Achieved by
1. osd superblock data is replicated in onode's OMAP - hence one can
   recover from that after onode's content is corrupted.
2. pg_num_history object gets full overwrite which eliminatess the need to
   merge with previous data (and hence reading corrupted data wouldn't
   kill OSD).

Signed-off-by: Igor Fedotov <ifedotov@croit.io>

Signed-off-by: Igor Fedotov <igor.fedotov@croit.io>

Commit 23db15d disabled pool snaps for the rados mksnap path. But
ceph osd pool mksnap was an alternate way that pool snaps could be
created.
This commit disables pool snaps via this alternate path as well.

NOTE:
Pool-level snaps and fs-level snaps can't co-exist since snap IDs are
likely to clash between the two different mechanisms and can result in
unintentional data loss when either of the snaps are deleted.

Fixes: https://tracker.ceph.com/issues/59552
Signed-off-by: Milind Changire <mchangir@redhat.com>

Signed-off-by: Milind Changire <mchangir@redhat.com>

Signed-off-by: Radoslaw Zarzynski <rzarzyns@redhat.com>

Fixes: https://tracker.ceph.com/issues/61865
Signed-off-by: Patrick Donnelly <pdonnell@redhat.com>

Signed-off-by: Yingxin Cheng <yingxin.cheng@intel.com>

Signed-off-by: Yingxin Cheng <yingxin.cheng@intel.com>

Signed-off-by: Yingxin Cheng <yingxin.cheng@intel.com>

Move client cores to the largest smp shards, and reserve core-0 to
reduce performance impacts.

Signed-off-by: Yingxin Cheng <yingxin.cheng@intel.com>

Signed-off-by: Yingxin Cheng <yingxin.cheng@intel.com>

Signed-off-by: Yingxin Cheng <yingxin.cheng@intel.com>

…with client

Signed-off-by: Yingxin Cheng <yingxin.cheng@intel.com>

…without client

Signed-off-by: Yingxin Cheng <yingxin.cheng@intel.com>

Signed-off-by: Yingxin Cheng <yingxin.cheng@intel.com>

Signed-off-by: Yingxin Cheng <yingxin.cheng@intel.com>

Signed-off-by: Yingxin Cheng <yingxin.cheng@intel.com>

Signed-off-by: Yingxin Cheng <yingxin.cheng@intel.com>

…ions

Signed-off-by: Yingxin Cheng <yingxin.cheng@intel.com>

…rver

Signed-off-by: Yingxin Cheng <yingxin.cheng@intel.com>

Signed-off-by: Yingxin Cheng <yingxin.cheng@intel.com>

…hput

Signed-off-by: Yingxin Cheng <yingxin.cheng@intel.com>

…cessary

Signed-off-by: Yingxin Cheng <yingxin.cheng@intel.com>

…d explicitly

Usually, we skip core 0 to avoid unwanted performance impacts, but that
may not be always the case. Add the configuration not to skip core 0.

Signed-off-by: Yingxin Cheng <yingxin.cheng@intel.com>

Signed-off-by: Zhang Song <zhangsong325@gmail.com>

Signed-off-by: Soumya Koduri <skoduri@redhat.com>

Signed-off-by: Vedansh Bhartia <vedanshbhartia@gmail.com>

Introduce the emdash character in
docs/rados/troubleshooting/troubleshooting-mon.rst.

Signed-off-by: Zac Dover <zac.dover@proton.me>

Edit docs/rados/operations/add-or-rm-osds.

Co-authored-by: Cole Mitchell <cole.mitchell.ceph@gmail.com>
Co-authored-by: Anthony D'Atri <anthony.datri@gmail.com>
Signed-off-by: Zac Dover <zac.dover@proton.me>

Signed-off-by: Frank S. Filz <ffilzlnx@mindspring.com>

Signed-off-by: Frank S. Filz <ffilzlnx@mindspring.com>

centos_latest was moved to centos9, we only schedule centos8
Crimson builds.
Use centos8 until Crimson will support 9.

Signed-off-by: Matan Breizman <mbreizma@redhat.com>

Signed-off-by: Matan Breizman <mbreizma@redhat.com>

Signed-off-by: Matan Breizman <mbreizma@redhat.com>

fs:upgrade tests use n/o/p release packages which aren't
available for centos 9.stream - switch using 8.stream instead.

Fixes: http://tracker.ceph.com/issues/62146
Signed-off-by: Venky Shankar <vshankar@redhat.com>

broken by bea9f4b
This commits fixes that regression.

Fixes: https://tracker.ceph.com/issues/62001

Signed-off-by: Guillaume Abrioux <gabrioux@ibm.com>

raw list shouldn't list lvm OSDs.
This confuses the generic activate process given that it relies
on the raw list to detect which kind of OSD is being activated.

Fixes: https://tracker.ceph.com/issues/62002

Signed-off-by: Guillaume Abrioux <gabrioux@ibm.com>

Fixes: cf5fdcd
Signed-off-by: Patrick Donnelly <pdonnell@redhat.com>

Introduce statfs for the CephFS REST API controller, we
can easily get statfs of the specified path by it, it returns
the used bytes, used files and used subdirs.

Fixes: https://tracker.ceph.com/issues/61883
Signed-off-by: Yite Gu <yitegu0@gmail.com>

Signed-off-by: Adam Kupczyk <akupczyk@ibm.com>

Fixes: https://tracker.ceph.com/issues/61929

Signed-off-by: Redouane Kachach <rkachach@redhat.com>

Signed-off-by: Redouane Kachach <rkachach@redhat.com>

Signed-off-by: Redouane Kachach <rkachach@redhat.com>

Signed-off-by: Redouane Kachach <rkachach@redhat.com>

There are cases in which we need low latency of an individual op, like
updating RBD image's object map which would block other ops for a
object not created yet. We need to make sure the rbd-object-map-update
like ops have as low latency as possible

Fixes: https://tracker.ceph.com/issues/62098
Signed-off-by: Xuehan Xu <xuxuehan@qianxin.com>

Fixes: https://tracker.ceph.com/issues/61868

Signed-off-by: Yuval Lifshitz <ylifshit@ibm.com>

Correct and improve ceph/README.md.

Signed-off-by: Zac Dover <zac.dover@proton.me>

Signed-off-by: Jos Collin <jcollin@redhat.com>

Bumps [@applitools/eyes-cypress](https://github.com/applitools/eyes.sdk.javascript1/tree/HEAD/js/packages/eyes-cypress) from 3.22.5 to 3.38.0.
- [Release notes](https://github.com/applitools/eyes.sdk.javascript1/releases)
- [Changelog](https://github.com/applitools/eyes.sdk.javascript1/blob/master/js/packages/eyes-cypress/CHANGELOG.md)
- [Commits](https://github.com/applitools/eyes.sdk.javascript1/commits/dotnet/images@3.38/js/packages/eyes-cypress)

---
updated-dependencies:
- dependency-name: "@applitools/eyes-cypress"
  dependency-type: direct:development
  update-type: version-update:semver-minor
...

Signed-off-by: dependabot[bot] <support@github.com>