Merge pull request #40127 from neha-ojha/wip-49761-octopus

octopus: pybind/mgr/balancer/module.py: assign weight-sets to all buckets before balancing

Reviewed-by: Deepika Upadhyay <dupadhya@redhat.com>

Merge pull request #40278 from singuliere/wip-48596-octopus

octopus: test: cancelling both noscrub *and* nodeep-scrub

Reviewed-by: Neha Ojha <nojha@redhat.com>

Merge pull request #40277 from singuliere/wip-49009-octopus

octopus: osd: fix potential null pointer dereference when sending ping

Reviewed-by: Kefu Chai <kchai@redhat.com>

Merge pull request #40276 from singuliere/wip-49527-octopus

octopus: mon/OSDMonitor: fix safety/idempotency of {set,rm}-device-class

Reviewed-by: xie xingguo <xie.xingguo@zte.com.cn>
Reviewed-by: Sage Weil <sage@redhat.com>

Merge pull request #40275 from singuliere/wip-49730-octopus

octopus: debian/ceph-common.postinst: do not chown cephadm log dirs

Reviewed-by: Kefu Chai <kchai@redhat.com>

Merge pull request #40274 from singuliere/wip-49795-octopus

octopus: osd: propagate base pool application_metadata to tiers

Reviewed-by: Kefu Chai <kchai@redhat.com>

Merge pull request #40013 from mfoliveira/wip-49681-octopus

octopus: osd: add osd_fast_shutdown_notify_mon option (default false)

Reviewed-by: Josh Durgin <jdurgin@redhat.com>

Merge pull request #39970 from singuliere/wip-48985-octopus

octopus: osd/OSDMap: An empty bucket or OSD is not an error

Reviewed-by: Brad Hubbard <bhubbard@redhat.com>

Merge pull request #39716 from sebastian-philipp/octopus-backport-39373

octopus: mgr/rook: Add timezone info

Reviewed-by: Varsha Rao <varao@redhat.com>

Merge pull request #39487 from ChristinaMeno/wip-49297-octopus

octopus: ceph.spec.in: Enable tcmalloc on IBM Power and Z

Reviewed-by: Nathan Cutler <ncutler@suse.com>

Merge pull request #40298 from tchaikov/octopus-48381

octopus: mon/ConfigMap: fix stray option leak

Reviewed-by: Sage Weil <sage@redhat.com>

Merge pull request #40534 from tchaikov/octopus-pr-40505

octopus: mgr/PyModule: put mgr_module_path before Py_GetPath()

Reviewed-by: Neha Ojha <nojha@redhat.com>

mgr/PyModule: put mgr_module_path before Py_GetPath()

pip comes with _vendor/progress. so there is chance to import the vendored
version of "progress" module instead of the "progress" mgr module, and
fail to import the latter.

in this change, the order of paths are rearranged so the configured
`mgr_module_path` is put before the return value of `Py_GetPath()`.

Fixes: https://tracker.ceph.com/issues/50058
Signed-off-by: Kefu Chai <kchai@redhat.com>
(cherry picked from commit 8638f526a9d04c3dfd758073980d709165070336)

Conflicts:
src/mgr/PyModule.cc: trivial resolution

Merge pull request #40492 from tchaikov/octopus-flake8

octopus: pybind/mgr/dashboard: bump flake8 to 3.9.0

Reviewed-by: Alfonso Martínez <almartin@redhat.com>

pybind/mgr/dashboard: bump up requests to 2.25.1

request 2.20 is not compatible with urllib3 v1.25.2 and up. this causes
trouble of incompatibility with other python modules. for instance, we
now have following error:

ERROR: pip's dependency resolver does not currently take into account
all the packages that are installed. This behaviour is the source of the
following dependency conflicts.
botocore 1.20.14 requires urllib3<1.27,>=1.25.4, but you have urllib3
1.24.3 which is incompatible.

see also https://github.com/psf/requests/pull/5092

Signed-off-by: Kefu Chai <kchai@redhat.com>
(cherry picked from commit 8bf07cd1408d0f407ef5e32717bfea159188670d)

mgr/dashboard: update pylint to 2.6.0

* Update pylint to 2.6.0.
* Fix pylint issues.

Fixes: https://tracker.ceph.com/issues/47647
Signed-off-by: Volker Theile <vtheile@suse.com>
s

(cherry picked from commit 298c91958a41674a928d53f010b20f174f16d68f)

Conflicts:
src/pybind/mgr/dashboard/requirements-lint.txt
src/pybind/mgr/dashboard/services/ceph_service.py
src/pybind/mgr/dashboard/services/ganesha.py
src/pybind/mgr/dashboard/services/rgw_client.py
src/pybind/mgr/dashboard/tests/test_access_control.py
src/pybind/mgr/dashboard/tests/test_ganesha.py
src/pybind/mgr/dashboard/tests/test_iscsi.py
src/pybind/mgr/dashboard/tests/test_rgw.py
src/pybind/mgr/dashboard/tests/test_settings.py

admin/build-doc: stop passing --use-feature=2020-resolver to pip

to silence the warning of

WARNING: --use-feature=2020-resolver no longer has any effect, since it is now the default dependency resolver in pip. This will become an error in pip 21.0.

Signed-off-by: Kefu Chai <kchai@redhat.com>
(cherry picked from commit 5cb9d5458745046aaea58cf4af50579925fbb1d0)

Conflicts:
admin/build-doc: trivial resolution

pybind/mgr/dashboard: bump flake8 to 3.9.0

to address the failure of

ERROR: Cannot install -r requirements-lint.txt (line 2) and -r requirements-lint.txt (line 8) because these package versions have conflicting dependencies.

The conflict is caused by:
    flake8 3.8.4 depends on pycodestyle<2.7.0 and >=2.6.0a1
    autopep8 1.5.6 depends on pycodestyle>=2.7.0

To fix this you could try to:
1. loosen the range of package versions you've specified
2. remove package versions to allow pip attempt to solve the dependency conflict

also, loosen the version of pytest:

The conflict is caused by:
    The user requested pytest<4
    The user requested pytest<4
    pytest-cov 2.11.1 depends on pytest>=4.6

To fix this you could try to:
1. loosen the range of package versions you've specified
2. remove package versions to allow pip attempt to solve the dependency
   conflict

Signed-off-by: Kefu Chai <kchai@redhat.com>
(cherry picked from commit 152964ca360293d9accd18f435efcd66d145063e)

Merge pull request #39914 from mgfritch/octopus-backport-37764-39739

octopus: cephadm: run containers using `--init`

Reviewed-by: Deepika Upadhyay <dupadhya@redhat.com>

Merge pull request #40138 from neha-ojha/wip-49402-octopus

octopus: qa/suites/rados/singletone: whitelist MON_DOWN when injecting msgr errors

Reviewed-by: Yuri Weinstein <yweinste@redhat.com>

Merge pull request #40009 from mgfritch/octopus-backport-39825

octopus: mgr/cephadm: alias rgw-nfs -> nfs

Reviewed-by: Sebastian Wagner <sebastian.wagner@suse.com>

Merge pull request #39940 from smithfarm/wip-49663-octopus

octopus: src/global/signal_handler.h: fix preprocessor logic for alpine

Reviewed-by: Kefu Chai <kchai@redhat.com>

Merge pull request #39922 from smithfarm/wip-49636-octopus

octopus: mgr/telemetry: check if 'ident' channel is active

Reviewed-by: Yaarit Hatuka <yaarithatuka@gmail.com>

Merge pull request #39919 from smithfarm/wip-49530-octopus

octopus: crush/CrushWrapper: update shadow trees on update_item()

Reviewed-by: Neha Ojha <nojha@redhat.com>

Merge pull request #39884 from singuliere/wip-49386-octopus

octopus: os/bluestore/BlueFS: use iterator_impl::copy instead of bufferlist::c_str() to avoid bufferlist rebuild

Reviewed-by: Igor Fedotov <ifedotov@suse.com>

Merge pull request #37972 from callithea/wip-48131-octopus

octopus: mgr/dashboard: additional logging to SMART data retrieval

Reviewed-by: Patrick Seidensal <pnawracay@suse.com>

Merge pull request #40406 from tchaikov/octopus-pr-40400

octopus: run-make-check.sh: let ctest generate XML output

Reviewed-by: Josh Durgin <jdurgin@redhat.com>

run-make-check.sh: let ctest generate XML output

to enable XUnit plugin of jenkins to consume the ctest output and
publish it in the dashboard, we need to

* let ctest generate XML output instead of plain text output
* do not fail the test if any test case fails. this allows the publisher
  to do its job by checking the XML output.
* prevent ctest from compressing the output. see
  https://issues.jenkins.io/browse/JENKINS-21737

Signed-off-by: Kefu Chai <kchai@redhat.com>
(cherry picked from commit 48ba39987d3958531589d7969750ea749e6a6d30)

Merge pull request #39360 from kamoltat/wip-octupus-del-period-arg

octopus:qa/tasks/mgr/test_progress: fix wait_until_equal

Reviewed-by: Josh Durgin <jdurgin@redhat.com>
Reviewed-by: Neha Ojha <nojha@redhat.com>

Merge pull request #40225 from kamoltat/wip-fix-39289-incomplete-backport

octopus: qa/tasks/mgr/test_progress.py:  remove calling of _osd_in_out_completed_events_count()

Reviewed-by: Josh Durgin <jdurgin@redhat.com>
Reviewed-by: Neha Ojha <nojha@redhat.com>

Merge pull request #40001 from rhcs-dashboard/wip-49703-octopus

octopus: mgr/dashboard: fix dashboard instance ssl certificate functionality

Reviewed-by: Nizamudeen A <nia@redhat.com>

qa/tasks/mgr/test_progress: fix wait_until_equal

Octopus ceph_test_case doesn't have period arg
so remove that in wait_until_equal. Also increase
time to wait for complete events by using RECOVERY_PERIOD
instead of EVENT_CREATION_PERIOD

Not needed in masters because only octopus and nautilus
doesn't have a period argument in qa/tasks/mgr/test_progress.py
wait_until_equals() function

Fixes: https://tracker.ceph.com/issues/48824
Signed-off-by: Kamoltat <ksirivad@redhat.com>

PendingReleaseNotes: document option osd_fast_shutdown_notify_mon

Let's add the ``osd_fast_shutdown_notify_mon`` option to PendingReleaseNotes
so it is documented.

Signed-off-by: Mauricio Faria de Oliveira <mfo@canonical.com>
(cherry picked from commit 7f5aaef5d4585d74535192192c56549bd023bc1f)

 Conflicts:
PendingReleaseNotes
 - Move snippet into new 15.2.11 section.

osd: add osd_fast_shutdown_notify_mon option (default false)

The osd_fast_shutdown option may cause the cluster log to receive
too many entries of 'osd.X reported immediately failed by osd.Y',
depending on cluster scale.

This might be an issue for LMA stacks/tools that check ceph logs
for failed lines, and then require additional logic to filter on
an intended OSD (fast) shutdown; might not be an option/possible,
and require an admin to analyze.

So, add osd_fast_shutdown_notify_mon option for OSD to also tell
the monitor it is shutting down (done in slow/non-fast shutdown)
under osd_fast_shutdown.

This introduces minimal delay (the ack from the mon is required
to prevent the messages), and addresses the cluster log issue.
Note: the osd_mon_shutdown_timeout option can be used to control
the maximum amount of time waiting for the monitor ack to arrive.

Fixes: http://tracker.ceph.com/issues/46978
Signed-off-by: Mauricio Faria de Oliveira <mfo@canonical.com>
(cherry picked from commit c75734729764868c5c501722fc8de08dac9ebd4a)

mon/ConfigMap: fix stray option leak

The const Option* needs to remain alive only until the next clear().  Keep
the reference in ConfigMap and clean it up then.

Signed-off-by: Sage Weil <sage@newdream.net>
(cherry picked from commit 9397a46aec88e287d56a6286ed4319f65d9c1f31)

Fixes: https://tracker.ceph.com/issues/48381
Conflicts:
src/mon/ConfigMap.h: trivial resolution

Merge pull request #40286 from tchaikov/octopus-pr-40272

octopus: install-deps.sh: remove existing ceph-libboost of different version

Reviewed-by: David Galloway <dgallowa@redhat.com>

install-deps.sh: remove existing ceph-libboost of different version

we install different versions of precompiled ceph-libboost packages
for different branches when building and testing them on ubuntu test
nodes. for instance,

- nautilus, octopus: v1.72
- pacific: v1.73

they share the same set of test nodes. and these ceph-libboost packages
conflict with each other, because they install files to the same places.

in order to avoid the confliction, we should uninstall existing packages
before installing a different version of ceph-libboost packages.

ceph-libboost${version}-dev is a package providing the shared headers of
boost library, so, in this change we check if it is installed before
returning or removing the existing packages.

Signed-off-by: Kefu Chai <kchai@redhat.com>
(cherry picked from commit 939b147a55192c21e98d21cb380d0ec0b2ca84d5)

Conflicts:
install-deps.sh: use 1.72

rpm: re-disable SUSE lttng build on z390x

This partially reverts 2b1e646f7aade3135a98c505111ac7ebef5e93a6 which
mistakenly changed a line inside an "%if 0%{?suse_version}" conditional.

Fixes: 2b1e646f7aade3135a98c505111ac7ebef5e93a6
Signed-off-by: Nathan Cutler <ncutler@suse.com>
(cherry picked from commit ffd202a08619fc535df593eb41c0769577a1586a)

test: cancelling both noscrub *and* nodeep-scrub

as part of osd-scrub-test.sh.

Signed-off-by: Ronen Friedman <rfriedma@redhat.com>
(cherry picked from commit 43b1129030823817e0b7a21c85de5d3da841510a)

osd: fix potential null pointer dereference when sending ping

Fixes: https://tracker.ceph.com/issues/48821
Signed-off-by: Mykola Golub <mgolub@suse.com>
(cherry picked from commit 86576b09973b857ec2fe8195069e21812992db26)

mon/OSDMonitor: fix safety/idempotency of {set,rm}-device-class

If the command is resent (e.g., due to network reconnect), the second
instance may find that the pending crush map already has the changes
and not wait for it to commit.

Note that the stderr message will be misleading in this case; that is a
problem with most of our mon commands.  :(

Fixes: https://tracker.ceph.com/issues/49212
Signed-off-by: Sage Weil <sage@newdream.net>
(cherry picked from commit db6c8f9ab32a7bc0dc8bca94f79812bfb9e7b123)

 Conflicts:
src/mon/OSDMonitor.cc
only because mon. in master is mon-> in octopus

debian/ceph-common.postinst: do not chown cephadm log dirs

The container uid/gid is different than the debian uid/gid (because the
container is centos-based and we got a different uid/gid allocation there).

Fixes: https://tracker.ceph.com/issues/49677
Signed-off-by: Sage Weil <sage@newdream.net>
(cherry picked from commit b89ffdcae51303f185e1b423a948df353497250f)

osd: propagate base pool application_metadata to tiers

If there is application metadata on the base pool, it should be mirrored
to any other tiers in the set.  This aligns with the fact that the
'ceph osd pool application ...' commands refuse to operate on a non-base
pool.

This fixes problems with accessing tiers (e.g., cache tiers) when the
cephx cap is written in terms of application metadata.

Fixes: https://tracker.ceph.com/issues/49788
Signed-off-by: Sage Weil <sage@newdream.net>
(cherry picked from commit 8bcb281a7d91e6daa842aed4ffb416d764dc6bf1)

Merge pull request #40211 from idryomov/wip-krbd-blkroset-5.12-octopus

octopus: qa: krbd_blkroset.t: update for separate hw and user read-only flags

Reviewed-by: Jason Dillaman <dillaman@redhat.com>

qa/tasks/mgr/test_progress.py: remove _osd_in_out_completed_events_count()

delete the part where  _osd_in_out_completed_events_count()
was called in test_osd_cannot_recover() and revert to initial
state of the function since we don't need to use this function
in octopus. Also delete a duplicate of _osd_in_out_events_count().
This must be added by mistake in #39289 as well.

No need to fix for the backport in Nautilus: #38173
since the bugs are occured by adding additional code to
the cherry-pick specifically for Octopus.

fixes: https://tracker.ceph.com/issues/49891

Signed-off-by: Kamoltat <ksirivad@redhat.com>

Merge pull request #40065 from idryomov/wip-krbd-xfstests-fixes-octopus

octopus: qa/suites/krbd: address recent issues caused by newer kernels

Reviewed-by: Jason Dillaman <dillaman@redhat.com>

Merge pull request #40029 from yuvalif/wip-yuval-fix-48461

octopus: rgw/notification: trigger notifications on changes from any user

Reviewed-by: Casey Bodley <cbodley@redhat.com>

Merge pull request #39968 from dillaman/wip-49438-octopus

octopus: krbd: make sure the device node is accessible after the mapping

Reviewed-by: Mykola Golub <mgolub@mirantis.com>
Reviewed-by: Ilya Dryomov <idryomov@redhat.com>

Merge pull request #39967 from dillaman/wip-49417-octopus

octopus: rbd-mirror: delay update snapshot mirror image state

Reviewed-by: Mykola Golub <mgolub@mirantis.com>

Merge pull request #39966 from dillaman/wip-49257-octopus

octopus: librbd: refuse to release exclusive lock when removing

Reviewed-by: Mykola Golub <mgolub@mirantis.com>

Merge pull request #39965 from dillaman/wip-49181-octopus

octopus: test/rbd-mirror: fix broken ceph_test_rbd_mirror_random_write

Reviewed-by: Mykola Golub <mgolub@mirantis.com>

Merge pull request #39964 from dillaman/wip-48677-octopus

octopus: librbd/api: avoid retrieving more than max mirror image info records

Reviewed-by: Mykola Golub <mgolub@mirantis.com>

Merge pull request #39962 from dillaman/wip-48674-octopus

octopus: librbd: explicitly disable readahead for writearound cache

Reviewed-by: Mykola Golub <mgolub@mirantis.com>

Merge pull request #39961 from dillaman/wip-48561-octopus

octopus: rbd-mirror: bad state and crashes in snapshot-based mirroring

Reviewed-by: Mykola Golub <mgolub@mirantis.com>

Merge pull request #39958 from smithfarm/wip-49199-octopus

octopus: librbd/io: conditionally disable move optimization

Reviewed-by: Jason Dillaman <dillaman@redhat.com>

Merge pull request #39954 from smithfarm/wip-49361-octopus

octopus: rgw: fix sts get_session_token duration check failed

Reviewed-by: Pritha Srivastava <prsrivas@redhat.com>

Merge pull request #39766 from ivancich/wip-49193-octopus

octopus: rgw: allow rgw-orphan-list to handle intermediate files w/ binary data

Reviewed-by: Ali Maredia <amaredia@redhat.com>

Merge pull request #39785 from ivancich/wip-49186-octopus

octopus: rgw: tooling to locate rgw objects with missing rados components

Reviewed-by: Casey Bodley <cbodley@redhat.com>

Merge pull request #39758 from href/wip-49070-octopus

octopus: rgw: add MD5 in forward_request

Reviewed-by: Casey Bodley <cbodley@redhat.com>

Merge pull request #39746 from pponnuvel/wip-49538-octopus

octopus: rgw: catch non int exception

Reviewed-by: Daniel Gryniewicz <dang@redhat.com>

Merge pull request #39611 from smithfarm/wip-49380-octopus

octopus: rgw: multisite: fix single-part-MPU object etag misidentify problem

Reviewed-by: Casey Bodley <cbodley@redhat.com>

Merge pull request #39935 from gerald-yang/octopus-49641

octopus: common: Fix assertion when disabling and re-enabling clog_to_monitors

Reviewed-by: Kefu Chai <kchai@redhat.com>

Merge pull request #39899 from singuliere/wip-49039-octopus

octopus: blk/kernel: fix io_uring got (4) Interrupted system call

Reviewed-by: Kefu Chai <kchai@redhat.com>

Merge pull request #39717 from sebastian-philipp/octopus-backport-39385

octopus: cephadm: allow redeploy of daemons in error state if container running

Reviewed-by: Adam King adking@redhat.com

Merge pull request #39887 from tchaikov/octopus-49533

octopus: mgr: relax osd ok-to-stop condition on degraded pgs

Reviewed-by: Sage Weil <sage@redhat.com>
Reviewed-by: David Zafman <dzafman@redhat.com>

Merge pull request #39804 from sebastian-philipp/octopus-backport-39770

octopus: cephadm: add docker.service dependency in systemd units

Reviewed-by: Sage Weil <sage@redhat.com>

Merge pull request #39578 from neha-ojha/wip-49320-octopus

octopus: qa/*/thrash_cache_writeback_proxy_none.yaml: disable writeback overlay tests

Reviewed-by: Josh Durgin <jdurgin@redhat.com>

Merge pull request #39343 from smithfarm/wip-48904-octopus

octopus: common/ipaddr: Allow binding on lo

Reviewed-by: Kefu Chai <kchai@redhat.com>
Reviewed-by: Neha Ojha <nojha@redhat.com>

Merge pull request #39341 from smithfarm/wip-48898-octopus

octopus: mgr: fix deadlock in ActivePyModules::get_osdmap()

Reviewed-by: Kefu Chai <kchai@redhat.com>

qa: krbd_blkroset.t: update for separate hw and user read-only flags

Since kernel 5.12, hardware read-only state and user read-only
policy (BLKROGET/SET ioctls) are tracked separately in the block
layer.  As the purpose of our ->set_read_only() method was exactly
that, it was removed.

As a side effect, BLKROSET no longer returns EROFS on an attempt
to make a read-only mapping read-write with "blockdev --setrw".
The policy gets updated, but the device remains read-only as before
because the hardware (== mapping) state is controlled by the driver.

Fixes: https://tracker.ceph.com/issues/49858
Signed-off-by: Ilya Dryomov <idryomov@gmail.com>
(cherry picked from commit d72fca26edcff49d203ed6fb940e0cf331e943dd)

krbd: check device node accessibility only if we actually mapped

Fix a braino that came with commit f6854ac65d2a ("krbd: make sure the
device node is accessible after the mapping").

Signed-off-by: Ilya Dryomov <idryomov@gmail.com>
(cherry picked from commit 8330c9fa4e27204c768777afe45af0eeb273c835)

15.2.10

Merge pull request #40155 from rhcs-dashboard/wip-49272-octopus

octopus: mgr/dashboard: delete EOF when reading passwords from file

Reviewed-by: Ernesto Puerta <epuertat@redhat.com>
Reviewed-by: Yuri Weinstein <yweinste@redhat.com>

mgr/dashboard: delete EOF when reading passwords from file

Signed-off-by: Alfonso Martínez <almartin@redhat.com>
(cherry picked from commit caeadf1397db00c6b7ba218b1910508099802e39)

qa/suites/rados/singletone: whitelist MON_DOWN when injecting msgr errors

add a 'none.yaml' facet which doesn't whitelist so we hopefully capture
other causes of MON_DOWN.

This fixes at least one source of https://tracker.ceph.com/issues/45441

Fixes: https://tracker.ceph.com/issues/45441
Signed-off-by: Sage Weil <sage@newdream.net>
(cherry picked from commit 3d2eb0d76ae396bec5f527e99bfecd9fa5223bbd)

pybind/mgr/balancer/module.py: assign weight-sets to all buckets before balancing

Add an additional check to make sure that the choose_args section has the same
number of buckets as the crushmap. If not, ensure that
get_compat_weight_set_weights assigns weight-sets to all buckets.

Without this change, if we end up with an orig_ws, which has fewer buckets
than the crushmap, the mgr will crash due a KeyError in do_crush_compat().

Fixes: https://tracker.ceph.com/issues/49576
Signed-off-by: Neha Ojha <nojha@redhat.com>
(cherry picked from commit 42c01f0c3932f7a9157797013e5db2e03d9c1fd9)

 Conflicts:
src/pybind/mgr/balancer/module.py - trivial resolution

Merge pull request #39701 from ifed01/wip-ifed-fix-huge-bluefs-oct

octopus: os/bluestore: fix huge reads/writes at BlueFS

Reviewed-by: Adam Kupczyk <akupczyk@redhat.com>

rgw/notification: trigger notifications on changes from any user

any user authorized to make changes to a bucket may trigger
notifications defined on that bucket.
manual test procedure of the fix is described here:
https://gist.github.com/yuvalif/39c183aa0f74d286ecef7844268817df

Fixes: https://tracker.ceph.com/issues/48461
Signed-off-by: Yuval Lifshitz <ylifshit@redhat.com>
(cherry picked from commit 658993efc16498c2106108fe407f6d44040f349d)

Conflics:
src/test/rgw/rgw_multi/tests_ps.py
manual changes were made to the tests so they can pass. this does not affect teuthology runs

Merge pull request #40074 from ideepika/wip-fix-ignorelist

octopus: qa/suites/upgrade: s/whitelist/ignorelist for octopus specific tests

Reviewed-by: Neha Ojha <nojha@redhat.com>
Reviewed-by: Yuri Weinstein <yweinste@redhat.com>

Merge pull request #39754 from ifed01/wip-ifed-bluefs-zero-read-retry-octa

octopus: os/bluestore: Add option to check BlueFS reads

Reviewed-by: Adam Kupczyk <akupczyk@redhat.com>

mgr/cephadm: add CEPHADM_STRAY_DAEMON unittest

Fixes: https://tracker.ceph.com/issues/49573
Signed-off-by: Michael Fritch <mfritch@suse.com>
(cherry picked from commit 88dc55266958d77ffa1da34a3d20c039cb249ac5)

mgr/cephadm: alias rgw-nfs -> nfs

RGW will register a daemon_type `rgw-nfs` which needs to map to
the corresponding cephadm nfs daemon to avoid stray daemon warnings

Fixes: https://tracker.ceph.com/issues/49573
Signed-off-by: Michael Fritch <mfritch@suse.com>
(cherry picked from commit 0f237223accbea49cf9792400a657cbf696a5d67)

Merge pull request #39872 from rhcs-dashboard/wip-48190-octopus

octopus: mgr/dashboard: add ssl verify option for prometheus and alert manager

Reviewed-by: Alfonso Martínez <almartin@redhat.com>
Reviewed-by: Nizamudeen A <nia@redhat.com>

Merge pull request #39868 from rhcs-dashboard/wip-49083-octopus

octopus: mgr/dashboard: Fix missing root path of each session for CephFS

Reviewed-by: Alfonso Martínez <almartin@redhat.com>
Reviewed-by: Tatjana Dehler <tdehler@suse.com>
Reviewed-by: Tatjana Dehler <tdehler@suse.com>

Merge pull request #39854 from rhcs-dashboard/wip-49324-octopus

octopus: mgr/dashboard: fix MTU Mismatch alert

Reviewed-by: Nizamudeen A <nia@redhat.com>
Reviewed-by: Aashish Sharma <aasharma@redhat.com>

Merge pull request #39852 from rhcs-dashboard/wip-49599-octopus

octopus: mgr/dashboard: report mgr fsid

Reviewed-by: Laura Paduano <lpaduano@suse.com>
Reviewed-by: Nizamudeen A <nia@redhat.com>

Merge pull request #39436 from rhcs-dashboard/wip-48654-octopus

octopus: mgr/dashboard: CLI commands: read passwords from file

Reviewed-by: Avan Thakkar <athakkar@redhat.com>
Reviewed-by: Laura Paduano <lpaduano@suse.com>

qa/suites/upgrade: s/whitelist/ignorelist for octopus only

some upgrade tests are only present for octopus and not for master and
hence we missed updating the ignorelist terminology for those cases.

Signed-off-by: Deepika Upadhyay <dupadhya@redhat.com>

Merge pull request #39885 from smithfarm/wip-49031-octopus

octopus: cmake: boost>=1.74 adds BOOST_ASIO_USE_TS_EXECUTOR_AS_DEFAULT to radosgw

Reviewed-by: Kefu Chai <kchai@redhat.com>

Merge pull request #39867 from trociny/wip-49452-octopus

octopus: rbd-mirror: reset update_status_task pointer in timer thread

Reviewed-by: Jason Dillaman <dillaman@redhat.com>

Merge pull request #39866 from trociny/wip-49454-octopus

octopus: librbd: use on-disk image name when storing mirror snapshot state

Reviewed-by: Jason Dillaman <dillaman@redhat.com>

Merge pull request #39864 from trociny/wip-49399-octopus

octopus: librbd: allow disabling journaling for snapshot based mirroring image

Reviewed-by: Jason Dillaman <dillaman@redhat.com>

Merge pull request #39863 from trociny/wip-49335-octopus

octopus: mgr/rbd_support: mirror snapshot schedule should skip non-primary images

Reviewed-by: Jason Dillaman <dillaman@redhat.com>

Merge pull request #39862 from trociny/wip-49263-octopus

octopus: librbd: don't log error if get mirror status fails due to mirroring disabled

Reviewed-by: Jason Dillaman <dillaman@redhat.com>

Merge pull request #39704 from smithfarm/wip-49397-octopus

octopus: qa/suites/rados/dashboard: whitelist TELEMETRY_CHANGED

Reviewed-by: Laura Paduano <lpaduano@suse.com>
Reviewed-by: Yaarit Hatuka <yaarithatuka@gmail.com>
Reviewed-by: Nizamudeen A <nia@redhat.com>

mgr: relax osd ok-to-stop condition on degraded pgs

Right now, the "ok-to-stop" condition is relatively rigorous, it allows
stopping an osd only if no PG on it is non-active or degraded. But there
are situations in which an OSD is part of a degraded pg and the pg still
still have > min_size complete replicas after the OSD is stopped.

In 9750061d5d4236aaba156d60790e0b8bcd7cfb64, we changed from considering
just acting to using avail_no_missing (OSDs that have no missing objects).
When the projected pg_acting is constructed this way, we can safely compare
to min_size... even for a PG marked degraded.

Fixes: https://tracker.ceph.com/issues/49392
Signed-off-by: Xuehan Xu <xxhdx1985126@gmail.com>
(cherry picked from commit 2f28fc58eb63e20c2a869f5407516cf8044dbc41)

qa/suites/krbd: bump scratch image size to 15G

Allow generic/038 and generic/048 to run.

Signed-off-by: Ilya Dryomov <idryomov@gmail.com>
(cherry picked from commit 37d56e13544fa06b88196b1a4d6edd1b3425325e)

qa/suites/krbd: exclude ext4/002

ext4/002 exercises obsolete EXT4_EOFBLOCKS_FL feature that was removed
in kernel 5.7 and therefore always fails.

Signed-off-by: Ilya Dryomov <idryomov@gmail.com>
(cherry picked from commit d2bdf0ac43a35f755ea70103f497625d882a4592)

qa/tasks/rbd: fix xfstests_branch/url example

- don't mention xfstests_branch together with xfstests_url as
  xfstests_branch gets superseded by xfstests_url
- make xfstests_url example more clear

Signed-off-by: Ilya Dryomov <idryomov@gmail.com>
(cherry picked from commit ec36d501ed5f875d26d1362ec097187697170a4d)

mgr/dashboard: fix dashboard instance ssl certificate functionality

Fixes: https://tracker.ceph.com/issues/46542
Signed-off-by: Avan Thakkar <athakkar@redhat.com>
(cherry picked from commit 0a88c62e76f80a4ad9c01a91161b77ac6f073ed4)

 Conflicts:
src/pybind/mgr/mgr_module.py
      - Adopting the master branch changes.