RepoMirrors/ceph
1
0
Fork 0
mirror of https://github.com/ceph/ceph synced 2025-03-11 02:39:05 +00:00
Code Issues Packages Projects Releases Wiki Activity
60,269 Commits 219 Branches 396 Tags 1.4 GiB
5aa8121d4e
Commit Graph

60269 Commits

This Branch
This Branch
All Branches
Author SHA1 Message Date
Matt Benjamin
5aa8121d4e Merge pull request #12143 from rook/wip-boost-ext 
cmake: fix boost build on ubuntu 16.10 yakkety
 2016-11-22 16:29:26 -05:00
Jason Dillaman
ac4a407cd2 Merge pull request #12078 from yangdongsheng/rbd_cleanup 
librbd: minor cleanup

Reviewed-by: Jason Dillaman <dillaman@redhat.com>
 2016-11-22 16:03:30 -05:00
Bassam Tabbara
3eef10b977 cmake: fix boost build on ubuntu yakkety 
If the host system has boost version 1.61 or higher (as in the case
for ubuntu 16.10 yakkety) the ceph build is currently broken. it
will pickup the system boost libraries but use the headers from the
submodule.

This commit ensure that when WITH_SYSTEM_BOOST is OFF we always
use the boost libraries and headers built from the submodule.

Signed-off-by: Bassam Tabbara <bassam.tabbara@quantum.com>
 2016-11-22 11:53:34 -08:00
Gregory Farnum
9d2fd4bd0e Merge pull request #12110 from batrick/client-failure-msg 
client: improve failure messages/debugging

Reviewed-by: Greg Farnum <gfarnum@redhat.com>
 2016-11-22 14:32:53 -05:00
Kefu Chai
f0b7d36a92 Merge pull request #12124 from branch-predictor/bp-fast-markdown-doc 
doc/rados/configuration/osd-config-ref.rst: document the fast mark down

Reviewed-by: Kefu Chai <kchai@redhat.com>
 2016-11-22 19:20:01 +08:00
Kefu Chai
99063bcf73 Merge pull request #12123 from branch-predictor/bp-fast-markdown-is-default 
test/osd-fast-mark-down.sh: wrong assumption on first subtest

Reviewed-by: Kefu Chai <kchai@redhat.com>
 2016-11-22 19:04:39 +08:00
Piotr Dałek
9c5d253a2e doc/rados/configuration/osd-config-ref.rst: document the fast mark down 
Document the "osd fast fail on connection refused" option.

Signed-off-by: Piotr Dałek <git@predictor.org.pl>
 2016-11-22 10:42:28 +01:00
Piotr Dałek
79f59e4fd5 test/osd-fast-mark-down.sh: wrong assumption on first subtest 
First subtest checks for proper behavior when "osd fast fail on
connection refused" is disabled, yet it doesn't disable it and
fails, as is it enabled by default.

Signed-off-by: Piotr Dałek <git@predictor.org.pl>
 2016-11-22 10:14:14 +01:00
Haomai Wang
69c29960a2 Merge pull request #8948 from cooboos/perf-reset 
test: add perf-reset test in test/perf_counters.cc

Reviewed-by: Haomai Wang <haomai@xsky.com>
 2016-11-22 14:31:08 +08:00
Kefu Chai
013d4b2053 Merge pull request #12072 from dzafman/wip-test-17830 
tests: fix osd-scrub-repair.sh

Reviewed-by: Kefu Chai <kchai@redhat.com>
 2016-11-22 11:17:24 +08:00
Dongsheng Yang
9a9b0288d9 librbd: cleanup: use a more proper variable name 
Signed-off-by: Dongsheng Yang <dongsheng.yang@easystack.cn>
 2016-11-21 19:41:42 -05:00
Yuri Weinstein
f4ecdc55af Merge pull request #11993 from ceph/wip-17809 
buffer: add advance(unsigned) back

Reviewed-by: Sage Weil <sage@redhat.com>
 2016-11-21 13:37:17 -08:00
Yuri Weinstein
3afc8538c0 Merge pull request #12049 from liewegas/wip-kv-iterators 
kv: remove snapshot iterator

Reviewed-by: Kari, Varada <varada.kari@sandisk.com>
Reviewed-by: Samuel Just <sjust@redhat.com>
 2016-11-21 13:36:36 -08:00
Casey Bodley
6266dfa9e0 Merge pull request #11719 from zhangsw/fix-rgw-requestpay-bug 
rgw: Anonymous users shouldn't be able to access requester pays buckets.

Reviewed-by: Matt Benjamin <mbenjamin@redhat.com>
Reviewed-by: Casey Bodley <cbodley@redhat.com>
 2016-11-21 15:35:34 -05:00
Matt Benjamin
7d3d1c4743 Merge pull request #11989 from idealguo/httpargs-parse-cleanup 
rgw: skip empty http args in method parse() to avoid extra effort
 2016-11-21 15:14:01 -05:00
Casey Bodley
f579a3e1be Merge pull request #12021 from rzarzynski/wip-rgw-comp-noimplacefact 
rgw: compression uses optional::emplace instead of in-place factories

Reviewed-by: Casey Bodley <cbodley@redhat.com>
 2016-11-21 14:11:15 -05:00
Casey Bodley
ded5f005e1 Merge pull request #12108 from adamemerson/wip-compile-compress 
compress: Fix compilation failure from missing header

Reviewed-by: Casey Bodley <cbodley@redhat.com>
 2016-11-21 13:20:23 -05:00
Casey Bodley
139452c6cb Merge pull request #11827 from oritwas/wip-rgw-upgrade-doc 
rgw: add recovery procedure for upgrade to older version of jewel

Reviewed-by: Casey Bodley <cbodley@redhat.com>
 2016-11-21 13:12:29 -05:00
Jason Dillaman
2357af93f8 Merge pull request #11940 from trociny/wip-17880 
librbd: release lock after demote

Reviewed-by: Jason Dillaman <dillaman@redhat.com>
 2016-11-21 12:38:27 -05:00
Jason Dillaman
5cd929aee3 Merge pull request #11960 from vshankar/wip-librbd-ec-support 
librbd: bug fixes for optional data pool support

Reviewed-by: Jason Dillaman <dillaman@redhat.com>
 2016-11-21 12:36:54 -05:00
Adam C. Emerson
9ad32939e8 compress: Fix compilation failure from missing header 
Compress was failing due to the random header not being included.

Signed-off-by: Adam C. Emerson <aemerson@redhat.com>
 2016-11-21 12:34:23 -05:00
Radoslaw Zarzynski
d7100c4171 rgw: compression uses boost::optional::emplace instead of in-place factories. 
Signed-off-by: Radoslaw Zarzynski <rzarzynski@mirantis.com>
 2016-11-21 17:56:30 +01:00
Matt Benjamin
9fd205d9f4 Merge pull request #11901 from ceph/wip-rgw-compression-type-random 
rgw: add rgw_compression_type=random for teuthology testing
 2016-11-21 11:35:25 -05:00
Matt Benjamin
1f9b698eca Merge pull request #12073 from rzarzynski/wip-rgw-compile-asio-by-default 
rgw: compilation of the ASIO front-end is enabled by default.
 2016-11-21 11:24:32 -05:00
Matt Benjamin
43eb7f307a Merge pull request #12103 from rzarzynski/wip-rgw-auth-static_vector 
rgw: avoid dynamic memory allocations in rgw_swift_auth.cc by using static_vector

verified: f23
 2016-11-21 10:53:56 -05:00
David Zafman
b9c85c4c54 test: Re-enable osd-scrub-repair.sh again 
Signed-off-by: David Zafman <dzafman@redhat.com>
 2016-11-21 07:52:12 -08:00
David Zafman
402e5146e8 osd: Fix broken ec_size_mismatch setting 
Caused by: 1e95f2ce64

Signed-off-by: David Zafman <dzafman@redhat.com>
 2016-11-21 07:52:12 -08:00
Kefu Chai
a70dc74fe3 Merge pull request #12101 from tchaikov/wip-mark-down-test-cleanup 
test/osd/osd-fast-mark-down.sh: remove unnecessary teardown() calls

Reviewed-by: Loic Dachary <ldachary@redhat.com>
Reviewed-by: Piotr Dałek <piotr.dalek@ts.fujitsu.com>
 2016-11-21 23:23:47 +08:00
Sage Weil
7c059f9a42 Merge pull request #12100 from tchaikov/wip-rocksdb 
rocksdb: update to latest

Reviewed-by: Sage Weil <sage@redhat.com>
 2016-11-21 09:14:27 -06:00
Sage Weil
32c4e125d1 Merge pull request #12093 from tchaikov/wip-yasm-wrapper 
yasm-wrapper: translate "-isystem $1" to "-i $1"

Reviewed-by: Sage Weil <sage@redhat.com>
 2016-11-21 09:11:59 -06:00
Kefu Chai
23c21238b8 Merge pull request #12005 from wjwithagen/wip-wjw-freebsd-ceph-helpers 
workunits/ceph-helpers.sh: FreeBSD returns a different errorstring.

Reviewed-by: Kefu Chai <kchai@redhat.com>
 2016-11-21 22:53:23 +08:00
Radoslaw Zarzynski
cb782e1855 Revert "rgw: temporarily use std::vector in place of static_vector of Boost." 
This reverts commit bc23e0f7fa.
We're doing that because the reverted commit was a makeshift
solution to not fail Ceph compilation on platforms lacking
Boost modern enough to ship the container/static_vector.hpp.
As we got the in-tree Boost the commit isn't necessary anymore.

Signed-off-by: Radoslaw Zarzynski <rzarzynski@mirantis.com>
 2016-11-21 15:30:31 +01:00
Kefu Chai
60bb32a7b0 test/osd/osd-fast-mark-down.sh: remove unnecessary teardown() calls 
main() does this for us.

Signed-off-by: Kefu Chai <kchai@redhat.com>
 2016-11-21 22:24:23 +08:00
Jason Dillaman
9369f209b2 Merge pull request #12035 from VictorDenisov/fix_snap_remove 
rbd: use snap_remove implementation from internal

Reviewed-by: Jason Dillaman <dillaman@redhat.com>
 2016-11-21 09:12:51 -05:00
Kefu Chai
cb0fd953c4 Merge pull request #12019 from branch-predictor/bp-fix-fastkill 
test/osd/osd-fast-mark-down.sh: introduce large timeout

Reviewed-by: Kefu Chai <kchai@redhat.com>
 2016-11-21 21:47:20 +08:00
Kefu Chai
0d891f5d5d rocksdb: update to latest 
Latest master includes

- CMakeLists.txt: FreeBSD has jemalloc as default malloc
- cmake: drop "-march=native" from CXX_FLAGS (#1429)

Signed-off-by: Kefu Chai <kchai@redhat.com>
 2016-11-21 21:43:29 +08:00
Kefu Chai
a6aa737953 Merge pull request #12095 from dachary/wip-17830-eio 
tests: fix ceph-helpers.sh wait_for_clean delays

Reviewed-by: Kefu Chai <kchai@redhat.com>
 2016-11-21 21:27:21 +08:00
John Spray
9773abb50f Merge pull request #11813 from ukernel/wip-17801 
mds: ignore 'session evict' when mds is replaying log

Reviewed-by: John Spray <john.spray@redhat.com>
 2016-11-21 12:53:46 +00:00
Igor Fedotov
71b1a41c09 Merge pull request #12065 from xiexingguo/xxg-wip-kill-gc-counters 
os/bluestore: kill dead gc-related counters
Reviewed-by: Sage Weil <sage@redhat.com>
Reviewed-by: Igor Fedotov <ifedotov@mirantis.com>
 2016-11-21 15:22:28 +03:00
Loic Dachary
5e625674a8 tests: fix ceph-helpers.sh wait_for_clean delays 
The TENTH_TIMEOUT was not delcared as an int and failed to be set with
the correct number. The test of the function did not catch this.

Implement computing of the increasingly large sleep delays in a separate
function so that it can be tested more easily. Give up on sub-second
sleep because a the function will not sleep at all if the cluster is
already clean. And if it is not already clean, it is very unlikely to
become clean within less than a second. The downside of having very
short sleep time is that it needlessly stress the machine and also
possibly spam the logs.

Refs: http://tracker.ceph.com/issues/17830

Signed-off-by: Loic Dachary <loic@dachary.org>
 2016-11-21 11:42:42 +01:00
Kefu Chai
af3e7143d7 yasm-wrapper: translate "-isystem $1" to "-i $1" 
this silences the warning of:

yasm: warning: can open only one input file, only the last file will be
processed

as yasm does not understand -isystem, but gcc does.

Signed-off-by: Kefu Chai <kchai@redhat.com>
 2016-11-21 12:00:25 +08:00
Venky Shankar
3a91d7bd2e qa / workunits: make rbd import/export work with image data pool 
Signed-off-by: Venky Shankar <vshankar@redhat.com>
 2016-11-20 21:21:50 +05:30
Venky Shankar
3f6ef037ce tests / rbd: fix rbd python binding test 
Signed-off-by: Venky Shankar <vshankar@redhat.com>
 2016-11-20 21:21:50 +05:30
Venky Shankar
e98b8ceb86 test / rbd-mirror: image data pool test support 
Signed-off-by: Venky Shankar <vshankar@redhat.com>
 2016-11-20 21:21:50 +05:30
Venky Shankar
3d3d0fb31e test / librbd: image data pool test support 
Signed-off-by: Venky Shankar <vshankar@redhat.com>
 2016-11-20 21:21:50 +05:30
Venky Shankar
1c1f4d2a34 test / librbd: use data pool IoCtx while listing objects 
Signed-off-by: Venky Shankar <vshankar@redhat.com>
 2016-11-20 21:21:50 +05:30
Venky Shankar
0206e15b74 librbd / ObjectMap: use data_ctx/md_ctx pool context as required 
Signed-off-by: Venky Shankar <vshankar@redhat.com>
 2016-11-20 21:21:50 +05:30
Venky Shankar
75def529ee librbd / CopyupRequest: use data pool ctx during copyup 
Signed-off-by: Venky Shankar <vshankar@redhat.com>
 2016-11-20 21:21:50 +05:30
Venky Shankar
cce335fccf librados: drop io_ctx_impl on ioctx_create/create2 
close() was never called for the passed in IoCtx which
could probably result in an IoCtx leak if the original
IoCtx was a valid pool context allocated earlier.

Its kind of better to do it here rather than to leave
the destruction on the caller for better (or cleaner)
common case handling.

Signed-off-by: Venky Shankar <vshankar@redhat.com>
 2016-11-20 21:21:50 +05:30
Kefu Chai
49bc71bacf Merge pull request #12080 from wjwithagen/wip-wjw-freebsd-install-deps 
install-deps.sh: JQ is needed in one script

Reviewed-by: Kefu Chai <kchai@redhat.com>
 2016-11-20 22:17:58 +08:00