]> git-server-git.apps.pok.os.sepia.ceph.com Git - ceph.git/log
ceph.git
2 days agocrimson/os/seastore/lba: merge the whole mapping into the delta
Xuehan Xu [Thu, 4 Jun 2026 05:59:36 +0000 (13:59 +0800)]
crimson/os/seastore/lba: merge the whole mapping into the delta

Signed-off-by: Xuehan Xu <xuxuehan@qianxin.com>
2 days agocrimson/os/seastore/extent_pinboard: don't add extents under promotion
Xuehan Xu [Thu, 4 Jun 2026 05:54:01 +0000 (13:54 +0800)]
crimson/os/seastore/extent_pinboard: don't add extents under promotion
to cache

This is to avoid an under-promotion extent from being added back to and
evicted from the cache again before the promotion ends.

Signed-off-by: Xuehan Xu <xuxuehan@qianxin.com>
2 days agocrimson/os/seastore/lba,TM: remove shadows when trimming dirty non-data
Xuehan Xu [Wed, 3 Jun 2026 09:36:42 +0000 (17:36 +0800)]
crimson/os/seastore/lba,TM: remove shadows when trimming dirty non-data
extents

Signed-off-by: Xuehan Xu <xuxuehan@qianxin.com>
2 days agocrimson/os/seastore/cached_extent: let pending cached extents hold pointers to the...
Xuehan Xu [Thu, 21 May 2026 13:52:46 +0000 (21:52 +0800)]
crimson/os/seastore/cached_extent: let pending cached extents hold pointers to the transactions

This is necessary for the promote transaction to find the transactions
that are retiring the extents promoted

Signed-off-by: Xuehan Xu <xuxuehan@qianxin.com>
2 days agocrimson/os/seastore: demote/promote background processes are also
Xuehan Xu [Mon, 18 May 2026 09:55:35 +0000 (17:55 +0800)]
crimson/os/seastore: demote/promote background processes are also
rewrite transactions

So we need to merge the extents promoted/demoted with their counterparts
in client transactions, just like trim/cleaner transaction do.

The difference between demote/promote transactions and trim/cleaner
transactions is that the former also involves remapping extents while the
latter only involves rewriting and mutating extents.

With the logical bucket cache in place, all rewrite transactions need to
handle shadow extents carafully, too.

Signed-off-by: Xuehan Xu <xuxuehan@qianxin.com>
2 days agocrimson/os/seastore/transaction_manager: wait for the demotion when the
Xuehan Xu [Sat, 16 May 2026 05:40:35 +0000 (13:40 +0800)]
crimson/os/seastore/transaction_manager: wait for the demotion when the
rbm main device is not enough

Signed-off-by: Xuehan Xu <xuxuehan@qianxin.com>
2 days agocrimson/os/seastore: drop backref when all backends are RBMs
Xuehan Xu [Wed, 13 May 2026 11:51:58 +0000 (19:51 +0800)]
crimson/os/seastore: drop backref when all backends are RBMs

Signed-off-by: Xuehan Xu <xuxuehan@qianxin.com>
2 days agoqa/config/crimson_qa_overrides.yaml: add test workload for LBC
Xuehan Xu [Wed, 13 May 2026 07:13:28 +0000 (15:13 +0800)]
qa/config/crimson_qa_overrides.yaml: add test workload for LBC

Signed-off-by: Xuehan Xu <xuxuehan@qianxin.com>
2 days agocrimson/os/seastore/cache: count promote/demote related metrics
Xuehan Xu [Mon, 11 May 2026 06:27:25 +0000 (14:27 +0800)]
crimson/os/seastore/cache: count promote/demote related metrics

Signed-off-by: Xuehan Xu <xuxuehan@qianxin.com>
2 days agotest/crimson/seastore/test_transaction_manager: the scatter_allocation case shouldn...
Xuehan Xu [Sat, 9 May 2026 02:15:24 +0000 (10:15 +0800)]
test/crimson/seastore/test_transaction_manager: the scatter_allocation case shouldn't submit the trans

In scatter_allocation, an extent is split into several
small ones, and the last allocation is supposed to fail,
in which case, the small extents that belong to the same
allocation wouldn't be attached to the parent lba leaf
node. Sumitting the transaction would trigger assertions,
and is not consistent with the behavior of nornal crimson
OSDs.

Signed-off-by: Xuehan Xu <xuxuehan@qianxin.com>
2 days agocrimson/os/seastore: add test workload to promote/evict aggressively
Zhang Song [Wed, 3 Sep 2025 08:10:54 +0000 (16:10 +0800)]
crimson/os/seastore: add test workload to promote/evict aggressively

Signed-off-by: Zhang Song <zhangsong02@qianxin.com>
Signed-off-by: Xuehan Xu <xuxuehan@qianxin.com>
2 days agocrimson/os/seastore/random_block_manager: try to allocate consecutive rbm space when...
Xuehan Xu [Wed, 3 Sep 2025 08:10:28 +0000 (16:10 +0800)]
crimson/os/seastore/random_block_manager: try to allocate consecutive rbm space when rewriting extents

Add the address of the last allocation as the hint to the current round
of allocation in rbm

Signed-off-by: Xuehan Xu <xuxuehan@qianxin.com>
2 days agocrimson/os/seastore/cache: add hit ratio metric
Zhang Song [Wed, 3 Sep 2025 08:06:54 +0000 (16:06 +0800)]
crimson/os/seastore/cache: add hit ratio metric

Signed-off-by: Zhang Song <zhangsong02@qianxin.com>
Signed-off-by: Xuehan Xu <xuxuehan@qianxin.com>
2 days agocrimson/os/seastore: make cold rbm cleaner use different metric prefix
Zhang Song [Fri, 8 Aug 2025 08:58:22 +0000 (16:58 +0800)]
crimson/os/seastore: make cold rbm cleaner use different metric prefix

Signed-off-by: Zhang Song <zhangsong02@qianxin.com>
Signed-off-by: Xuehan Xu <xuxuehan@qianxin.com>
2 days agocrimson/os/seastore: allow RBM backend as hot tier
Zhang Song [Fri, 8 Aug 2025 08:36:27 +0000 (16:36 +0800)]
crimson/os/seastore: allow RBM backend as hot tier

Signed-off-by: Zhang Song <zhangsong02@qianxin.com>
Signed-off-by: Xuehan Xu <xuxuehan@qianxin.com>
2 days agocrimson/os/seastore: update logical bucket cache when processing
Zhang Song [Fri, 8 Aug 2025 07:50:52 +0000 (15:50 +0800)]
crimson/os/seastore: update logical bucket cache when processing
read/write transactions

All client accessed extents are to be tracked by logical bucket cache
before eviction

Signed-off-by: Zhang Song <zhangsong02@qianxin.com>
Signed-off-by: Xuehan Xu <xuxuehan@qianxin.com>
2 days agocrimson/os/seastore: add write through policy
Zhang Song [Wed, 3 Sep 2025 08:05:38 +0000 (16:05 +0800)]
crimson/os/seastore: add write through policy

Write extents larger than some threshold to the cold tier directly

Signed-off-by: Zhang Song <zhangsong02@qianxin.com>
Signed-off-by: Xuehan Xu <xuxuehan@qianxin.com>
2 days agocrimson/osd: create cold devices when mkfs
Zhang Song [Tue, 26 Mar 2024 03:52:15 +0000 (11:52 +0800)]
crimson/osd: create cold devices when mkfs

Signed-off-by: Zhang Song <zhangsong02@qianxin.com>
Signed-off-by: Xuehan Xu <xuxuehan@qianxin.com>
3 days agocrimson/os/seastore: throttle bandwidth to secondary devices
Zhang Song [Thu, 31 Jul 2025 08:22:28 +0000 (16:22 +0800)]
crimson/os/seastore: throttle bandwidth to secondary devices

Signed-off-by: Zhang Song <zhangsong02@qianxin.com>
Signed-off-by: Xuehan Xu <xuxuehan@qianxin.com>
3 days agocrimson/os/seastore: introduce RotationalDevice
Zhang Song [Thu, 31 Jul 2025 08:16:56 +0000 (16:16 +0800)]
crimson/os/seastore: introduce RotationalDevice

Signed-off-by: Zhang Song <zhangsong02@qianxin.com>
Signed-off-by: Xuehan Xu <xuxuehan@qianxin.com>
3 days agocrimson/os/seastore: set backend type explicitly
Zhang Song [Thu, 31 Jul 2025 08:06:13 +0000 (16:06 +0800)]
crimson/os/seastore: set backend type explicitly

Signed-off-by: Zhang Song <zhangsong02@qianxin.com>
Signed-off-by: Xuehan Xu <xuxuehan@qianxin.com>
3 days agocrimson/os/seastore/EPM/BackgroundProcess: add promote and demote process
Zhang Song [Thu, 7 Aug 2025 11:22:58 +0000 (19:22 +0800)]
crimson/os/seastore/EPM/BackgroundProcess: add promote and demote process

Signed-off-by: Zhang Song <zhangsong02@qianxin.com>
Signed-off-by: Xuehan Xu <xuxuehan@qianxin.com>
3 days agocrimson/os/seastore/EPM: initialize logical bucket and pinboard
Zhang Song [Thu, 31 Jul 2025 03:07:44 +0000 (11:07 +0800)]
crimson/os/seastore/EPM: initialize logical bucket and pinboard

Signed-off-by: Zhang Song <zhangsong02@qianxin.com>
Signed-off-by: Xuehan Xu <xuxuehan@qianxin.com>
3 days agocrimson/os/seastore: don't add shadow extent to cache
Zhang Song [Wed, 30 Jul 2025 07:39:55 +0000 (15:39 +0800)]
crimson/os/seastore: don't add shadow extent to cache

Signed-off-by: Zhang Song <zhangsong02@qianxin.com>
Signed-off-by: Xuehan Xu <xuxuehan@qianxin.com>
3 days agocrimson/os/seastore: avoid evicting promoted extent to the cold tier
Zhang Song [Wed, 3 Sep 2025 08:04:12 +0000 (16:04 +0800)]
crimson/os/seastore: avoid evicting promoted extent to the cold tier

Signed-off-by: Zhang Song <zhangsong02@qianxin.com>
Signed-off-by: Xuehan Xu <xuxuehan@qianxin.com>
3 days agocrimson/os/seastore: consolidate parameters on extent allocation path
Zhang Song [Wed, 30 Jul 2025 06:38:51 +0000 (14:38 +0800)]
crimson/os/seastore: consolidate parameters on extent allocation path

Signed-off-by: Zhang Song <zhangsong02@qianxin.com>
Signed-off-by: Xuehan Xu <xuxuehan@qianxin.com>
3 days agocrimson/os/seastore: implement demote region
Zhang Song [Wed, 3 Sep 2025 08:02:29 +0000 (16:02 +0800)]
crimson/os/seastore: implement demote region

Signed-off-by: Zhang Song <zhangsong02@qianxin.com>
Signed-off-by: Xuehan Xu <xuxuehan@qianxin.com>
3 days agocrimson/os/seastore: implement promote extent
Zhang Song [Wed, 3 Sep 2025 08:02:01 +0000 (16:02 +0800)]
crimson/os/seastore: implement promote extent

Signed-off-by: Zhang Song <zhangsong02@qianxin.com>
Signed-off-by: Xuehan Xu <xuxuehan@qianxin.com>
3 days agocrimson/os/seastore/lba: introduce shadow paddr
Zhang Song [Wed, 3 Sep 2025 08:01:35 +0000 (16:01 +0800)]
crimson/os/seastore/lba: introduce shadow paddr

Signed-off-by: Zhang Song <zhangsong02@qianxin.com>
Signed-off-by: Xuehan Xu <xuxuehan@qianxin.com>
3 days agocrimson/os/seastore: introduce logical bucket and demote process
Zhang Song [Thu, 3 Jul 2025 10:09:17 +0000 (18:09 +0800)]
crimson/os/seastore: introduce logical bucket and demote process

Signed-off-by: Zhang Song <zhangsong02@qianxin.com>
Signed-off-by: Xuehan Xu <xuxuehan@qianxin.com>
3 days agocrimson/os/seastore: support promote extents purged from pinboard
Zhang Song [Thu, 3 Jul 2025 07:41:43 +0000 (15:41 +0800)]
crimson/os/seastore: support promote extents purged from pinboard

Signed-off-by: Zhang Song <zhangsong02@qianxin.com>
Signed-off-by: Xuehan Xu <xuxuehan@qianxin.com>
3 days agocrimson/os/seastore: rename extent_2q_state_t to extent_pin_state_t
Zhang Song [Wed, 3 Sep 2025 07:58:59 +0000 (15:58 +0800)]
crimson/os/seastore: rename extent_2q_state_t to extent_pin_state_t

This is because, even for LRU caches, extents still have pin states,
although it's only Fresh. What's more, in the logical bucket cache
world, extents in both LRU and 2Q caches may be in the
Fresh/PendingPromote/Promoting state.

Signed-off-by: Zhang Song <zhangsong02@qianxin.com>
Signed-off-by: Xuehan Xu <xuxuehan@qianxin.com>
3 days agocrimson/os/seastore: introduce PROMOTE and DEMOTE transaction type
Zhang Song [Wed, 10 Apr 2024 03:17:30 +0000 (11:17 +0800)]
crimson/os/seastore: introduce PROMOTE and DEMOTE transaction type

Signed-off-by: Zhang Song <zhangsong02@qianxin.com>
Signed-off-by: Xuehan Xu <xuxuehan@qianxin.com>
3 days agodoc/dev/crimson: add design the document for LogicalBucketCache
Xuehan Xu [Wed, 1 Jul 2026 11:52:31 +0000 (19:52 +0800)]
doc/dev/crimson: add design the document for LogicalBucketCache

Signed-off-by: Xuehan Xu <xuxuehan@qianxin.com>
3 days agoMerge pull request #70051 from mheler/wip-copyobject-replace
mheler [Wed, 29 Jul 2026 18:30:53 +0000 (13:30 -0500)]
Merge pull request #70051 from mheler/wip-copyobject-replace

3 days agoMerge pull request #70659 from bluikko/wip-doc-cephadm-install-substitutions
bluikko [Wed, 29 Jul 2026 16:05:19 +0000 (23:05 +0700)]
Merge pull request #70659 from bluikko/wip-doc-cephadm-install-substitutions

doc/cephadm: enable substitutions on CLI command example

3 days agoMerge pull request #69533 from qiuxinyidian/main
Casey Bodley [Wed, 29 Jul 2026 15:44:29 +0000 (11:44 -0400)]
Merge pull request #69533 from qiuxinyidian/main

rgw:adding bucket index Indexless check when bi list

Reviewed-by: Casey Bodley <cbodley@redhat.com>
3 days agoMerge pull request #70097 from tchaikov/wip-mgr-testcase-single-bounce
Afreen Misbah [Wed, 29 Jul 2026 12:56:36 +0000 (18:26 +0530)]
Merge pull request #70097 from tchaikov/wip-mgr-testcase-single-bounce

Reviewed-by: Afreen Misbah <afreen@ibm.com>
3 days agoMerge pull request #70515 from rhcs-dashboard/fix-17067
Afreen Misbah [Wed, 29 Jul 2026 12:56:28 +0000 (18:26 +0530)]
Merge pull request #70515 from rhcs-dashboard/fix-17067

Reviewed-by: Afreen Misbah <afreen@ibm.com>
3 days agoMerge pull request #70604 from rhcs-dashboard/service-form-text-label-fix
Pedro Gonzalez Gomez [Wed, 29 Jul 2026 12:11:46 +0000 (14:11 +0200)]
Merge pull request #70604 from rhcs-dashboard/service-form-text-label-fix

mgr/dashboard: fix cd-text-label-list string writeValue

Reviewed-by: Nizamudeen A <nia@redhat.com>
3 days agodoc/cephadm: enable substitutions on CLI command example 70659/head
Ville Ojamo [Wed, 29 Jul 2026 09:40:54 +0000 (16:40 +0700)]
doc/cephadm: enable substitutions on CLI command example

Substitutions were not enabled on a prompt directive that tried to use
the stable-release substitution so the substitution string was rendered
verbatim.

Signed-off-by: Ville Ojamo <git2233+ceph@ojamo.eu>
3 days agoMerge pull request #70625 from rhcs-dashboard/fix/dashboard-gzip-mime-type
Afreen Misbah [Wed, 29 Jul 2026 09:37:01 +0000 (15:07 +0530)]
Merge pull request #70625 from rhcs-dashboard/fix/dashboard-gzip-mime-type

mgr/dashboard: add text/javascript to gzip mime types

Reviewed-by: Afreen Misbah <afreen@ibm.com>
3 days agoMerge pull request #70606 from rhcs-dashboard/add-doc-url
Afreen Misbah [Wed, 29 Jul 2026 08:02:30 +0000 (13:32 +0530)]
Merge pull request #70606 from rhcs-dashboard/add-doc-url

mgr/dashboard: Add alerts doc url to active alert list

Reviewed-by: Aashish Sharma <aasharma@redhat.com>
3 days agoMerge pull request #68519 from edwinzrodriguez/cephfs-bench
Venky Shankar [Wed, 29 Jul 2026 07:39:51 +0000 (13:09 +0530)]
Merge pull request #68519 from edwinzrodriguez/cephfs-bench

cephfs-tool: Improve reporting and config options

Reviewed-by: Venky Shankar <vshankar@redhat.com>
3 days agoMerge pull request #70486 from guits/fix-nvme-reformat
Guillaume Abrioux [Wed, 29 Jul 2026 07:34:24 +0000 (09:34 +0200)]
Merge pull request #70486 from guits/fix-nvme-reformat

ceph-volume: fix nvme preformat wiping VG on osds-per-device > 1

3 days agoMerge pull request #70603 from rhcs-dashboard/snap-edit-fix
Pedro Gonzalez Gomez [Wed, 29 Jul 2026 06:54:00 +0000 (08:54 +0200)]
Merge pull request #70603 from rhcs-dashboard/snap-edit-fix

mgr/dashboard: fix snap schedule startDate

Reviewed-by: Afreen Misbah <afreen@ibm.com>
Reviewed-by: Nizamudeen A <nia@redhat.com>
3 days agomgr/dashboard: fix cd-text-label-list string writeValue 70604/head
root [Tue, 28 Jul 2026 06:12:40 +0000 (08:12 +0200)]
mgr/dashboard: fix cd-text-label-list string writeValue

When editing services such as oauth2-proxy, backend fields like scope
arrive as a single string. Spreading that string into the list turned
each character into its own row. Normalize string values to one item.

Fixes: https://tracker.ceph.com/issues/78808
Signed-off-by: Pedro Gonzalez Gomez <pegonzal@ibm.com>
4 days agoMerge pull request #69680 from sunyuechi/wip-bufferlist-omap
Kefu Chai [Wed, 29 Jul 2026 03:47:24 +0000 (11:47 +0800)]
Merge pull request #69680 from sunyuechi/wip-bufferlist-omap

test/crimson/omap: enlarge values under ASan to shrink depth-driven tests

Reviewed-by: Kefu Chai <k.chai@proxmox.com>
4 days agoMerge pull request #66757 from senolcolak/wip-v2-mgr-inventory-crush-host
SrinivasaBharathKanta [Wed, 29 Jul 2026 03:28:21 +0000 (08:58 +0530)]
Merge pull request #66757 from senolcolak/wip-v2-mgr-inventory-crush-host

mgr: override OSD hostname with CRUSH physical host in all metadata paths

4 days agoMerge pull request #69885 from sunyuechi/wip-rados-fix-std-fd-close
SrinivasaBharathKanta [Wed, 29 Jul 2026 03:26:20 +0000 (08:56 +0530)]
Merge pull request #69885 from sunyuechi/wip-rados-fix-std-fd-close

tools/rados: fix stdio/stream cleanup

4 days agoMerge pull request #67023 from trociny/wip-74490
SrinivasaBharathKanta [Wed, 29 Jul 2026 03:25:52 +0000 (08:55 +0530)]
Merge pull request #67023 from trociny/wip-74490

mon: allow getpoolstats for mgr modules

4 days agoMerge pull request #69973 from ljflores/wip-tracker-77382
SrinivasaBharathKanta [Wed, 29 Jul 2026 03:25:12 +0000 (08:55 +0530)]
Merge pull request #69973 from ljflores/wip-tracker-77382

qa/suites/upgrade: ignore expected telemetry warning

4 days agoMerge pull request #70206 from sunyuechi/wip-dockerfile-sccache-dnf
Kefu Chai [Wed, 29 Jul 2026 03:22:01 +0000 (11:22 +0800)]
Merge pull request #70206 from sunyuechi/wip-dockerfile-sccache-dnf

Dockerfile.build: install sccache from distro packages when available

Reviewed-by: Zack Cerza <zack@redhat.com>
4 days agoMerge pull request #69325 from NitzanMordhai/wip-nitzan-mgr-memory-grow-debug-high...
SrinivasaBharathKanta [Tue, 28 Jul 2026 23:54:49 +0000 (05:24 +0530)]
Merge pull request #69325 from NitzanMordhai/wip-nitzan-mgr-memory-grow-debug-high-level

mgr/ClusterState: remove debug 30 memory spike

4 days agoMerge pull request #69099 from MaxKellermann/osd_SnapMapper__fmt_no_sprintf
SrinivasaBharathKanta [Tue, 28 Jul 2026 23:54:11 +0000 (05:24 +0530)]
Merge pull request #69099 from MaxKellermann/osd_SnapMapper__fmt_no_sprintf

osd/SnapMapper: use fmt::format() instead of fmt::sprintf()

4 days agoMerge pull request #69098 from MaxKellermann/common_options__long_desc__c_string
SrinivasaBharathKanta [Tue, 28 Jul 2026 23:53:55 +0000 (05:23 +0530)]
Merge pull request #69098 from MaxKellermann/common_options__long_desc__c_string

common/options: convert `long_desc` to C string

4 days agoMerge pull request #70231 from bill-scales/issue58417
SrinivasaBharathKanta [Tue, 28 Jul 2026 23:43:48 +0000 (05:13 +0530)]
Merge pull request #70231 from bill-scales/issue58417

osd: Fix race hazard causing duplicated message after OSD marked down

4 days agoMerge pull request #70233 from bill-scales/issue54932
SrinivasaBharathKanta [Tue, 28 Jul 2026 23:43:18 +0000 (05:13 +0530)]
Merge pull request #70233 from bill-scales/issue54932

osd: Fix PeeringState::proc_lease crash caused by race hazard

4 days agoMerge pull request #70333 from bill-scales/issue76953
Radoslaw Zarzynski [Tue, 28 Jul 2026 20:22:56 +0000 (22:22 +0200)]
Merge pull request #70333 from bill-scales/issue76953

osd: Fix valgrind uninitialized variable issues in fast EC

Reviewed-by: Matty Williams <Matty.Williams@ibm.com>
Reviewed-by: Radoslaw Zarzynski <rzarzyns@redhat.com>
4 days agoMerge pull request #70505 from ivancich/wip-fix-ceph-spec-in
J. Eric Ivancich [Tue, 28 Jul 2026 20:16:58 +0000 (16:16 -0400)]
Merge pull request #70505 from ivancich/wip-fix-ceph-spec-in

rgw: update rpm and debian builds

Reviewed-by: Thomas Serlin <tserlin@redhat.com>
4 days agomgr/dashboard: add text/javascript to gzip mime types 70625/head
Afreen Misbah [Tue, 28 Jul 2026 12:47:56 +0000 (18:17 +0530)]
mgr/dashboard: add text/javascript to gzip mime types

CherryPy's serve_file() uses Python's mimetypes module which returns
text/javascript for .js files (RFC 9239), but the gzip config only
listed application/javascript. This meant the main Angular bundle
was being served uncompressed. With Cheroot write timeouts,
the uncompressed transfer times out over slower networks, causing
ERR_CONTENT_LENGTH_MISMATCH in the browser.

Signed-off-by: Afreen Misbah <afreen@ibm.com>
4 days agoMerge pull request #70544 from rhcs-dashboard/switcher
Afreen Misbah [Tue, 28 Jul 2026 13:33:46 +0000 (19:03 +0530)]
Merge pull request #70544 from rhcs-dashboard/switcher

mgr/dashboard: rgw gateway switcher read-only for resource page

Reviewed-by: Afreen Misbah <afreen@ibm.com>
4 days agoMerge pull request #70547 from xxhdx1985126/wip-crimson-recover-delete-fix
Xuehan Xu [Tue, 28 Jul 2026 13:19:26 +0000 (21:19 +0800)]
Merge pull request #70547 from xxhdx1985126/wip-crimson-recover-delete-fix

crimson/osd/pg_recovery: call on_global_recover on replica-recover-deleted objects only when it's not missing on the primary too

Reviewed-by: Kefu Chai <k.chai@proxmox.com>
4 days agoMerge PR #70581 into main
Patrick Donnelly [Tue, 28 Jul 2026 12:43:05 +0000 (08:43 -0400)]
Merge PR #70581 into main

* refs/pull/70581/head:
doc/governance: update CEC term length

Reviewed-by: Ilya Dryomov <idryomov@redhat.com>
4 days agoMerge PR #70574 into main
Patrick Donnelly [Tue, 28 Jul 2026 12:40:32 +0000 (08:40 -0400)]
Merge PR #70574 into main

* refs/pull/70574/head:
.github/workflows: bump checkout action version
.github/workflows/qa-symlink: sparse checkout verify-qa
.github/workflows/qa-symlink: run only for active branches

Reviewed-by: Joseph Mundackal <jmundackal@bloomberg.net>
4 days agoMerge PR #70582 into main
Patrick Donnelly [Tue, 28 Jul 2026 12:40:04 +0000 (08:40 -0400)]
Merge PR #70582 into main

* refs/pull/70582/head:
doc/governance: add Redouane to CSC

Reviewed-by: Anthony D'Atri <anthony.datri@gmail.com>
Reviewed-by: Ilya Dryomov <idryomov@redhat.com>
4 days agoMerge PR #70586 into main
Patrick Donnelly [Tue, 28 Jul 2026 12:38:54 +0000 (08:38 -0400)]
Merge PR #70586 into main

* refs/pull/70586/head:
.githubmap: add BBoozmen

Reviewed-by: Oguzhan Ozmen <oozmen@bloomberg.net>
4 days agoMerge PR #70598 into main
Patrick Donnelly [Tue, 28 Jul 2026 12:36:50 +0000 (08:36 -0400)]
Merge PR #70598 into main

* refs/pull/70598/head:
.githubmap: add mheler

Reviewed-by: Patrick Donnelly <pdonnell@ibm.com>
4 days agoMerge pull request #70614 from VallariAg/wip-nvmeof-upgrade-refresh
Vallari Agrawal [Tue, 28 Jul 2026 10:33:38 +0000 (16:03 +0530)]
Merge pull request #70614 from VallariAg/wip-nvmeof-upgrade-refresh

qa/suites/nvmeof/upgrade: use refresh True for wait_for_service

4 days agoMerge pull request #69662 from aainscow/truncate_and_write_tests
Alex Ainscow [Tue, 28 Jul 2026 10:27:43 +0000 (11:27 +0100)]
Merge pull request #69662 from aainscow/truncate_and_write_tests

test/osd: Add truncate-then-write tests

Reviewed-by: Bill Scales <bill_scales@uk.ibm.com>
4 days agoMerge pull request #69758 from tchaikov/wip-bumpup-seastar
Kefu Chai [Tue, 28 Jul 2026 10:24:32 +0000 (18:24 +0800)]
Merge pull request #69758 from tchaikov/wip-bumpup-seastar

crimson: bump seastar submodule and adapt to its new future API

Reviewed-by: Matan Breizman <mbreizma@redhat.com>
4 days agomgr/dashboard: Add alerts doc url to active alert list 70606/head
Aashish Sharma [Tue, 28 Jul 2026 07:00:59 +0000 (12:30 +0530)]
mgr/dashboard: Add alerts doc url to active alert list

Add documentation link in Learn more column for every alert in the
active alerts list table

Fixes: https://tracker-origin.ceph.com/issues/78747
Signed-off-by: Aashish Sharma <aasharma@redhat.com>
4 days agoqa/suites/nvmeof/upgrade: use refresh True for wait_for_service 70614/head
Vallari Agrawal [Mon, 27 Jul 2026 06:21:39 +0000 (11:51 +0530)]
qa/suites/nvmeof/upgrade: use refresh True for wait_for_service

Add refresh: True to upgrade tests, same as we added for
https://tracker.ceph.com/issues/73538

Fixes: https://tracker.ceph.com/issues/78770
Co-authored-by: barakda <barak.davidov@gmail.com>
Signed-off-by: Vallari Agrawal <vallari.agrawal@ibm.com>
4 days agocrimson: migrate off deprecated seastar::smp::count and all_cpus() 69758/head
Kefu Chai [Fri, 26 Jun 2026 11:35:15 +0000 (19:35 +0800)]
crimson: migrate off deprecated seastar::smp::count and all_cpus()

The updated seastar deprecates the global smp::count and smp::all_cpus() in
favour of the per-instance smp::shard_count() and smp::all_shards(). Under the
CI -Werror build these deprecation warnings are fatal.

Replace the uses with the reactor-local free functions
seastar::this_smp_shard_count() and seastar::this_smp_all_shards(), which read
the current smp instance's shard count and shard-id range. crimson runs a
single smp instance and every migrated site executes on a reactor thread (the
alienstore worker threads do not touch these), so the behaviour is unchanged.

Signed-off-by: Kefu Chai <k.chai@proxmox.com>
4 days agocrimson: use the seastar span sink API and build at API level 10
Kefu Chai [Fri, 26 Jun 2026 07:56:35 +0000 (15:56 +0800)]
crimson: use the seastar span sink API and build at API level 10

At SEASTAR_API_LEVEL >= 9 seastar dropped output_stream::write(net::packet) in
favour of write(std::span<temporary_buffer<char>>). Convert Socket's writes to
the span form, writing the packet's released fragments and keeping them alive
across the write with do_with, then raise Seastar_API_LEVEL from 6 to 10 to
match the updated seastar. The span write overloads are unconditional, so the
Socket change is valid at every API level.

Signed-off-by: Kefu Chai <k.chai@proxmox.com>
4 days agocrimson: bump seastar submodule and adapt to its new future API
Kefu Chai [Fri, 26 Jun 2026 07:56:35 +0000 (15:56 +0800)]
crimson: bump seastar submodule and adapt to its new future API

Update src/seastar to ceph-umbrella-25.05.0-1074-gcced0236, which is
scylladb/seastar master (fd71a5b4c7) with the ceph patches reapplied,
and adapt crimson to the future API changes the bump brings in:

  - interruptible_future.h: seastar split its single ready_future_marker into
    set_ready_future_marker and set_from_tuple_ready_future_marker; route
    futurize::from_tuple through the matching markers.
  - thread_pool.h: future_state::set() is now templated and get_value() is
    ref-qualified; pass seastar::internal::monostate{} for the void case and
    move the state before reading its value.

These crimson changes need the new seastar, so they share one commit with the
submodule bump. The build stays at Seastar_API_LEVEL 6 here; the level is
raised in the following commit.

Signed-off-by: Kefu Chai <k.chai@proxmox.com>
4 days agoMerge pull request #70180 from Matan-B/wip-matanb-seastore-coll-batch
Matan Breizman [Tue, 28 Jul 2026 07:13:32 +0000 (10:13 +0300)]
Merge pull request #70180 from Matan-B/wip-matanb-seastore-coll-batch

crimson/seastore: batch same-collection transactions

Reviewed-by: Josh Durgin <jdurgin@redhat.com>
Reviewed-by: Xuehan Xu <xuxuehan@qianxin.com>
4 days agoMerge pull request #70467 from bluikko/wip-doc-nginx-capitalization
bluikko [Tue, 28 Jul 2026 07:04:30 +0000 (14:04 +0700)]
Merge pull request #70467 from bluikko/wip-doc-nginx-capitalization

doc: standardize capitalization of NGINX following the vendor

5 days agoMerge pull request #70209 from Hezko/warn-on-deprecated-param
Hezko [Tue, 28 Jul 2026 06:25:41 +0000 (09:25 +0300)]
Merge pull request #70209 from Hezko/warn-on-deprecated-param

mgr/dashboard: add deprecation warning on ns add cli command

5 days agoMerge pull request #70560 from rhcs-dashboard/fix-gateway-readonly
Afreen Misbah [Tue, 28 Jul 2026 06:18:23 +0000 (11:48 +0530)]
Merge pull request #70560 from rhcs-dashboard/fix-gateway-readonly

Reviewed-by: Naman Munet <nmunet@redhat.com>
5 days agoMerge pull request #70566 from rhcs-dashboard/service-form-helper-fix
Afreen Misbah [Tue, 28 Jul 2026 06:18:18 +0000 (11:48 +0530)]
Merge pull request #70566 from rhcs-dashboard/service-form-helper-fix

Reviewed-by: Nizamudeen A <nia@redhat.com>
5 days agomgr/dashboard: fix snap schedule startDate 70603/head
Pedro Gonzalez Gomez [Tue, 28 Jul 2026 06:05:53 +0000 (08:05 +0200)]
mgr/dashboard: fix snap schedule startDate

Fixes date time picker that was wrongly setting the form date

Fixes: https://tracker-origin.ceph.com/issues/78742
Signed-off-by: Pedro Gonzalez Gomez <pegonzal@ibm.com>
5 days agoMerge PR #70375 into main
Venky Shankar [Tue, 28 Jul 2026 05:55:58 +0000 (11:25 +0530)]
Merge PR #70375 into main

* refs/pull/70375/head:
src/pybind: fixing the test_invalid_client_id failure

Reviewed-by: Jos Collin <jcollin@redhat.com>
Reviewed-by: Venky Shankar <vshankar@redhat.com>
5 days agoMerge pull request #69726 from tchaikov/wip-drop-bluestore-pmem
Kefu Chai [Tue, 28 Jul 2026 05:27:17 +0000 (13:27 +0800)]
Merge pull request #69726 from tchaikov/wip-drop-bluestore-pmem

blk,build: drop BlueStore PMEM support

Reviewed-by: Igor Fedotov <igor.fedotov@croit.io>
5 days agomgr/dashboard: add deprecation warning on ns add nvme cli command 70209/head
Tomer Haskalovitch [Mon, 13 Jul 2026 22:34:38 +0000 (01:34 +0300)]
mgr/dashboard: add deprecation warning on ns add nvme cli command

Fixes: https://tracker.ceph.com/issues/78228
Signed-off-by: Tomer Haskalovitch <tomer.haska@ibm.com>
5 days agoMerge pull request #68209 from sunyuechi/riscv-bench-crc
Kefu Chai [Tue, 28 Jul 2026 01:40:42 +0000 (09:40 +0800)]
Merge pull request #68209 from sunyuechi/riscv-bench-crc

test/common: add RISC-V CRC32C performance benchmark to unittest_crc32c

Reviewed-by: Kefu Chai <k.chai@proxmox.com>
5 days agoMerge pull request #70578 from gbregman/main
Gil Bregman [Mon, 27 Jul 2026 23:04:26 +0000 (02:04 +0300)]
Merge pull request #70578 from gbregman/main

mgr/cephadm: Add CNC fields to NVMEoF spec file

5 days ago.githubmap: add mheler 70598/head
Matthew N. Heler [Mon, 27 Jul 2026 22:24:32 +0000 (17:24 -0500)]
.githubmap: add mheler

Signed-off-by: Matthew N. Heler <matthew.heler@hotmail.com>
5 days agomgr/dashboard: rgw gateway switcher read-only for resource page 70544/head
Syed Ali Ul Hasan [Sat, 25 Jul 2026 22:08:19 +0000 (03:38 +0530)]
mgr/dashboard: rgw gateway switcher read-only for resource page

- Fixes: https://tracker.ceph.com/issues/78513

Signed-off-by: Syed Ali Ul Hasan <syedaliulhasan19@gmail.com>
5 days agoMerge pull request #69790 from syedali237/dashboard/user-management
Afreen Misbah [Mon, 27 Jul 2026 20:19:50 +0000 (01:49 +0530)]
Merge pull request #69790 from syedali237/dashboard/user-management

mgr/dashboard : migrated User Table tabs to resource pages

Reviewed-by: Afreen Misbah <afreen@ibm.com>
Reviewed-by: Abhishek Desai <abhishek.desai1@ibm.com>
5 days agotest/osd: cover truncate+write EC transaction behaviour 69662/head
Alex Ainscow [Tue, 9 Jun 2026 14:45:25 +0000 (15:45 +0100)]
test/osd: cover truncate+write EC transaction behaviour

Add unit-tests to recreate the problems found by the IO Sequencer in the previous commit.

Some tests here rely on frameworks added in Umbrella and as such we are keeping them
independent (in their own PRs).

Signed-off-by: Alex Ainscow <aainscow@uk.ibm.com>
Signed-off-by: Matty Williams <Matty.Williams@ibm.com>
Assisted-by: IBM Bob:Claude/GPT
5 days agomgr/cephadm: Add CNC fields to NVMEoF spec file 70578/head
Gil Bregman [Mon, 27 Jul 2026 14:08:35 +0000 (17:08 +0300)]
mgr/cephadm: Add CNC fields to NVMEoF spec file
Fixes: https://tracker-origin.ceph.com/issues/78675
Signed-off-by: Gil Bregman <gbregman@il.ibm.com>
5 days ago.githubmap: add BBoozmen 70586/head
Patrick Donnelly [Mon, 27 Jul 2026 17:23:31 +0000 (13:23 -0400)]
.githubmap: add BBoozmen

Signed-off-by: Patrick Donnelly <pdonnell@ibm.com>
5 days agorgw: update build config to streamline builds 70505/head
J. Eric Ivancich [Thu, 23 Jul 2026 19:32:41 +0000 (15:32 -0400)]
rgw: update build config to streamline builds

Some binaries and man pages were not included in rpm builds or debian
builds.  This adds those artifacts to ceph.spec.in
debian/ceph-common.install and debian/radosgw.install.

Signed-off-by: J. Eric Ivancich <ivancich@redhat.com>
5 days agoMerge PR #70576 into main
Patrick Donnelly [Mon, 27 Jul 2026 17:06:57 +0000 (13:06 -0400)]
Merge PR #70576 into main

* refs/pull/70576/head:
script/ptl-tool: add more refs for main branch

Reviewed-by: Yuri Weinstein <yweins@redhat.com>
5 days agomgr/dashboard: migrated user table tabs to resource pages 69790/head
Syed Ali Ul Hasan [Wed, 22 Jul 2026 14:22:20 +0000 (19:52 +0530)]
mgr/dashboard: migrated user table tabs to resource pages

- Fixes: https://tracker.ceph.com/issues/77475

Signed-off-by: Syed Ali Ul Hasan <syedaliulhasan19@gmail.com>
5 days agodoc/governance: update CEC term length 70581/head
Patrick Donnelly [Mon, 27 Jul 2026 15:10:21 +0000 (11:10 -0400)]
doc/governance: update CEC term length

Fixes: https://tracker.ceph.com/issues/77966
Signed-off-by: Patrick Donnelly <pdonnell@ibm.com>
5 days agodoc/governance: add Redouane to CSC 70582/head
Patrick Donnelly [Mon, 27 Jul 2026 15:12:17 +0000 (11:12 -0400)]
doc/governance: add Redouane to CSC

Fixes: https://tracker.ceph.com/issues/77965
Signed-off-by: Patrick Donnelly <pdonnell@ibm.com>
5 days agoscript/ptl-tool: add more refs for main branch 70576/head
Patrick Donnelly [Mon, 27 Jul 2026 13:41:33 +0000 (09:41 -0400)]
script/ptl-tool: add more refs for main branch

Start with "upstream", then "origin", then finally the local main branch.

Signed-off-by: Patrick Donnelly <pdonnell@ibm.com>
5 days agoMerge pull request #69844 from kginonredhat/issue-77835-fail-fs-mds-upgrade-wait
Redouane Kachach [Mon, 27 Jul 2026 13:39:40 +0000 (15:39 +0200)]
Merge pull request #69844 from kginonredhat/issue-77835-fail-fs-mds-upgrade-wait

mgr/cephadm: wait for active MDS during fail_fs upgrade

Reviewed-by: Patrick Donnelly <pdonnell@redhat.com>