librbd: use actual monitor addresses when creating a peer bootstrap token
Reviewed-by: Mykola Golub <mgolub@suse.com>
Reviewed-by: Christopher Hoffman <choffman@redhat.com>
Use `main` instead of `master` in the workunit scripts for the
Windows Teuthology job.
Signed-off-by: Ionut Balutoiu <ibalutoiu@cloudbasesolutions.com>
Relying on mon_host config option is fragile, as the user may confuse
v1 and v2 addresses, group them incorrectly, etc. Get mon_host value
only as a fallback.
Fixes: https://tracker.ceph.com/issues/57317
Signed-off-by: Ilya Dryomov <idryomov@gmail.com>
The shell commands we test beforehand can create the
/var/lib/ceph/00000000-0000-0000-0000-0000deadbeef directory
and that directory being present will block bootstrap as
it will think a cluster with this fsid alreayd exists
Fixes: https://tracker.ceph.com/issues/57290
Signed-off-by: Adam King <adking@redhat.com>
mds/client: fail the request if the peer MDS doesn't support getvxattr op
Reviewed-by: Venky Shankar <vshankar@redhat.com>
Reviewed-by: Patrick Donnelly <pdonnell@redhat.com>
Reviewed-by: Milind Changire <mchangir@redhat.com>
When the osd is full, if the snapshot has metadata set, it
can't be removed as user metadata can't be removed when osd
is full. This patch provides a way to remove the snapshot
with 'force' option while keeping the corresponding metadata
which gets removed on subvolume discover when it finds space.
Fixes: https://tracker.ceph.com/issues/55976
Signed-off-by: Kotresh HR <khiremat@redhat.com>
Existing object-dedup command always creates temporary object and
perform set-chunk to make the input object manifest---this is because
current tier-flush, which is called by object-dedup, works only if
the target object is manifest, otherwise it just return 0.
So, object-dedup may not work if the target is already manifest.
Moreover, it causes unnecessary overhead if the target object is not manifest.
To solve these, this commit makes existing tier-flush to
change object's state to manifest implicitly.
With this change, the only required operation when calling object-dedup
is a tier-flush and object-dedup can work with the manifest object.
Signed-off-by: Myoungwon Oh <myoungwon.oh@samsung.com>
Missed this one, it seems the test was still passing until
recently but I guess the "master" builds are now gone
Signed-off-by: Adam King <adking@redhat.com>
Commit 96f05a7956b3 ("rbd: delay determination of default pool name")
broke "rbd perf image iostat" and "rbd perf image iotop" GLOBAL_POOL_KEY
support (the ability to blend all rbd pools together into a single
view).
Fixes: https://tracker.ceph.com/issues/56561
Signed-off-by: Ilya Dryomov <idryomov@gmail.com>
1. If data or metadata pool is already in-use by filesystem
then it is not allowed to reuse the same pool for another
filesystems.
2. Test is failing because above(1) restrictions/checks comes
before checking erasure-code pools. Hence test is failing
and not finding expected error string in output.
3. Proposed fix checks newly added error string instead of
'erasure-code'.
4. Also adding new tests to verify string 'erasure-code'
by passing --force option so that check for pools reuse(1)
will be skipped and check for 'erasure-code' will be hit.
Fixes: https://tracker.ceph.com/issues/56384
Signed-off-by: Nikhilkumar Shelke <nshelke@redhat.com>
Establishing a watch on rbd_mirroring object and skipping rescanning
image mirror snapshots on periodic refresh unless rbd_mirroring object
gets notified in the interim is flawed. rbd_mirroring object is
notified when mirroring is enabled or disabled on some image (including
when the image is removed), but it is not notified when images are
promoted or demoted. However, load_pool_images() discards images that
are not primary at the time of the scan. If the image is promoted
later, no snapshots are created even if the schedule is in place. This
happens regardless of whether the schedule is added before or after the
promotion.
This effectively reverts commit 69259c8d37 ("mgr/rbd_support: make
mirror_snapshot_schedule rescan only updated pools"). An alternative
fix could be to stop discarding non-primary images (i.e. drop
if not info['primary']:
continue
check added in commit d39eb283c5 ("mgr/rbd_support: mirror snapshot
schedule should skip non-primary images")), but that would clutter the
queue and therefore "rbd mirror snapshot schedule status" output with
bogus entries. Performing a rescan roughly every 60 seconds should be
manageable: currently it amounts to a single mirror_image_status_list
request, followed by mirror_image_get, get_snapcontext and snapshot_get
requests for each snapshot-based mirroring enabled image and concluded
by a single dir_list request. Among these, per-image get_snapcontext
and snapshot_get requests are necessary for determining primaryness.
Fixes: https://tracker.ceph.com/issues/53914
Signed-off-by: Ilya Dryomov <idryomov@gmail.com>
Modify Iscsi tcmu-runner container to be run demonized in the same
systemd slice as all other ceph processes
Signed-off-by: Teoman ONAY <tonay@redhat.com>
