os/bluestore: fix potential uninitialized nid of onode
The _zero() process may implicitly create a new onode,
thus we shall call _assign_nid() to initialize the nid
properly. And if the onode already has one, _assign_nid()
does nothing.
So it is proper to call _assign_nid() here under any case.
Mark's comments:
This passed "ceph_test_objectstore --gtest_filter=*/2".
This PR did not appear to have a significant impact on performance tests.
Closes #10236
Signed-off-by: xie xingguo <xie.xingguo@zte.com.cn>
os/bluestore: check against we don't overflow
Signed-off-by: xie xingguo <xie.xingguo@zte.com.cn>
os/bluestore: try to reap as many collections as we can
So if there is one collection getting contiguously stucking,
we don't abort at the same point each time.
Signed-off-by: xie xingguo <xie.xingguo@zte.com.cn>
os/bluestore: make device size of BitFreelistManager is block-size aligned
Otherwise if we try to set past-eof blocks as allocated durint create(),
the call to _xor() will trigger the firing of the following assert:
Mark Nelson [Sun, 17 Jul 2016 11:39:41 +0000 (06:39 -0500)]
os/bluestore: revert preferred csum behavior
This passes "ceph_test_objectstore --gtest_filter=*/2".
This restores 4K random read performance to previous levels when objects
are were previously written out using large IOs (4MB in this case):
Somnath Roy [Sat, 9 Jul 2016 02:41:46 +0000 (22:41 -0400)]
Bluestore: Fixed a Bluestore crash
A bluestore race condition is been fixed by protecting txc structures
within _txc_state_poc with collection lock.
Mark's comments:
This fixes segfaults during random write tests with bluestore.
This passes "ceph_test_objectstore --gtest_filter=*/2".
This may introduce a small performance regresion, though there is enough
noise in the results to make it inconclusive.
Closes #10220
Signed-off-by: Somnath Roy <somnath.roy@sandisk.com>
test: run_seed_to_range.sh: check existance of a directory using [ -d "$dir" ]
sadly, sh evalutes `[ -d ]` to true. as it takes "-d" as a non-empty
string as true.
this fixes following failure
```
2016-07-12T23:22:02.839 INFO:teuthology.orchestra.run.mira084.stderr:cp:
missing destination file operand after β.β
2016-07-12T23:22:02.839 INFO:teuthology.orchestra.run.mira084.stderr:Try
'cp --help' for more information.
```
see
http://pulpito.ceph.com/kchai-2016-07-12_23:09:35-rados-wip-kefu-testing2---basic-mira/311334/
this fixes the test failure of
```
2016-07-12T23:29:40.935
INFO:tasks.workunit.client.0.mira101.stderr:/home/ubuntu/cephtest/workunit.client.0/cephtool/test.sh:
line 153: /CEPH_WATCH_9445: Permission denied
```
see
http://pulpito.ceph.com/kchai-2016-07-12_23:09:35-rados-wip-kefu-testing2---basic-mira/311333/
Fixes: http://tracker.ceph.com/issues/16665
Instead of drain_all_but() that specifies number of stacks to leave behind,
added drain_all_but_stack() that has a specific stack specified. This is needed
so that we don't call wakeup() through lease_cr->go_down() on a cr stack that
was already collected.
cmake: disable -fvar-tracking-assignments for ceph_dencoder.cc
there are too many variables involved in main() of ceph_dencoder.cc, but
-fvar-tracking-assignments is enabled by default, when the compiler
finds that the number of variables exceeds the limit
"max-vartrack-size", it gives up, and tries again disabling this option.
this change avoids the recompilation of ceph_dencoder.cc and also silences
the warning of
```
/home/jenkins-build/build/workspace/ceph-pull-requests/src/test/encoding/ceph_dencoder.cc:
In function βint main(int, const char**)β:
/home/jenkins-build/build/workspace/ceph-pull-requests/src/test/encoding/ceph_dencoder.cc:289:5:
note: variable tracking size limit exceeded with
-fvar-tracking-assignments, retrying without
int main(int argc, const char **argv)
^
```
include/ceph_fs.h: guard `#define CEPH_SETATTR_*` with #ifndef
this silences the warnings like:
```
from
/srv/autobuild-ceph/gitbuilder.git/build/out~/ceph-11.0.0-513-gfe98b0b/src/java/native/libcephfs_jni.cc:33:
warning:
/srv/autobuild-ceph/gitbuilder.git/build/out~/ceph-11.0.0-513-gfe98b0b/src/include/ceph_fs.h:373:0:
"CEPH_SETATTR_MODE" redefined [enabled by default]
^
In file included from
/srv/autobuild-ceph/gitbuilder.git/build/out~/ceph-11.0.0-513-gfe98b0b/src/java/native/libcephfs_jni.cc:32:0:
/srv/autobuild-ceph/gitbuilder.git/build/out~/ceph-11.0.0-513-gfe98b0b/src/include/cephfs/libcephfs.h:106:0:
note: this is the location of the previous definition
^
```
Josh Durgin [Thu, 23 Jun 2016 00:54:33 +0000 (17:54 -0700)]
googletest: use ceph branch
Added the same functionality as the previous gtest/gmock branches -
.gitignore and not installing shared libs that we only want to
statically link to test programs.
the mock_open() does not support "for i in open()", and readline() and
readlines() support was added in py3.3 and py3.4 so for better backwards
compatibility, we should change the code being tested to use the plain
read() call.
and also use open(path) instead of open(path, 'rb') for simplicity,
otherwise we need to use bytestring for comparison and pass the same
parameters to mock.
this fixes the pep8 failure:
```
pep8 runtests: commands[0] | flake8 ceph_detect_init tests
tests/test_all.py:121:17: E128 continuation line under-indented for
visual indent
tests/test_all.py:125:17: E128 continuation line under-indented for
visual indent
tests/test_all.py:129:17: E128 continuation line under-indented for
visual indent
tests/test_all.py:133:17: E128 continuation line under-indented for
visual indent
```
Haomai Wang [Thu, 7 Jul 2016 11:12:46 +0000 (19:12 +0800)]
ceph_test_msgr: clean previous sent queue infos
1. A -> B
2. goto standby
3. B mark down
4. A reconnect to B
5. got reset session and dispatch remote reset
6. because remote reset is executed in DispatchQueue, it will be delayed
7. A -> B successfully and begin to send message
8. assert because we found the first message is missing but it's reasonble
Haomai Wang [Wed, 29 Jun 2016 09:14:16 +0000 (17:14 +0800)]
msg/async: make sure worker started before let msgr ready
When we create event thread, it need a little time to enter event loop(like
calling set_owner), if caller is going to call create_file_event before event
thread enter event loop, it will trigger assert.
Haomai Wang [Wed, 29 Jun 2016 08:54:16 +0000 (16:54 +0800)]
msg/async: make EventCenter notify file event creating when set_owner
EventCenter::init is called by other thread instead of event thread, so we
need to move create_file_event to set_owner which is called by event thread.
Haomai Wang [Wed, 29 Jun 2016 08:26:29 +0000 (16:26 +0800)]
msg/async/AsyncConnection: swap eventcenter when replacing
Previously we only exchange fd when replacing, now we will introduce dpdk
plugin in the near future. It needs all fd used locally which not like
kernel socket shared by all cores.
So we need to add EventCenter swapping to let each socket is associated to
EventCenter.
Haomai Wang [Tue, 8 Mar 2016 05:59:50 +0000 (13:59 +0800)]
AsyncMessenger: make create/delete_file_event within event thread
We are make each AsyncConnection/AsyncMessenger only modify its file event
in event thread. So make sure create/delete_file_event aren't directly called.
projects / ceph-ci.git / log