Fixes: http://tracker.ceph.com/issues/17635
Before accessing json entity, need to check that iterator is valid.
If there is no entry return appropriate error code.
Signed-off-by: Yehuda Sadeh <yehuda@redhat.com>
When ceph-* drops drops privileges via setuid, core dumps are no longer
generated because its DUMPABLE flag is cleared. We have to manually
turn that back on.
From prctl(2):
Normally, this flag is set to 1. However, it is reset to the current value contained in the file /proc/sys/fs/suid_dumpable (which by default has the value 0), in the fol‐
lowing circumstances:
* The process's effective user or group ID is changed.
* The process's filesystem user or group ID is changed (see credentials(7)).
* The process executes (execve(2)) a set-user-ID or set-group-ID program, or a program that has capabilities (see capabilities(7)).
Fixes: http://tracker.ceph.com/issues/17650
Signed-off-by: Patrick Donnelly <pdonnell@redhat.com>
Passing by value tries to copy from a null and segfaults.
Fixes: http://tracker.ceph.com/issues/17636
Signed-off-by: John Spray <john.spray@redhat.com>
prior to this change, librbd tests link against librados_api_obj and
librados_objs, but since librados_a offers all the goodies, we can
just link against it instead.
Signed-off-by: Kefu Chai <kchai@redhat.com>
so tests referencing internal symbols can use librados_a instead of
librados to avoid violating ODR (one definition rule).
Fixes: http://tracker.ceph.com/issues/17260
Signed-off-by: Kefu Chai <kchai@redhat.com>
Callers expect run_in_thread() to return a tuple, but it
returns an int. When interrupt the function it would raise
an iterable exception.
This patch fix the bug, and it would just output a message
like: "Error EINTR: Interrupted!"
Signed-off-by: Javeme <javaloveme@gmail.com>
* group some paragraphs into sections, so we can link to them via its
html anchor.
* remove unrelated contents.
* add cross-refs.
Signed-off-by: Kefu Chai <kchai@redhat.com>
in the `RGWBucketShardFullSyncCR::operate`, inc_marker will assigned with remote bilog's max_marker.
but the sync_status's inc_marker cant be assigned.so the next step inc sync will always sync
from null log,which means at beginning log.
Fixes: http://tracker.ceph.com/issues/17624
Signed-off-by: Zengran Zhang <zhangzengran@h3c.com>
The 0 sized segment can arise depending on how
the client uploads the object in the first place.
The cpu loop then happens during a swift `GET'.
Signed-off-by: Marcus Watts <mwatts@redhat.com>
Sometimes users know that particular data pool PGs
have been damaged, and they would like to scan
their files to work out which ones might have
been affected.
Fixes: http://tracker.ceph.com/issues/17249
Signed-off-by: John Spray <john.spray@redhat.com>
Previously, if you used ceph_create_with_context, then
it would try and destroy your context when you called
ceph_release. Bad behaviour: caller created the
context, caller should decide when to destroy it.
Signed-off-by: John Spray <john.spray@redhat.com>
Changes scoping of json section, and push the --object param in so that
an object can be specified as a filter.
Signed-off-by: Yehuda Sadeh <yehuda@redhat.com>
