This prevents partial results from concatenating with the error message.
Fixes: http://tracker.ceph.com/issues/21252
Signed-off-by: Patrick Donnelly <pdonnell@redhat.com>
The teuthology machines are periodically running out of space
due to the aggressive log settings.
Fixes: http://tracker.ceph.com/issues/21251
Signed-off-by: Jason Dillaman <dillaman@redhat.com>
_should_compact_log uses new_log != nullptr to tell whether compaction is
already in progress, but we don't set it until we are midway through the
process. Set it at the top of the method to prevent reentry.
See 455cc6cea2, which failed to implement
this properly.
Fixes: http://tracker.ceph.com/issues/21250
Signed-off-by: Sage Weil <sage@redhat.com>
max in eversion_t::max() can be static, no need to re constructs
it every time when calling eversion_t::max(), and max can be
constructed using eversion_t(epoch_t e, version_t v), it better than
firstly construct it by eversion_t() and then reassign it to (-1,-1)
Change-Id: I85de5685024e61ebc2d1816b0969be67bdaf45f5
Signed-off-by: yang.wang yang.wang@easystack.cn
Idea of this is to allow scripts to lookup the contributor name/email by GitHub
username. This is useful in particular for adding appropriate "Reviewed-by"s
for each GitHub style "review".
Signed-off-by: Patrick Donnelly <pdonnell@redhat.com>
During up:resolve, the MDS tries to merge each subtree with its parent. During
testing, QE found that many thousands of subtrees in a directory (made possible
using pins) would cause the MDS to spend minutes printing out subtree maps to
the debug log. This causes the heartbeat code to consider the MDS as stalled so
beacons are no longer sent to the mons resulting in the MDS being removed from
the rank.
A more complete solution to this problem is to selectively print subtrees
relating to the operation (e.g. the subtree and its parents).
Fixes: http://tracker.ceph.com/issues/21221
Fixes: https://bugzilla.redhat.com/show_bug.cgi?id=1485783
Signed-off-by: Patrick Donnelly <pdonnell@redhat.com>
As reported by Ademar de Souza Reis Jr <areis@redhat.com>:
This symlink should not be necessary anymore. QEMU is properly linked to
ceph/librbd these days (look at the qemu-block-rbd sub-package in Fedora and
the respective package in RHEL).
The symlink was a hack from a time when librbd was distributed and supported by
ceph and the rbd driver would be enabled at runtime only when/if the symlink
was present.
So the right fix is to actually get rid of the symlink altogether and never
touch (much less own) /usr/lib64/qemu.
Signed-off-by: Nathan Cutler <ncutler@suse.com>
os/bluestore: pass strict flag to bluestore_blob_use_tracker_t::equal()
Reviewed-by: Sage Weil <sage@redhat.com>
Reviewed-by: Igor Fedotov <ifedotov@mirantis.com>
refactor OSD::build_initial_pg_history() so we update the info only if a
new interval is identified. also, this is consistent with
OSD::build_past_intervals_parallel().
Signed-off-by: Kefu Chai <kchai@redhat.com>
we need to manage the ip addr of the "standbys" state,
because the hostname/gid is insufficient to locate the
Standby node. we add ip of the mgr standby to metadata.
Signed-off-by: huanwen ren <ren.huanwen@zte.com.cn>
because it will also statistic the latency of sync write
Signed-off-by: Pan Liu <wanjun.lp@alibaba-inc.com>
Signed-off-by: Ziye Yang <optimistyzy@gmail.com>
nvme-device-manager, to avoid name duplication with main thread
"ceph-osd"
Signed-off-by: Pan Liu <wanjun.lp@alibaba-inc.com>
Signed-off-by: Ziye Yang <optimistyzy@gmail.com>
os/bluestore: separate finisher for deferred_try_submit
Reviewed-by: xie xingguo <xie.xingguo@zte.com.cn>
Reviewed-by: Varada Kari <varada.kari@sandisk.com>
