* refs/pull/70675/head:
mgr/nfs: Updated enable_nfsv3 check while export creation
mgr/nfs: set nfs export protocols based on cluster's protocol settings
mgr/cephadm: Updated NFS default protocol to v4 and v3 is enabled only when enable_nfsv3 is set in the spec
Reviewed-by: Adam King <adking@redhat.com>
Fixes: https://tracker.ceph.com/issues/74492
Signed-off-by: Shweta Bhosale <Shweta.Bhosale1@ibm.com>
(cherry picked from commit 44c1362028)
Conflicts:
src/pybind/mgr/nfs/export.py - due to code refactoring in main
Fixes: https://tracker.ceph.com/issues/74492
Signed-off-by: Shweta Bhosale <Shweta.Bhosale1@ibm.com>
(cherry picked from commit 9b0becfdf2)
Conflicts:
doc/cephadm/services/nfs.rst - due to documentation related to RDMA
doc/mgr/nfs.rst - due to params related to RDMA
src/pybind/mgr/cephadm/services/nfs.py - due to other extra params
src/pybind/mgr/cephadm/templates/services/nfs/ganesha.conf.j2 - due to RDMA
src/pybind/mgr/cephadm/tests/services/test_ingress.py - not exists on tentacle
src/pybind/mgr/cephadm/tests/services/test_nfs.py - not exists on tentacle
src/pybind/mgr/nfs/cluster.py - due to other nfs parameters which are not supported in tentacle
src/pybind/mgr/nfs/module.py - due to other nfs parameters which are not supported in tentacle
src/python-common/ceph/deployment/service_spec.py - due to extra variable not supported in tentacle
These workflows (with pull_request) are confusing the CI checks for
release branches. Remove all of them and let `main` be the sole
definition.
Signed-off-by: Patrick Donnelly <pdonnell@ibm.com>
* refs/pull/68426/head:
os/bluestore/bluefs: Make sure write buffer is aligned
test/bluefs: add cases for wal v2 envelope header mislocation.
os/bluestore/bluefs: Add assert if write buffer is not page aligned
Reviewed-by: Adam Kupczyk <akupczyk@redhat.com>
* refs/pull/69024/head:
rgw: avoid doubled ARN in GetBucketReplication for pre-existing data
Reviewed-by: Casey Bodley <cbodley@redhat.com>
Reviewed-by: Patrick Donnelly <pdonnell@ibm.com>
* refs/pull/70108/head:
rgw: return additional checksum headers too
rgw: return an etag header for all successful complete-multipart
Reviewed-by: Adam C. Emerson <aemerson@redhat.com>
* refs/pull/68444/head:
rgw/iam: match value of Null condition
rgw/iam: typed_any/typed_none take x as forwarding reference
Reviewed-by: Patrick Donnelly <pdonnell@ibm.com>
* refs/pull/70262/head:
rgw/beast: dispatch connection close on strand during pause
common/async: `io_context_pool` no longer moves from init function
rgw/rados: Initialize things before starting threads that use them
rgw: Check `going_down` in other-than-`interval_set` loop
test/rgw/beast: add unit test for ConnectionList shutdown behavior
rgw/beast: (non-functional change) add size method to ConnectionList
rgw/beast: fix shutdown crash in ConnectionList::close()
rgw/beast: (non-functional) refactor Connection and ConnectionList to header
Reviewed-by: Casey Bodley <cbodley@redhat.com>
* refs/pull/69277/head:
test/rgw/s3-tests: add CopyObject and UploadPart encryption tests
qa/rgw: crypt subsuite tests compress-encrypted
qa/rgw: rgw task can enable/disable zone features
qa/rgw: don't duplicate 'user list' commands for default zone
rgw: always decompress in CopyObject when source is compressed
rgw: fix fixup_range cascade for compressed+encrypted objects
rgw: fix offset calculation in copy_obj_data
rgw: sanitize copy source attrs
rgw: verify sse-c headers on multipart completion
rgw: always return crypt headers on put obj
rgw: expose copy_source from rgw_s3_prepare_decrypt
rgw: set compression attrs in copy_obj_data()
rgw: implement CopyObject for encrypted object
rgw: introduce RGWCopyObjDPF for RGWCopyObj
rgw: introduce DataProcessorFilter
rgw: pass DataProcessorFactory to copy_object
rgw/sal: introduce DataProcessorFactory
rgw: gather all get_decrypt_filter together
rgw: teach rgw_s3_prepare_decrypt about source decrypt headers
rgw: involve next on RGWGetObj_BlockDecrypt::flush
rgw: involve next on RGWGetObj_BlockDecrypt::fixup_range
Reviewed-by: J. Eric Ivancich <ivancich@redhat.com>
* refs/pull/68853/head:
doc: added a note for damaged hard links in scrub documentation
qa: add a test to verify that a damage hard link id detected during scrub
mds: identify damaged hard links during scrub
Reviewed-by: Rishabh Dave <ridave@redhat.com>
* refs/pull/65533/head:
mds/MDSMap: Enhance set_min_compat_client to validate iterator range before sorting feature bits
Reviewed-by: Rishabh Dave <ridave@redhat.com>
* refs/pull/65812/head:
src/common: add helper to prepend "..." to trimmed paths
mds/ScrubStack: avoid generating inode path since it is unused
mds: fix few log entries
client: trim path before logging it
mds: log trimmed path wherever generating full path is necessary
mds: for logging generate only 10 final components of dentry path
mds: for logging generate only 10 final components of inode path
qa, test: run unit tests for cephfs.pyx with non-root user
test/pybind: add unit tests for rmtree() in cephfs python bindings
pybind/cephfs, mgr/volumes: refactor purge() to be non-recursive
Reviewed-by: Patrick Donnelly <pdonnell@ibm.com>
quoting https://docs.aws.amazon.com/IAM/latest/UserGuide/reference_policies_elements_condition_operators.html#Conditions_Null:
> Use a Null condition operator to check if a condition key is absent at the time of authorization. In the policy statement, use either true (the key doesn't exist โ it is null) or false (the key exists and its value is not null).
previously, this check acted as if it was always given a value of
"true". instead, convert this decision to a "true"/"false" string
and use `typed_any()` to match that against the parsed `vals`
`runtime_vals` are ignored here because the same document says:
> You can not use a policy variable with the Null condition operator.
Fixes: https://tracker.ceph.com/issues/74736
Signed-off-by: Casey Bodley <cbodley@redhat.com>
(cherry picked from commit 1d0c8c286c)
* refs/pull/66269/head:
messages: fix MOSDOpReplay decode of old version bad_replay_version
include/types: Update set wire function parameter
messages: MOSDOpReply encode and decode errorcode32_t with PGID64 features on
Reviewed-by: Laura Flores <lflores@redhat.com>
* refs/pull/66491/head:
test/rgw/notify: cover case with more than 1K topics
rgw/notify: fix reading the entries in a loop
Reviewed-by: Yuval Lifshitz <ylifshit@redhat.com>
* refs/pull/67214/head:
rgw/iam: add NotEquals UT
rgw/iam: fix NotEquals conditions to use AND logic instead of OR
rgw/iam: rename condition helpers for clarity
Reviewed-by: Adam C. Emerson <aemerson@redhat.com>
When no OSDs within a CRUSH bucket can be found to upgrade, the earlier error
message indicated the minimum number of PGs affected if any OSD was removed
from the set. For e.g.,
"Error EBUSY: unsafe to upgrade osd(s) at this time (at least X PG(s) will
become offline if any OSD out of the Y in CRUSH bucket 'foo' is stopped)"
This was inaccurate in some cases because the number of PGs affected is
determined only from the last tested OSD in the CRUSH bucket based on the sort
order (i.e., OSD with least number of PGs per OSD). Based on the CRUSH rules
and placement, this doesn't mean that the least number of PGs are affected if
this OSD is stopped. There may be other OSDs in the set that affect a smaller
number of PGs if made offline.
But with the way ok-to-upgrade logic uses the convergence factor, not all OSDs
would be tested for the offline pg check. Therefore, it is not possible to
accurately determine the minimum number affected PGs.
In view of the above, the error message is modified as shown below to be
slightly more generic but still convey the reason:
"Error EBUSY: unsafe to upgrade osd(s) at this time (one or more PG(s) will
become offline if any OSD out of the Y in CRUSH bucket 'foo' is stopped)"
Fixes: https://tracker.ceph.com/issues/78425
Signed-off-by: Sridhar Seshasayee <sridhar.seshasayee@ibm.com>
(cherry picked from commit afd0b924e2)
* refs/pull/69589/head:
mgr/dashboard: Fix HTML lint, OpenAPI spec, and RGW Account Role unit tests
mgr/dashboard: align RGW role management with Carbon and fix API routing
mgr/dashboard: Remove global RGW tenant Roles tab and decommission routes
Reviewed-by: Afreen Misbah <afreen@ibm.com>
