d187c6cef6d5d24a7a
2949 Author: 9seconds <nineseconds@yandex.ru>
2950 Date: Wed Apr 1 09:12:06 2015 +0300
2952 rgw: quota not respected in POST object
2954 Signed-off-by: Sergey Arkhipov <nineseconds@yandex.ru>
2955 Backport: hammer, firefly
2957 (cherry picked from commit e76f84e179d2ba8bfc0dc5abf4e620fef14bc8a0)
2961 discard the whitespace modification hunks that were creating
2962 conflicts and ignore the conflict due to an unrelated cast
2963 modification in the context
2965 commit 177ef55d61c9b08783ae746aa521d2535324d6ba
2966 Author: Orit Wasserman <owasserm@redhat.com>
2967 Date: Fri Apr 3 13:43:31 2015 +0200
2969 rgw: remove meta file after deleting bucket
2970 The meta file is deleted only if the bucket meta data is not synced
2972 Signed-off-by: Orit Wasserman <owasserm@redhat.com>
2975 Backport: hammer, firefly
2976 (cherry picked from commit dfdc7afb59cc8e32cf8bff55faa09076c853de06)
2978 commit 9845258f184c823d1e48da1dcb97bdde97220bb6
2979 Author: Jason Dillaman <dillaman@redhat.com>
2980 Date: Wed Apr 22 11:27:35 2015 -0400
2982 librbd: updated cache max objects calculation
2984 The previous calculation was based upon the image's object size.
2985 Since the cache stores smaller bufferheads, the object size is not
2986 a good indicator of cache usage and was resulting in objects being
2987 evicted from the cache too often. Instead, base the max number of
2988 objects on the memory load required to store the extra metadata
2992 Backport: firefly, hammer
2993 Signed-off-by: Jason Dillaman <dillaman@redhat.com>
2994 (cherry picked from commit 0b378942c4f1b79cb65967f2d3466728ca1c8d5b)
2997 src/librbd/ImageCtx.cc
2998 because firefly has cct->_conf->rbd_cache_size
2999 instead of cache_size
3001 commit 0a7f2b2a859aefebc995ad5c7b699eb117d6fa12
3002 Author: Nathan Cutler <ncutler@suse.cz>
3003 Date: Sun May 10 17:13:22 2015 +0200
3005 PGLog: split divergent priors as well
3007 Signed-off-by: Samuel Just <sjust@redhat.com>
3008 (cherry picked from commit bbe231a61fc06a1a668e6b831fb8224526721f93)
3013 The cherry-picked commit did not compile as-is, because the hobject_t
3014 class in firefly lacks a get_hash() method, which was added in 6de83d4.
3015 To get the patch to compile, I replaced i->second.get_hash() with
3018 commit 424289dc4ee8fb09dd00d507b5a9d1fecc786937
3019 Author: Kefu Chai <kchai@redhat.com>
3020 Date: Fri May 8 12:01:16 2015 +0800
3022 json_sprit: fix the FTBFS on old gcc
3025 Signed-off-by: Kefu Chai <kchai@redhat.com>
3026 (cherry picked from commit 6b68b27146852f057a3373cd04b08cd6917f3eea)
3028 commit 84b00f189554fc37413c990ac4011079bb5cdb60
3029 Author: Tim Serong <tserong@suse.com>
3030 Date: Sat May 2 01:59:53 2015 +1000
3032 json_spirit: use utf8 intenally when parsing \uHHHH
3034 When the python CLI is given non-ASCII characters, it converts them to
3035 \uHHHH escapes in JSON. json_spirit parses these internally into 16 bit
3036 characters, which could only work if json_spirit were built to use
3037 std::wstring, which it isn't; it's using std::string, so the high byte
3038 ends up being zero'd, leaving the low byte which is effectively garbage.
3040 This hack^H^H^H^H change makes json_spirit convert to utf8 internally
3041 instead, which can be stored just fine inside a std::string.
3043 Note that this implementation still assumes \uHHHH escapes are four hex
3044 digits, so it'll only cope with characters in the Basic Multilingual
3045 Plane. Still, that's rather a lot more characters than it could cope
3048 (For characters outside the BMP, Python seems to generate escapes in the
3049 form \uHHHHHHHH, i.e. 8 hex digits, which the current implementation
3050 doesn't expect to see)
3054 Signed-off-by: Tim Serong <tserong@suse.com>
3055 (cherry picked from commit 8add15b86e7aaef41397ab8fa9e77ee7957eb607)
3058 src/test/mon/osd-pool-create.sh
3060 Changed $CEPH_MON to 127.0.0.1 -- the CEPH_MON was introduced after
3061 firefly to allow tests to run in parallel. Back in firefly all tests
3062 use the same port because 127.0.0.1 was hardcoded. We can't
3063 conveniently backport all that's necessary for tests to run in
3064 parallel, therefore we keep the 127.0.0.1 hardcoded.
3066 commit 10d85e1126768f6c72f7d2a301ea7ca9ba59f8e0
3067 Author: Owen Synge <osynge@suse.com>
3068 Date: Tue Mar 17 15:41:33 2015 +0100
3070 Fix "disk zap" sgdisk invocation
3074 If the metadata on the disk is truly invalid, sgdisk would fail to zero
3075 it in one go, because --mbrtogpt apparently tried to operate on the
3076 metadata it read before executing --zap-all.
3078 Splitting this up into two separate invocations to first zap everything
3079 and then clear it properly fixes this issue.
3081 Based on patch by Lars Marowsky-Bree <lmb@suse.com> in ceph-deploy.
3082 Created by Vincent Untz <vuntz@suse.com>
3084 Signed-off-by: Owen Synge <osynge@suse.com>
3085 Signed-off-by: Thorsten Behrens <tbehrens@suse.com>
3086 (cherry picked from commit fdd7f8d83afa25c4e09aaedd90ab93f3b64a677b)
3088 commit b06114088a354f9dbcfc9182643733399c31f2a8
3089 Author: Jon Bernard <jbernard@tuxion.com>
3090 Date: Fri May 8 11:54:06 2015 -0400
3092 common/admin_socket: close socket descriptor in destructor
3094 Long-running processes that do not reuse a single client connection will
3095 see accumulating file descriptors as a result of not closing the
3096 listening socket. In this case, eventually the system will reach
3097 file-max and subsequent connections will fail.
3101 Signed-off-by: Jon Bernard <jbernard@tuxion.com>
3102 (cherry picked from commit 88fabb1ee6cd3c7d211ca46919055195e32245db)
3104 commit 8372f1d53b847224bf4c61894d18851e84ccd357
3105 Author: Jianpeng Ma <jianpeng.ma@intel.com>
3106 Date: Fri Mar 6 15:54:07 2015 +0800
3108 osd: refuse to write a new erasure coded object with an offset > 0
3110 Even if the offset is properly aligned.
3112 http://tracker.ceph.com/issues/11507 Fixes: #11507
3114 Signed-off-by: Jianpeng Ma <jianpeng.ma@intel.com>
3115 Signed-off-by: Loic Dachary <ldachary@redhat.com>
3116 (cherry picked from commit a4f1256c214ee0e7ebb91ac4ea8655f5d9642dc8)
3118 commit 909096713dcef5dc4fdbb22a9971609554688e23
3119 Author: Loic Dachary <ldachary@redhat.com>
3120 Date: Sun Mar 8 15:15:35 2015 +0100
3122 ceph-disk: more robust parted output parser
3124 In some cases, depending on the implementation or the operating system,
3126 parted --machine -- /dev/sdh print
3128 may contain empty lines. The current parsing code is fragile and highly
3129 depends on output details. Replace it with code that basically does the
3130 same sanity checks (output not empty, existence of units, existence of
3131 the dev entry) but handles the entire output instead of checking line by
3134 http://tracker.ceph.com/issues/10983 Fixes: #10983
3136 Signed-off-by: Loic Dachary <ldachary@redhat.com>
3137 (cherry picked from commit f5acf6bb6a342b05897605703d7d9cb7c09714fd)
3139 commit 8c573c8826096d90dc7dfb9fd0126b9983bc15eb
3140 Author: Yehuda Sadeh <yehuda@redhat.com>
3141 Date: Tue Sep 30 14:15:47 2014 -0700
3143 rgw: swift GET / HEAD object returns X-Timestamp field
3146 Backport: giant, firefly, dumpling
3147 Swift clients expect X-Timestamp header, dump it.
3149 Signed-off-by: Yehuda Sadeh <yehuda@redhat.com>
3150 (cherry picked from commit 5b41d80b7fb9ed96c26801fc42c044191bb18d84)
3152 commit 9614e4dc5be745dab367d17f29a07377e348b2ca
3153 Author: Daniel J. Hofmann <daniel@trvx.org>
3154 Date: Mon May 5 16:06:45 2014 +0200
3156 Variable length array of std::strings (not legal in C++) changed to std::vector<std::string>
3158 Signed-off-by: Daniel J. Hofmann <daniel@trvx.org>
3159 (cherry picked from commit 08a4e8889783673436972a979c2fc780466893c8)
3161 commit 2cb2333ecdf2106925d74e024697f73ad947f378
3162 Author: Thorsten Behrens <tbehrens@suse.com>
3163 Date: Thu Jul 3 00:15:03 2014 +0200
3165 ceph_argparse_flag has no regular 3rd parameter.
3167 With clang warning: 'va_start' has undefined behavior with reference
3168 types, noticing copy-paste mistake on ceph_argparse_flag.
3170 Signed-off-by: Thorsten Behrens <tbehrens@suse.com>
3171 (cherry picked from commit 0c7c7223a2beb4ffb953d89d316f87d350677063)
3173 commit f6022639758ec13b9a25b03cd831882db0b517b3
3174 Author: Yehuda Sadeh <yehuda@redhat.com>
3175 Date: Wed Dec 17 17:12:43 2014 -0800
3177 rgw: use correct objv_tracker for bucket instance
3179 When trying to create a bucket that already existed, use the
3180 objv_tracker of the newly created instance, and not of the original
3183 Signed-off-by: Yehuda Sadeh <yehuda@redhat.com>
3184 (cherry picked from commit fe158ecc25feefcea8aea4133118e4a84900a8ec)