Sage Weil
855955e58e
osd: reduce size of osdmap cache, messages
...
On large clusters, these large caches can be problematic (as maps get big).
We've seen good results with extremely small caches (10s of maps). Make
a more modest reduction.
Signed-off-by: Sage Weil <sage@redhat.com>
2017-05-25 23:13:37 -04:00
Sage Weil
c1b92cdcbb
test/osd/osd-config.sh: fix test to isolate cases
...
The third test (increasing osd_map_max_advance)
was triggering a warning from the 4th case (which
it didn't before).
Signed-off-by: Sage Weil <sage@redhat.com>
2017-05-25 23:13:37 -04:00
Sage Weil
c489d45113
osd: fadvise hint WILL_NEED when reading encoded osdmaps
...
This way will ensure we cache data for recent osdmaps if we need to for
the benefit of laggy clients... even if (in bluestore's case)
bluestore_default_buffered_reads = false (it's true by default). This
should mitigate any tail latency/work even if the osdmap cache size is too
small.
Signed-off-by: Sage Weil <sage@redhat.com>
2017-05-25 22:52:33 -04:00
Sage Weil
73985a5b10
osd: instrument osdmap bl cache hits and misses
...
Add perfcounters so we can see whether we are missing osdmaps in the
cache. This will let us tell whether, given a workload or environment,
our osdmap cache might be too small.
Signed-off-by: Sage Weil <sage@redhat.com>
2017-05-25 22:52:33 -04:00
Sage Weil
0ff3d4291c
osd: do not bother with misdirected op diagnosis by default
...
We enable osd_debug_misdirected_ops in QA, but this is wasted effort on
a production cluster. In particular, it means that a idle client that
sends an op to the wrong OSD based on an old map will require that OSD to
load that old map into memory to decide whether to print a warning... all
on the off-chance that the client is buggy.
Signed-off-by: Sage Weil <sage@redhat.com>
2017-05-25 22:29:22 -04:00
Sage Weil
c150cba817
osd: remove osd_enxio_on_misdirected_op option
...
There is no good reason anyone would want this turned on.
Introduced 923e7f5ce5
(post-kraken), but
backported to kraken and jewel (10.2.6).
Signed-off-by: Sage Weil <sage@redhat.com>
2017-05-25 22:29:22 -04:00
Yuri Weinstein
25b8e0f63f
Merge pull request #15146 from tchaikov/wip-split-test-tiering
...
qa: split test_tiering into smaller pieces
Reviewed-by: Sage Weil <sage@redhat.com>
2017-05-25 13:43:53 -07:00
Yuri Weinstein
0f7e269880
Merge pull request #15156 from tchaikov/wip-ceph.in-assert-connected
...
ceph.in: assert(state==connected) before help_for_target()
Reviewed-by: Jos Collin <jcollin@redhat.com>
Reviewed-by: John Spray <john.spray@redhat.com>
2017-05-25 13:43:17 -07:00
Yuri Weinstein
516c3ec452
Merge pull request #15224 from tchaikov/wip-denc-no-deep-copy-for-segmented-buffer
...
denc: add need_contiguous to denc_traits
Reviewed-by: Sage Weil <sage@redhat.com>
Reviewed-by: Kefu Chai <kchai@redhat.com>
Reviewed-by: Casey Bodley <cbodley@redhat.com>
2017-05-25 13:40:23 -07:00
John Spray
f80e0973f5
Merge pull request #15062 from ukernel/wip-19912
...
qa/tasks/cephfs: use getattr to guarantee inode is in client cache
Reviewed-by: John Spray <john.spray@redhat.com>
2017-05-25 18:44:54 +01:00
Ken Dreyer
5b40557a2c
Merge pull request #15270 from ceph/wip-subman-cron-attr-tserlin
...
Set subman cron attributes in spec file
Reviewed-by: Ken Dreyer <kdreyer@redhat.com>
2017-05-25 11:43:24 -06:00
Dan Mick
19197d88aa
Merge pull request #15253 from tchaikov/wip-ceph-daemon-termsize
...
pybind/ceph_daemon.py: fix Termsize.update
Reviewed-by: Dan Mick <dmick@redhat.com>
2017-05-25 10:09:14 -07:00
Jos Collin
4ef5559fe8
Merge pull request #15279 from joscollin/wip-cleanup-rename-filejournal-object
...
test: Rename FileJournal object to distinguish
2017-05-25 21:27:40 +05:30
Thomas Serlin
a9eb6ecea7
Set subman cron attributes in spec file
...
Fixes: http://tracker.ceph.com/issues/20074
Signed-off-by: Thomas Serlin <tserlin@redhat.com>
2017-05-25 10:16:51 -04:00
Nathan Cutler
e9a81ba41c
Merge pull request #15200 from smithfarm/wip-rdma-devel
...
build/ops: rpm: move RDMA and python-prettytables build dependencies to distro-conditional section
Reviewed-by: Kefu Chai <kchai@redhat.com>
Reviewed-by: Ken Dreyer <kdreyer@redhat.com>
2017-05-25 11:18:29 +02:00
Jos Collin
9c431c7878
test: Rename FileJournal object to distinguish
...
Renamed FileJournal object for easier searching and identifying with in test_filejournal.cc
Signed-off-by: Jos Collin <jcollin@redhat.com>
2017-05-25 12:06:14 +05:30
Jos Collin
2f9ee0badd
Merge pull request #15258 from big-henry/wip-mds-misc-fix
...
mds: fix client ID truncation
2017-05-25 09:29:33 +05:30
Nathan Cutler
9e297686f7
Merge pull request #15235 from smithfarm/wip-doc-empowers
...
doc: fix factual inaccuracy in doc/architecture.rst
Reviewed-by: Kefu Chai <kchai@redhat.com>
Reviewed-by: Sage Weil <sage@redhat.com>
2017-05-25 03:32:30 +02:00
Josh Durgin
814a6f3b4b
Merge pull request #15275 from ktdreyer/wip-17396
...
docs: document "osd recovery max single start" setting
Reviewed-by: Josh Durgin <jdurgin@redhat.com>
2017-05-24 17:05:10 -07:00
Sage Weil
6d520e9166
Merge pull request #15276 from liewegas/wip-ext4-again
...
qa/suites/rados/singleton-nomsgr: fix syntax
2017-05-24 17:09:25 -05:00
Sage Weil
5d80c74e63
Merge pull request #15252 from liewegas/wip-cleanup-tell
...
qa/tasks/ceph_manager: 'ceph $service tell ...' is obsolete
Reviewed-by: Kefu Chai <kchai@redhat.com>
2017-05-24 17:05:32 -05:00
Sage Weil
ee84a23eef
Merge pull request #15250 from neha-ojha/wip-19989
...
osd/OSDMap.cc: check if osd is out in subtree_type_is_down
Reviewed-by: Sage Weil <sage@redhat.com>
2017-05-24 17:04:52 -05:00
Sage Weil
035cad4ff1
Merge pull request #14944 from yuyuyu101/wip-msgr-throttle-ops
...
ceph_osd: remove client message cap limit
Reviewed-by: Sage Weil <sage@redhat.com>
2017-05-24 17:03:36 -05:00
Sage Weil
499c1afc81
Merge pull request #14655 from chardan/jfw-wip-halflife_atomic_t-gojira
...
test: migrate atomic_t to std::atomic
Reviewed-by: Kefu Chai <kchai@redhat.com>
2017-05-24 17:03:13 -05:00
Sage Weil
8e72e0ee7a
Merge pull request #11966 from aclamk/wip-linear-zeros-crc
...
common: Improved CRC calculation for zero buffers
2017-05-24 17:02:28 -05:00
Sage Weil
d2dac535f5
qa/suites/rados/singleton-nomsgr/health...: whitelist wrongly marked me down
...
Signed-off-by: Sage Weil <sage@redhat.com>
2017-05-24 17:58:12 -04:00
Sage Weil
85e2f3f31d
qa/suites/rados/singleton-nomsgr: fix syntax
...
This parsed out as
tasks:
- install: null
- ceph:
conf:
osd: osd max object name len = 400 osd max object namespace len = 64
- workunit:
clients:
all:
- rados/test_health_warnings.sh
which is clearly not correct.
Signed-off-by: Sage Weil <sage@redhat.com>
2017-05-24 16:46:00 -04:00
Ken Dreyer
87d47b0dfa
docs: document "osd recovery max single start" setting
...
Fixes: http://tracker.ceph.com/issues/17396
Signed-off-by: Ken Dreyer <kdreyer@redhat.com>
2017-05-24 14:41:17 -06:00
Matt Benjamin
cc52f051fe
Merge pull request #15028 from pritha-srivastava/wip-rgw-bucket-policy-test
...
rgw: Adding code to create tenanted user for s3 bucket policy tests.
2017-05-24 15:57:31 -04:00
Matt Benjamin
d06d13472c
Merge pull request #15070 from ceph/wip-turn-off-fcgi-tserlin
...
rgw: Turn off fcgi as a frontend
2017-05-24 15:53:57 -04:00
Sage Weil
773a3f3db2
Merge pull request #15175 from cloudbau/update-rgw-docs
...
doc: Update some RGW documentation
Reviewed-by: Casey Bodley <cbodley@redhat.com>
2017-05-24 11:38:34 -05:00
John Spray
9dd2bc8974
Merge pull request #15213 from jcsp/wip-20040
...
mon: fix accesing pending_fsmap from peon
Reviewed-by: Yan, Zheng <zyan@redhat.com>
2017-05-24 16:50:42 +01:00
John Spray
0f0fdeb18e
Merge pull request #15130 from ukernel/wip-19946
...
mds: fix CDir::merge() for mds_debug_auth_pins
Reviewed-by: John Spray <john.spray@redhat.com>
2017-05-24 16:49:42 +01:00
John Spray
ef9d555916
Merge pull request #15105 from ukernel/wip-19892
...
qa/cephfs: disable mds_bal_frag for TestStrays.test_purge_queue_op_rate
Reviewed-by: John Spray <john.spray@redhat.com>
2017-05-24 16:41:45 +01:00
John Spray
ee75318807
Merge pull request #15122 from batrick/test-fragment-error
...
qa: fix float parse error in test_fragment
Reviewed-by: John Spray <john.spray@redhat.com>
Reviewed-by: Kefu Chai <kchai@redhat.com>
2017-05-24 16:40:50 +01:00
John Spray
6ca92c502b
Merge pull request #15106 from ukernel/wip-export-pin-cleanup
...
ceph: simplify CInode::maybe_export_pin()
Reviewed-by: Patrick Donnelly <pdonnell@redhat.com>
2017-05-24 16:37:53 +01:00
John Spray
db6890e2c9
Merge pull request #15129 from ukernel/wip-mds-standby-for-rank
...
mon/MDSMonitor: respect mds_standby_for_rank config
Reviewed-by: Patrick Donnelly <pdonnell@redhat.com>
2017-05-24 16:37:03 +01:00
John Spray
4f5d7aafe1
Merge pull request #15131 from ukernel/wip-19955
...
mds: limit client writable range increment
Reviewed-by: Sage Weil <sage@redhat.com>
Reviewed-by: John Spray <john.spray@redhat.com>
2017-05-24 16:34:50 +01:00
John Spray
8535249bad
Merge pull request #15198 from renhwztetecs/renhw-wip-client-inode
...
client/inode: fix the dump type of Inode::dump()
Reviewed-by: Yan, Zheng <zyan@redhat.com>
2017-05-24 16:13:53 +01:00
Sage Weil
da7a99f5d1
Merge pull request #15264 from xiexingguo/wip-kill-typo
...
os/bluestore/KernelDevice: fix comments
Reviewed-by: Sage Weil <sage@redhat.com>
2017-05-24 08:43:35 -05:00
Jos Collin
cc7dea594d
Merge pull request #15266 from ztczll/master
...
test:Check make_writeable() return value
2017-05-24 16:11:48 +05:30
zhanglei
0bcb3d2f34
test/:Check make_writeable return value
...
Signed-off-by: zhanglei <243290414@qq.com>
2017-05-24 05:22:02 -04:00
xie xingguo
ba3d6b48e1
os/bluestore/KernelDevice: fix comments
...
Signed-off-by: xie xingguo <xie.xingguo@zte.com.cn>
2017-05-24 17:15:18 +08:00
Mykola Golub
1d74182a29
Merge pull request #15206 from ifed01/wip-fix-rbd-bench
...
rbd/bench: fix write gaps when doing sequential writes with io-threads > 1
Reviewed-by: Jason Dillaman <dillaman@redhat.com>
2017-05-24 09:57:02 +03:00
Kefu Chai
7f7ae7ac7c
pybind/ceph_daemon.py: fix Termsize.update
...
it's a regression introduced by 84f9c12
Signed-off-by: Kefu Chai <kchai@redhat.com>
2017-05-24 12:25:50 +08:00
Kefu Chai
77fbfc29d1
denc: add need_contiguous to denc_traits
...
so the decode() will not try to copy the bufferlist for a continous one
if the bufferlist is segmented *and* its length is greater than
CEPH_PAGE_SIZE (4K) if the decoded object does not "need_contiguous" per
its denc_traits<>.
copying a memory chunk could be expensive if the decoded bufferlist is
huge, so we should try to avoid this. this could happen when we read the
buffer from bluestore.
and drop the partial specialization for denc() which tries to differentiate
traits::featured and !traits::featured, it does not matter to
decode() if the type supports feature or not. the encode() does.
also use denc() in interval_set<>::decode(). unlike encode(), decode(),
encode_nohead() or decode_nohead(), denc() is not part of our legacy
dencoder, so i think it's fine and encouraged to use it.
Signed-off-by: Kefu Chai <kchai@redhat.com>
2017-05-24 11:29:07 +08:00
Sage Weil
ba3e590f7b
Merge pull request #15238 from xiexingguo/wip-fix-blue-mkfs
...
os/bluestore: write "mkfs_done" into disk only if we pass fsck() tests
Reviewed-by: Sage Weil <sage@redhat.com>
2017-05-23 22:08:49 -05:00
Sage Weil
2ebd3e8dcc
Merge pull request #15228 from chardan/jfw-wip-halflife_atomic_t-filestore
...
filestore: migrate atomic_t to std::atomic<>
Reviewed-by: Sage Weil <sage@redhat.com>
2017-05-23 22:08:32 -05:00
Sage Weil
5ab996ab3c
qa/tasks/ceph_manager: 'ceph $service tell ...' is obsolete
...
This died forever ago; no need for the fallback here.
Signed-off-by: Sage Weil <sage@redhat.com>
2017-05-23 22:53:53 -04:00
Sage Weil
0aa04f523b
Merge pull request #9307 from cxwshawn/wip-tc7-fix
...
crush: optimize header file dependency
Reviewed-by: Brad Hubbard <bhubbard@redhat.com>
2017-05-23 21:49:46 -05:00