rgw: fix rgw hang when do RGWRealmReloader::reload after go SIGHUP
Reviewed-by: Matt Benjamin <mbenjami@redhat.com>
Reviewed-by: Casey Bodley <cbodley@redhat.com>
rgw: fix asctime when logging in rgw_lc
Reviewed-by: Daniel Gryniewicz <dang@fprintf.net>
Reviewed-by: Radoslaw Zarzynski <rzarzyns@redhat.com>
Reviewed-by: Adam C. Emerson <aemerson@linuxbox.com>
Due to http://tracker.ceph.com/issues/20751 it is possible for the UP
state bit to be set for OSDs without EXISTS. Make sure it is cleared as
needed when the OSD id is reused.
Signed-off-by: Sage Weil <sage@redhat.com>
Rephrase as "Monitor daemon marked osd.<id> down, but it is
still running", which is more descriptive. Demote the
epoch information to DEBUG level.
Signed-off-by: John Spray <john.spray@redhat.com>
These were dead code and could cause confusion.
The actual default values live in options.cc since
it was added.
Signed-off-by: John Spray <john.spray@redhat.com>
...and reinstate the set_daemon_default versions of config options
in place of the special casing in common_init.
Signed-off-by: John Spray <john.spray@redhat.com>
This avoids compiler warning about one oversized
literal, allows us to add_service en-masse to
the whole collection, and avoids potential merge
conflicts when config changes in separate subsystems
are made.
Signed-off-by: John Spray <john.spray@redhat.com>
Check the mon_max_pool_pg_num when creating pools, but did not
check in the the modification. So we can create a pool and then
modify pg_num exceeds the mon_max_pool_pg_num limit.
By following these steps can verify this problem
step 1: set mon_max_pool_pg_num = 65536
step 2: ceph osd pool create testpool 65536 65536
step 3: ceph osd pool set testpool pg_num 65537
Signed-off-by: chenhg <c744402859@gmail.com>
foobar_ is taken as a hyper link by RST. so escape the underscores.
this change can avoid the warnings like:
ceph/doc/release-notes.rst:4593: ERROR: Unknown target name: "rgwgetusage".
where the source code looks like:
```
rgw: need to 'open_object_section' before dump stats in 'RGWGetUsage_…(`issue#17499 <http://tracker.ceph.com/issues/17499>`_, `pr#11325 <http://github.com/ceph/ceph/pull/11325>\
`_, weiqiaomiao)
```
and apparently, "RGWGetUsage_" is not a hyper link here.
Signed-off-by: Kefu Chai <kchai@redhat.com>
if use "ceph-authtool keyring -n mon. --cap mon allow 'allow *'"
as err:
[root]# ceph-authtool keyring -n mon. --cap mon allow 'allow *'
ceph-authtool: unexpected 'allow *'
Excessive "allow" is the cause of the error
Signed-off-by huanwen ren <ren.huanwen@zte.com.cn>
ceph-disk.sh has its own setup/teardown, so update them also. so asok
files will have their parent directory when being created.
Signed-off-by: Kefu Chai <kchai@redhat.com>
These are interpreted as CEPH_OSD_UP in apply_incremental for legacy
reasons--we don't want them!
Fixes: http://tracker.ceph.com/issues/20751
Signed-off-by: Sage Weil <sage@redhat.com>
This avoids leaking the callback passed to remove_missing_object().
Use EAGAIN to signal that the work was not completed, so we don't have
to take the pg lock and check if the interval changed again in
PGBackend.
Fixes: http://tracker.ceph.com/issues/20747
Signed-off-by: Josh Durgin <jdurgin@redhat.com>
