mgr/telemetry: add stretch_mode information

Stretch Mode information helps us learn how deployments are done
for stretch clusters.

We add a basic_stretch_cluster collection fo the "basic" channel
for this purpose.

Fixes: https://tracker.ceph.com/issues/67812
Signed-off-by: Naveen Naidu <naveen.naidu@ibm.com>
(cherry picked from commit 6472b6b9f94affb96be341c9d595e543d734f30b)

Merge pull request #66357 from k0ste/wip-70542-squid

squid: os/bluestore: Disable invoking unittest_deferred

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

os/bluestore: Disable invoking unittest_deferred

There is no value in invoking unittest_deferred expect via
run_test_deferred.sh script.

Fixes: https://tracker.ceph.com/issues/68718
Signed-off-by: Adam Kupczyk <akupczyk@ibm.com>
(cherry picked from commit 163d8297bd566d25a6897f3b9be0d97f95c3c12b)

Merge pull request #66003 from cbodley/wip-73598-squid

squid: rgw: fix 'bucket rm --bypass-gc' for copied objects

Reviewed-by: Krunal Chheda <kchheda3@bloomberg.net>

Merge pull request #66152 from ivancich/wip-73747-squid

squid: rgw: fix `radosgw-admin object unlink ...`

Reviewed-by: Krunal Chheda <kchheda3@bloomberg.net>
Reviewed-by: Jane Zhu <jzhu116@bloomberg.net>

rgw: fix 'bucket rm --bypass-gc' for copied objects

the `--bypass-gc` argument to `radosgw-admin bucket rm` causes us to
call `RadosBucket::remove_bypass_gc()`, which loops over the tail
objects and removes each with `RGWRados::delete_raw_obj_aio()`

however, this was removing the objects with `cls_rgw_remove_obj()`,
which is for head objects, not tails. tail objects must be removed with
`cls_refcount_put()`, which preserves them until the last copy is
removed

rename `delete_raw_obj_aio()` to `delete_tail_obj_aio()` to clarify its
purpose

Fixes: https://tracker.ceph.com/issues/73348
Signed-off-by: Casey Bodley <cbodley@redhat.com>
(cherry picked from commit 1fba459071da9f7ec13defe2c666f0df8174c8da)

Merge pull request #66104 from shraddhaag/wip-73694-squid

squid: tasks/cbt_performance: Tolerate exceptions during performance data up…

tasks/cbt_performance: Tolerate exceptions during performance data updates

If an exception occurs during the POST request to update CBT performance,
log the error instead of failing the entire job. This ensures that
intermittent update failures do not block the main workflow.

Fixes: https://tracker.ceph.com/issues/68843
Signed-off-by: Nitzan Mordechai <nmordech@ibm.com>
(cherry picked from commit b47880f82de436776acab7ff13fb5e6496e49170)

Merge pull request #66033 from thuvh/bug-kafka-expected-sizes

squid: rgw-test: fix bug kafka unexpected keyword argument 'expected_sizes'

rgw: fix `radosgw-admin object unlink ...`

The unlink subcommand did not handle unsharded bucket indices
appropriately. These are when the number of shards listed in the
bucket instance object is 0. In that case there will actually be 1
shard.

When number of shards as 0 is passed into the function that maps
object names to shards, it returns -1. And that was not handled
properly. That is now fixed.

Signed-off-by: J. Eric Ivancich <ivancich@redhat.com>
(cherry picked from commit 9eeb71e28526df8ce2b920d4f50763a734264416)

squid: rgw-testing: fix unexpected keyword argument 'expected_sizes' for kafka test

Signed-off-by: Hoai-Thu Vuong <thuvh87@gmail.com>

Merge pull request #66242 from kshtsk/wip-73816-squid

squid: rgw: update keystone repo stable branch to 2024.2

Reviewed-by: J. Eric Ivancich <ivancich@redhat.com>

Merge pull request #66251 from kshtsk/wip-73582-squid

squid: qa/tasks/workunit: fix no module named 'pipes'

Merge pull request #65922 from guits/wip-73514-squid

squid: ceph-volume: use udev data instead of LVM subprocess in get_devices()

qa/tasks/workunit: fix no module named 'pipes'

Python 3.13 finally dropped the 'pipes' module,
replacing it with shlex.

Fixes: https://tracker.ceph.com/issues/73498
Signed-off-by: Kyr Shatskyy <kyrylo.shatskyy@clyso.com>
(cherry picked from commit da40484b9577aba21833e6b09c6129fe2d3c596c)

rgw: update keystone repo stable branch to 2024.2

The stable/2024.1 is gone from the github repo:

  https://github.com/openstack/keystone.git

Fixes: https://tracker.ceph.com/issues/73800
Signed-off-by: Kyr Shatskyy <kyrylo.shatskyy@clyso.com>
(cherry picked from commit 998e4299af417221ac8102fe46c46967e88f0da4)

Merge pull request #65584 from rzarzynski/wip-bug-72412-squid

squid: osd: stop scrub_purged_snaps() from ignoring osd_beacon_report_interval

Reviewed-by: Prashant D <pdhange@redhat.com>