container/Containerfile: purge .repo files with secrets before commit

ceph.repo had creds in it for download.ceph.com/prerelease.
Remove the .repo files we construct, since they're not necessary
once the container is built (no one should be dnf'ing anything
in the container).

Signed-off-by: Dan Mick <dan.mick@redhat.com>
(cherry picked from commit 08432ef54212aa2aba724102854082e76f0cd847)

container/Containerfile fix boolean IS_RELEASE usage

Existing syntax was just wrong.

Signed-off-by: Dan Mick <dan.mick@redhat.com>
(cherry picked from commit ce74f14de5042774b8386b0eff04497b517eaacc)

container/make-manifest-list.py: use LOCALMANIFEST throughout

Signed-off-by: Dan Mick <dan.mick@redhat.com>
(cherry picked from commit a3a9f188cabf714de2c39e310bccbe848ed66671)

containers/make-manifest.py: Add --promote

The promote function finds the latest image on the prerelease repo,
finds all its tags, and copies them all to the release repo.

Signed-off-by: Dan Mick <dan.mick@redhat.com>
(cherry picked from commit 4f050d7811fa6ec347c3414d5749e0f86480d7f4)

container/Containerfile, build.sh: Add PRERELEASE_USERNAME/PASSWORD

Used for accessing download.ceph.com

Signed-off-by: Dan Mick <dan.mick@redhat.com>
(cherry picked from commit 7374b9b1114da08f1fdb9dfcfe2ef37c2178c8f5)

container/build.sh:

- change repo path to prerelease
- remove image id arg for release push
- include 'v' in version name for release push

Signed-off-by: Dan Mick <dan.mick@redhat.com>
(cherry picked from commit 6c8e9984253a61c261cfe6e0981dda58895500ae)

container/make-manifest-list.py

- don't print command failure in worker; let the caller print them
  if desired (allow silent failure)
- allow for empty tags list
- look for CEPH_SHA1.  GIT_COMMIT was the sha1 of the ceph-container.git
  commit
- change default paths to prerelease
- add --dry-run to avoid final push
- rename 'HOST' to 'CONTAINER_HOST'
- Use ARCH_SPECIFIC_HOST instead of CONTAINER_HOST (which is used by podman)

Signed-off-by: Dan Mick <dan.mick@redhat.com>
(cherry picked from commit 44b2fc4d5e1fd909370018ca9ec210686f98cefa)

Merge pull request #60949 from zdover23/wip-doc-2024-12-04-backport-60930-to-squid

squid: doc/rados: fix sentences in health-checks (3 of x)

doc/rados: fix sentences in health-checks (3 of x)

Make sentences agree at the head of each section in
doc/rados/operations/health-checks.rst. The sentences were sometimes in
the imperative mood and sometimes in the declarative mood.

This commit edits the second third of
doc/rados/operations/health-checks.rst.

Note to (I hope soon) future Zac: There are a a couple of places near
the end of this file where the sentences are ungrammatical. Update these
in a separate PR (in isolation, so that the grammar and technical
accuracy of these sentences can be the primary focus of the reviewers).

Zac: cf. 000228

Signed-off-by: Zac Dover <zac.dover@proton.me>
(cherry picked from commit 97df4479d90b7f1b659b71ac172ff1094d58a50e)

Merge pull request #60931 from zdover23/wip-doc-2024-12-04-backport-60619-to-squid

squid: doc/rados: fix sentences in health-checks (2 of x)

doc/rados: fix sentences in health-checks (2 of x)

Make sentences agree at the head of each section in
doc/rados/operations/health-checks.rst. The sentences were sometimes in
the imperative mood and sometimes in the declarative mood.

This commit edits the second third of
doc/rados/operations/health-checks.rst.

Zac: cf. 000228

Signed-off-by: Zac Dover <zac.dover@proton.me>
(cherry picked from commit ee0ef765349ed8e170f8cd31c4fa0c7570105a96)

Merge pull request #60920 from zdover23/wip-doc-2024-12-03-backport-60917-to-squid

squid: doc/rados: make sentences agree in health-checks.rst

doc/rados: make sentences agree in health-checks.rst

Make sentences agree at the head of each section in
doc/rados/operations/health-checks.rst. The sentences were sometimes in
the imperative mood and sometimes in the declarative mood.

This commit edits the first third of
doc/rados/operations/health-checks.rst.

Zac: cf. 000228

Signed-off-by: Zac Dover <zac.dover@proton.me>
(cherry picked from commit aec87b9188a92d2722c7310e753462a054cb12c7)

Merge pull request #60880 from ShwetaBhosale1/fix_issue_69075_cephadm_bootstrap_message_with_default_image

Squid: cephadm: bootstrap should not have "This is a development version of cephadm" message

Reviewed-by: Adam King <adking@redhat.com>
Reviewed-by: Redouane Kachach <rkachach@ibm.com>

Merge pull request #60893 from zdover23/wip-doc-2024-12-01-backport-60882-to-squid

squid: os/bluestore: Improve documentation introduced by #57722

Merge pull request #60896 from zdover23/wip-doc-2024-12-01-backport-60892-to-squid

squid: doc/glossary.rst: add "Dashboard Plugin"

doc/glossary.rst: add "Dashboard Plugin"

Add an entry below the (Mimic-era and therefore outdated but
nonetheless historically important) Dashboard Plugin key word in the
glosssary, which before now had never been added to the glossary.

Signed-off-by: Zac Dover <zac.dover@proton.me>
(cherry picked from commit a1cec3de5704cedd7f7c8ce185144a96d72520d3)

os/bluestore: Improve documentation introduced by #57722

Signed-off-by: Anthony D'Atri <anthonyeleven@users.noreply.github.com>
(cherry picked from commit b6eb98cc10895e0cdc4bc147f6f83145c10cba41)

Merge pull request #60662 from zdover23/wip-doc-2024-11-07-backport-60642-to-squid

squid: doc/rgw/notification: clarified the notification_v2 behavior upon upg…

Reviewed-by: Anthony D'Atri <anthony.datri@gmail.com>

Merge pull request #60885 from zdover23/wip-doc-2024-11-29-backport-60884-to-squid

squid: doc/radosgw: update rgw_dns_name doc

doc/radosgw: update rgw_dns_name doc

Update doc/radosgw/s3/commons.rst with the changes made by Jiffin Tony
Thottan in https://github.com/ceph/ceph/pull/54524 and the suggestions
made in that same PR by Anthony D'Atri.

Explain how to set rgw_dns_name to a domain name in order to configure
access to virtual hosted buckets.

Signed-off-by: Zac Dover <zac.dover@proton.me>
(cherry picked from commit 45ff7e3365eaaa01ccb4ead71a14dd1dd9741c2f)

Merge pull request #60867 from zdover23/wip-doc-2024-11-28-backport-60860-to-squid

squid: doc/radosgw: Improve archive-sync-module.rst more

Merge pull request #60107 from cbodley/wip-68378-squid

squid: cmake: use ExternalProjects to build isa-l and isa-l_crypto libraries

Merge pull request #59147 from NitzanMordhai/wip-67473-squid

squid: suites: test should ignore osd_down warnings

cephadm: bootstap should not have "This is a development version of cephadm" message

Fixes: https://tracker.ceph.com/issues/69075
Signed-off-by: Shweta Bhosale <Shweta.Bhosale1@ibm.com>

Merge pull request #60864 from zdover23/wip-doc-2024-11-28-backport-59225-to-squid

squid: doc/glossary: add "flapping OSD"

Reviewed-by: Anthony D'Atri <anthony.datri@gmail.com>

doc/radosgw: Improve archive-sync-module.rst more

Signed-off-by: Anthony D'Atri <anthonyeleven@users.noreply.github.com>
(cherry picked from commit c24dc78484cbca8660c87742a6612c15935953c4)

doc/glossary: add "flapping OSD"

Add an entry for "Flapping OSD" to the glossary.

Signed-off-by: Zac Dover <zac.dover@proton.me>
(cherry picked from commit 1ca89e6ca3d9e0042702b7ac3af9a3e596330c1d)

Merge pull request #60616 from YiteGu/wip-68790-squid

squid: blk/KernelDevice: using join() to wait thread end is more safe

Merge pull request #60660 from sseshasa/wip-68856-squid

squid: common,osd: Use last valid OSD IOPS value if measured IOPS is unrealistic

Merge pull request #59372 from NitzanMordhai/wip-67646-squid

squid: mgr/rest: Trim  requests array and limit size

Merge pull request #60852 from zdover23/wip-doc-2024-11-27-backport-60846-to-squid

squid: doc/radosgw: Improve archive-sync-module.rst

doc/radosgw: Improve archive-sync-module.rst

Signed-off-by: Anthony D'Atri <anthony.datri@gmail.com>
(cherry picked from commit 303ce155a34fdf2cde63b948d204aca339dd5682)

Merge pull request #60823 from cbodley/wip-69025-squid

squid: rgw: RGWAccessKey::decode_json() preserves default value of 'active'

Reviewed-by: Shilpa Jagannath <smanjara@redhat.com>

Merge pull request #60824 from zdover23/wip-doc-2024-11-25-backport-60813-to-squid

squid: doc/rados/troubleshooting: Improve log-and-debug.rst

Merge pull request #60774 from yuvalif/wip-68984-squid

squid: RGW|Bucket notification: fix for v2 topics rgw-admin list operation

Reviewed-by: Adam Emerson <aemerson@redhat.com>

Merge pull request #60792 from smanjara/wip-69003-squid

squid: rgw/multisite: metadata polling event based on unmodified mdlog_marker

Reviewed-by: Adam Emerson <aemerson@redhat.com>

qa/rgw: the rgw/verify suite runs java tests last

a band-aid for https://tracker.ceph.com/issues/68991 to make sure all
other tests run before the failing java tests

Signed-off-by: Casey Bodley <cbodley@redhat.com>
(cherry picked from commit 49e76c23de1b5c1f1e84e50a6d8207329ce4093e)

doc/rados/troubleshooting: Improve log-and-debug.rst

Signed-off-by: Anthony D'Atri <anthonyeleven@users.noreply.github.com>
(cherry picked from commit 44c9519cc3eb013104c52014c302ba71ba806f8f)

rgw: RGWAccessKey::decode_json() preserves default value of 'active'

squid added the new "bool active" member that defaults to true, but
RGWAccessKey::decode_json() sets it to false when "active" isn't present

this is an issue for multisite when a squid zone replicates user
metadata from a master zone running reef

Fixes: https://tracker.ceph.com/issues/68985
Signed-off-by: Casey Bodley <cbodley@redhat.com>
(cherry picked from commit 8ba6c2de830129ec0533652a3cb33391b584006d)

Merge pull request #60737 from zdover23/wip-doc-2024-11-15-backport-60703-to-squid

squid: doc/rados: correct "full ratio" note

Reviewed-by: Anthony D'Atri <anthony.datri@gmail.com>

Merge pull request #60809 from zdover23/wip-doc-2024-11-24-backport-60804-to-squid-test-2

squid: doc/cephadm: Clarify "Deploying a new Cluster"

doc/cephadm: Clarify "Deploying a new Cluster"

Change the title of the section "Deploying a new Ceph cluster" to "Using
cephadm to Deploy a New Ceph Cluster". This is part of the initiative to
separate package-related documentation from container-based
documenation.

Signed-off-by: Zac Dover <zac.dover@proton.me>
(cherry picked from commit 8cf77d03a1f3ecb73931ddf96d13ef098aa61c1a)

Merge pull request #60751 from idryomov/wip-68743-squid

squid: Make mon addrs consistent with mon info

Reviewed-by: Laura Flores <lflores@redhat.com>

Merge pull request #60718 from ljflores/wip-68910-squid

squid: mgr/balancer: optimize 'balancer status detail'

Merge pull request #58914 from ljflores/wip-67245-squid

squid: mon: Remove any pg_upmap_primary mapping during remove a pool

rgw/multisite: in order to sleep between mdlog polling events, we check if the mdlog_marker is not modified by comparing
mdlog_marker and max_marker. but max_marker is exposed to changes from RGWReadMDLogEntriesCR, and if there is a race
coming from mdlog trimming which could render max_marker empty, then its comparison with mdlog polling can be incorrect.
To fix this, we now save the previous mdlog marker and compare with the updated mdlog marker.

Signed-off-by: Shilpa Jagannath <smanjara@redhat.com>
(cherry picked from commit bae10f22e8b008088f0861db2785d8c0ba7ee506)

Merge pull request #60766 from zdover23/wip-doc-2024-11-19-backport-60765-to-squid

squid: doc/start: update os-recommendations.rst

Reviewed-by: Anthony D'Atri <anthony.datri@gmail.com>

rgw/notification: fix segmentation fault and topic listing logic

- Fixed a segmentation fault caused by a null bucket pointer in RGWPSListTopicsOp::execute()
- Corrected logic to use get_topics_v2 when supported, with fallback otherwise

Fixes: https://tracker.ceph.com/issues/68756
Signed-off-by: Oshrey Avraham <oshrey16@gmail.com>
(cherry picked from commit d27dab240d568a705ff45e2231be650673279ebe)

RGW|Bucket notification: fix for v2 topics rgw-admin list operation

Signed-off-by: Ali Masarwa <amasarwa@redhat.com>
(cherry picked from commit 575a19d2a81e959d21f64d1a9ed39a5db3b92957)

doc/start: update os-recommendations.rst

Remove information about the operating systems that support Ceph's
official container images from the "Platforms" table in
doc/start/os-recommendations.rst and add that information to the (new)
table that shows the operating systems that support Ceph's official
container images.

Credit for this change should go to Enrico Bocchi, who noticed a
discrepancy that motivated it.

Signed-off-by: Zac Dover <zac.dover@proton.me>
(cherry picked from commit c2f1fac3b94e946b253bc64e579429660cbc2d59)

Merge pull request #60164 from cbodley/wip-68425-squid

squid: cls/user: reset stats only returns marker when truncated

Reviewed-by: Adam Emerson <aemerson@redhat.com>

Merge pull request #59302 from yuvalif/wip-67574-squid

squid: rgw/notifications: fixing radosgw-admin notification json

Reviewed-by: Adam Emerson <aemerson@redhat.com>

Merge pull request #59120 from k0ste/wip-64499-squid

squid: rgw: add s3select usage to log usage

Reviewed-by: Adam Emerson <aemerson@redhat.com>

Merge pull request #59931 from adk3798/wip-68136-squid

squid: mgr/cephadm: add command to expose systemd units of all daemons

Reviewed-by: John Mulligan <jmulligan@redhat.com>

Merge pull request #60734 from guits/wip-68961-squid

squid: ceph-volume: fix dmcrypt activation regression

Merge pull request #60130 from cbodley/wip-68399-squid

squid: rgw: decrypt multipart get part when encrypted

Reviewed-by: Adam Emerson <aemerson@redhat.com>

Merge pull request #59754 from yuvalif/wip-68047-squid

squid: rgw/kafka: refactor topic creation to avoid rd_kafka_topic_name()

Reviewed-by: Adam Emerson <aemerson@redhat.com>

Merge pull request #59678 from jzhu116-bloomberg/wip-67959-squid

squid: rgw/multipart: use cls_version to avoid racing between part upload and multipart complete

Reviewed-by: Adam Emerson <aemerson@redhat.com>

Merge pull request #59123 from k0ste/wip-67099-squid

squid: [RGW] Fix the handling of HEAD requests that do not comply with RFC standards

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

Make mon addrs consistent with mon info

For legacy librbd clients, sanitize_mons will panic if addrs are not
consistent with mon_info_t.

Signed-off-by: shenjiatong <yshxxsjt715@163.com>
fixes: https://tracker.ceph.com/issues/68743
(cherry picked from commit 14b09519739f3b2dc715647fd0f04dd22c17aeb3)

doc/rados: correct "full ratio" note

Correct a note that directed users not to add an OSD after the cluster
has reached its "full ratio". The note now says "Do not let your cluster
reach its full ratio before adding an OSD."

Hat tip: Oskar Berggren

Fixes: https://tracker.ceph.com/issues/68900
Co-authored-by: Oskar Berggren <oskar.berggren@gmail.com>
Signed-off-by: Zac Dover <zac.dover@proton.me>
(cherry picked from commit f1a2637c79a15c26a769661dd72ca68d766b2f0d)

Merge pull request #60671 from zdover23/wip-doc-2024-11-08-backport-56771-to-squid

squid: doc: Document which options are disabled by mClock.

Reviewed-by: Anthony D'Atri <anthony.datri@gmail.com>

ceph-volume: fix dmcrypt activation regression

`ceph.with_tpm` tag is actually a `str`, which means that
`bool('0')` will return `True`.

Fixes: https://tracker.ceph.com/issues/68944
Signed-off-by: Guillaume Abrioux <gabrioux@ibm.com>
(cherry picked from commit 3cf00cde820ec838e4fbda1799d2ce336c9e2d86)

Merge pull request #59948 from aaSharma14/wip-68203-squid

squid: mgr/cephadm: use host address while updating rgw zone endpoints

Reviewed-by: Adam King <adking@redhat.com>

Merge pull request #59945 from aaSharma14/wip-68201-squid

squid: mgr/rgwam: use realm/zonegroup/zone method arguments for period update

Reviewed-by: Adam King <adking@redhat.com>

Merge pull request #59935 from adk3798/squid-cephadm-no-exception-cert-store-get

squid: mgr/cephadm: add --no-exception-when-missing flag to cert-store cert/key get

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

Merge pull request #59934 from adk3798/squid-cephadm-iscsi-teuth-extension

squid: qa/cephadm: extend iscsi teuth test

Reviewed-by: John Mulligan <jmulligan@redhat.com>

Merge pull request #59933 from adk3798/wip-68159-squid

squid: cephadm: Support Docker Live Restore

Reviewed-by: John Mulligan <jmulligan@redhat.com>

Merge pull request #59930 from adk3798/wip-68114-squid

squid: qa/task: update alertmanager endpoints version

Reviewed-by: John Mulligan <jmulligan@redhat.com>

Merge pull request #59929 from adk3798/wip-68019-squid

squid: cephadm: emit warning if daemon's image is not to be used

Reviewed-by: John Mulligan <jmulligan@redhat.com>

Merge pull request #59710 from adk3798/cephadm-squid-ssh-keepalive-configurable

squid: mgr/cephadm: make ssh keepalive settings configurable

Reviewed-by: John Mulligan <jmulligan@redhat.com>

Merge pull request #59472 from adk3798/wip-67462-squid

squid: qa/upgrade: fix checks to make sure upgrade is still in progress

Reviewed-by: John Mulligan <jmulligan@redhat.com>

Merge pull request #58711 from rhcs-dashboard/bump-monitoring-stack-squid

squid: mgr/cephadm: bump monitoring stacks version

Reviewed-by: John Mulligan <jmulligan@redhat.com>

Merge pull request #58515 from adk3798/squid-cephadm-ingress-fix-ipv6-binding

squid: mgr/cephadm/services/ingress Fix HAProxy to listen on IPv4 and IPv6

Reviewed-by: John Mulligan <jmulligan@redhat.com>

Merge pull request #58381 from adk3798/squid-cephadm-haproxy-bugged-config

squid: cephadm/services/ingress: fixed keepalived config bug

Reviewed-by: John Mulligan <jmulligan@redhat.com>

Merge pull request #60466 from rhcs-dashboard/wip-68686-squid

squid: mgr/dashboard: fix bucket get for s3 account owned bucket

Reviewed-by: Afreen Misbah <afreen@ibm.com>

Merge pull request #60620 from YiteGu/wip-67094-squid

squid: mgr/dashboard: add cephfs rename REST API

Reviewed-by: Afreen Misbah <afreen@ibm.com>

Merge pull request #60632 from rhcs-dashboard/wip-68844-squid

squid: mgr/dashboard: remove cherrypy_backports.py

doc: document balancer configurable "update_pg_upmap_activity"

Signed-off-by: Laura Flores <lflores@ibm.com>
(cherry picked from commit c4274c0c3a608eeb0b1183c8a96ef3a314bf797c)

mgr/balancer: tie `update_pg_upmap_activity` to a configurable

This addition gives users the option of enabling/disabling this feature.

They can do so by running:
ceph config set mgr mgr/balancer/update_pg_upmap_activity <True|False>

The feature is off by default.

Signed-off-by: Laura Flores <lflores@ibm.com>
(cherry picked from commit 62c9f152e31affe3e19eb50e917cbeb10624f8fd)

mgr/balancer: optimize 'balancer status detail'

Before, we were updating the balancer status by
iterating through all pg upmap entires. This was
affecting the loading time of other mgr modules
on clusters with a large number of pgs (600+).

This can be optimized by simply pulling from
the incremental.

Fixes: https://tracker.ceph.com/issues/68657
Signed-off-by: Laura Flores <lflores@ibm.com>
(cherry picked from commit 31c498da96fa1acc1700515edcd544dead86dab9)

Merge pull request #60707 from zdover23/wip-doc-2024-11-12-backport-60705-to-squid

squid: doc/radosgw: s/Poliicy/Policy/

Merge pull request #60708 from zdover23/wip-doc-2024-11-12-backport-60704-to-squid

squid: doc/start: fix "are are" typo

doc/start: fix "are are" typo

Fix typo reading "They are are single-threaded".
s/are are/are/

Fixes: https://tracker.ceph.com/issues/68901
Signed-off-by: Zac Dover <zac.dover@proton.me>
(cherry picked from commit 22010719d7edf2a6cf7de7b115ed797e7942ec47)

doc/radosgw: s/Poliicy/Policy/

Correct "AttachGroupPoliicy" to "AttachGroupPolicy".

Fixes: https://tracker.ceph.com/issues/68902
Signed-off-by: Zac Dover <zac.dover@proton.me>
(cherry picked from commit 961aef1bb4e59912ac1c8827a392c18b912c71f0)

Merge pull request #60698 from zdover23/wip-doc-2024-11-12-backport-60529-to-squid

squid: doc/start: separate package chart from container chart

doc/start: separate package chart from container chart

Separate the packages-and-containers chart into two charts:
   (1) a chart that shows which OSes Ceph builds packages for
   (2) a chart that shows which OSes support Ceph's containers

Signed-off-by: Zac Dover <zac.dover@proton.me>
(cherry picked from commit e5ff1916e3d5e2cc3f47b4cea2762730dfd335c0)

doc/mgr/restful: update max_request config

Signed-off-by: Nitzan Mordechai <nmordech@redhat.com>
(cherry picked from commit dd7e8bbde6df5efdf4706279e5f8b8281ebaee16)

Merge pull request #60679 from zdover23/wip-doc-2024-11-09-backport-60605-to-squid

squid: doc/cephadm: clean "Adv. OSD Service Specs"

doc/cephadm: clean "Adv. OSD Service Specs"

Clean the English grammar and the .RST in the section "Advanced OSD
Service Specifications" in the file doc/cephadm/services/osd.rst.

Signed-off-by: Zac Dover <zac.dover@proton.me>
(cherry picked from commit f6b56f9ee5d9f0b54dcaf1832d7a977895eb9c94)

Merge pull request #59035 from sseshasa/wip-67355-squid

squid: osd/scheduler: add mclock queue length perfcounter

Merge pull request #58785 from sseshasa/wip-67108-squid

squid: qa/suites/rados/verify/validater: increase heartbeat grace timeout

doc: Document which options are disabled by mClock.

Not only in the mClock docs, but also in the reference
of the options that are disabled.

Otherwise users are bound to miss it, and surprised
why their options are ignored or reset.

Signed-off-by: Niklas Hambüchen <mail@nh2.me>
(cherry picked from commit 790afd277730f24801a3b59c08d2dce0f2a9659f)

doc/rgw/notification: clarified the notification_v2 behavior upon upgrade

Fixes: https://tracker.ceph.com/issues/68858
Signed-off-by: Yuval Lifshitz <ylifshit@ibm.com>
(cherry picked from commit fccf7fd25e57772e91975b1b2038b8cecf68c440)

qa/task: update alertmanager endpoints version

the v1 endpoint is deprecated and v2 is suggested to use.

Fixes: https://tracker.ceph.com/issues/67183
Signed-off-by: Nizamudeen A <nia@redhat.com>
(cherry picked from commit ac998b58487f6f9176b7f934ed61ae57e59c2ee5)

mgr/dashboard: adapt dashboard e2e for grafana

Signed-off-by: Nizamudeen A <nia@redhat.com>
(cherry picked from commit a4374ab53806f353e49f3f818421b2630aad6f0b)

mgr/cephadm: bump monitoring stacks versions

Signed-off-by: Nizamudeen A <nia@redhat.com>
(cherry picked from commit c719d8ac2d7ca70b6c33d9f5893b99360029c947)

common,osd: Use last valid OSD IOPS value if measured IOPS is unrealistic

The OSD's IOPS capacity is used by the mClock scheduler to determine the
quantum of bandwidth allocation for the various operations on the OSD.
Prior to this commit, maybe_override_max_osd_capacity_for_qos() only
checked if the measured IOPS capacity exceeded the higher threshold defined
by 'osd_mclock_iops_capacity_threshold_[hdd|ssd]' and if so fallback to the
last valid or the default IOPS capacity as defined by
osd_mclock_max_capacity_iops_[hdd|ssd].

It's quite possible that the reported IOPS is unrealistically low. This
could be due to transient factors on the underlying device or it could
indicate bad health of the device. Either way, the safer option would be
to fallback to the last valid or the default IOPS setting for that OSD in
order to avoid cluster performance (slow or stalled ops) issues down the
line.

Therefore, to handle this case, the commit introduces additional config
options viz.,
 - osd_mclock_iops_capacity_low_threshold_hdd - set to 50 IOPS and
 - osd_mclock_iops_capacity_low_threshold_ssd - set to 1000 IOPS

If the measured IOPS capacity doesn't fall within the low and high
threshold range, the default or the last valid IOPS capacity is used.
The existing cluster log warning is suitably modified to convey the
reason.

Additionally, for a couple of valgrind related teuthology tests, the
cluster warning is added to the ignorelist since the reported IOPS can
be very low due to slowness.

Fixes: https://tracker.ceph.com/issues/67421
Signed-off-by: Sridhar Seshasayee <sseshasa@redhat.com>
(cherry picked from commit da4b85c55a15f49b241f3fc44dda2263b42dc637)

Merge pull request #60644 from zdover23/wip-doc-2024-11-07-backport-60643-to-squid

squid: doc/cephadm: link to "host pattern" matching sect