Commit Graph
226 Commits
Author SHA1 Message Date
travisn 27265ff279 helm: remove cpu limits from all pods
When CPU requests and limits are assigned to a pod,
the pod will be guaranteed the requests, up to the limits.
Even if there are spare CPU cycles, the pod cannot use
them. Thus, pods can be unnecessarily denied compute
when they need to burst if the limits are set.

Therefore, it is not recommended to set CPU limits
since the CPU requests are already guaranteeing that
no pod will be starved at least for its requests.

Signed-off-by: travisn <tnielsen@redhat.com>
2024-02-07 17:36:59 -07:00
Joshua Hoblitt a670fd9671 docs: fix CephCluster.spec.cephVersion.imagePullPolicy list item
Signed-off-by: Joshua Hoblitt <josh@hoblitt.com>
2024-02-07 11:40:30 -07:00
Liang Zheng ee05e82cea osd: support create osd with metadata partition
Currently, when rook provisions OSDs(in the OSD prepare job), rook effectively run a
c-v command such as the following.
```console
ceph-volume lvm batch --prepare <deviceA> <deviceB> <deviceC> --db-devices <metadataDevice>
```
but c-v lvm batch only supports disk and lvm, instead of disk partitions.

We can resort to `ceph-volume lvm prepare` to implement it.

Signed-off-by: Liang Zheng <zhengliang0901@gmail.com>
2024-02-07 11:34:09 +08:00
Raghavendra Talur cb702d3d52 external: allow specifying the kubernetes context
If the import script is run on a system that has many kubernetes configs
in the kubeconfig file, it is useful to be able to specify the context
to use.

Signed-off-by: Raghavendra Talur <raghavendra.talur@gmail.com>
2024-02-06 12:52:02 -05:00
travisn f0c30e2edb manifest: remove pg_autoscaler from mgr modules examples
The pg_autoscaler mgr module is no longer possible to disable.
Ceph requires the module to always be enabled. Now the
autoscaling configuration is done with a per-pool setting.

Now the rook module is shown as an example to enable.

Signed-off-by: travisn <tnielsen@redhat.com>
2024-01-19 14:26:41 -07:00
travisn 56a3370d83 docs: update all examples to ceph v18.2.1
Since v18.2.1 is the default recommendation for the ceph version,
update all the docs and examples to that version.

Signed-off-by: travisn <tnielsen@redhat.com>
2024-01-03 14:37:11 -07:00
Yuma Ogami 3a55526d55 doc: add that resizing encryptedDevice is not yet supported for host-based clusters
Caused by the following issue:
https://github.com/rook/rook/issues/13262

Signed-off-by: Yuma Ogami <yuma-ogami@cybozu.co.jp>
2023-12-21 01:01:49 +00:00
Blaine Gardner 61d0922010 Merge pull request #12369 from sp98/mon-failover
mon: allow changing hostNetwork settings
2023-12-18 08:54:59 -07:00
sp98 42ea2897cf mon: allow changing hostNetwork settings
Allow changing spec.network.hostNetwork settings on a running cluster
by failing over the mons

Signed-off-by: sp98 <sapillai@redhat.com>
2023-12-18 10:30:48 +05:30
Rakshith R 24430f989a doc: document spec.csi section in cephcluster crd
This commit documents the newly added `spec.csi`
section containing CSI driver related options.

Signed-off-by: Rakshith R <rar@redhat.com>
2023-12-14 19:27:18 +05:30
Travis Nielsen 85d94d4643 Merge pull request #13359 from parth-gr/documet-svg-pinning
docs: document the subvolumegroup pinning
2023-12-11 12:38:59 -07:00
parth-gr c9fd382c13 subvolumegroup: add pinning spec in subvolumegroup CRD
subvolumegroup can be pinned by pintype and pinsetting
So adding the spec to enhance its functionality

Closes: https://github.com/rook/rook/issues/12607

Signed-off-by: parth-gr <paarora@redhat.com>
2023-12-11 21:54:25 +05:30
parth-gr a3e2b3355e docs: document the subvolumegroup pinning
add the name and pinning spec in the docs

Closes: https://github.com/rook/rook/issues/13329

Signed-off-by: parth-gr <paarora@redhat.com>
2023-12-11 15:05:01 +05:30
Blaine Gardner 685a60ed76 docs: remove lingering notes about required quincy ver
Remove some lingering notes stating that features require Ceph Quincy.

Signed-off-by: Blaine Gardner <blaine.gardner@ibm.com>
2023-12-08 14:15:51 -07:00
Travis Nielsen 21efbf31c6 Merge pull request #13325 from cybozu-go/doc-mention-pgHealthCheckTimeout
doc: add the description of some fields and update the table of supported configurations for OSD creation
2023-12-05 13:42:00 -07:00
Satoru Takeuchi 2059a04fbd doc: update the table of supported configurations for OSD creation
In some cases, it's desired to use `must` than `should`.

Signed-off-by: Satoru Takeuchi <satoru.takeuchi@gmail.com>
2023-12-05 20:36:31 +00:00
Satoru Takeuchi ab1f1d97ac doc: describe pgHealthCheckTimeout pgHealthyRegex
Add the missing explanation of pgHealthCheckTimeout and pgHealthyRegex
in the CephCluster' CRD doc. In addition, remove a verbose sentence
from the explanation of `osdMaintenanceTimeout`.

Signed-off-by: Satoru Takeuchi <satoru.takeuchi@gmail.com>
2023-12-05 20:30:54 +00:00
Travis Nielsen 2edaf3a1f7 Merge pull request #13270 from parth-gr/radosnamespace-name
namespace: add name spec in cephBlockPoolRadosNamespace CRD
2023-12-05 10:46:55 -07:00
Alexander Trost 3097455d78 Merge pull request #13246 from koor-tech/ceph_config_via_cluster_crd_impl
operator: allow setting ceph config options via ceph cluster crd
2023-12-02 11:27:20 +01:00
Alexander Trost 4ed35d6bd6 operator: allow setting ceph config options via ceph cluster crd
This implements the "Ceph Config via Ceph Cluster CRD" design document
as a `cephConfig:` structure on the CRD.
This also fixes the `yq` commands used to manipulate the
`cluster-test.yaml` that caused CI issues for this PR and potentially
unknowingly others.

Signed-off-by: Alexander Trost <galexrt@googlemail.com>
2023-12-02 00:05:48 +01:00
Travis Nielsen 25f5a23072 Merge pull request #13196 from parth-gr/external-rados
external: add support for rados namespace for external cluster
2023-12-01 12:21:25 -07:00
Blaine Gardner f4d67fd4db Merge pull request #13261 from subhamkrai/remove-controller-runtime
core: remove webhook & controller-runtime from apis
2023-12-01 09:55:34 -07:00
subhamkrai 28cc1ebc55 core: remove webhook & controller-runtime from apis
This commits removes controller-runtime dependencies
from the apis dir and to achieve that we are removing
webhook.

Signed-off-by: subhamkrai <srai@redhat.com>
2023-12-01 14:15:40 +05:30
parth-gr 939a1b7d20 subvolumegroup: add name spec in subvolumegroup
Originally we create it using this cmd
ceph fs subvolume create <vol_name> <subvol_name>
So we can have 2 variables filesystem and subvolume name,
Currently the CR doesn't allow us to make subvolume-name
as constant as needed to "csi" because of k8s limitations

Signed-off-by: parth-gr <paarora@redhat.com>
2023-11-30 20:39:11 +05:30
parth-gr 5bf3847b13 namespace: add name spec in cephBlockPoolRadosNamespace CRD
instead of use metadata name as the backend resource name
we need to add a new name in the spec which can be the
actual backend name

Closes: https://github.com/rook/rook/issues/13220

Signed-off-by: parth-gr <paarora@redhat.com>
2023-11-29 22:35:21 +05:30
Rakshith R 59cb0dd4bf csi: add CSIDriverOptions section in cephCluster CR
This commit adds new CSIDriverOptions section in
cephCluster CR. This section contains settings
for read affinity and kernel+fuse Mount options
These settings will be injected directly into
rook-ceph-csi-config cm to be applicable per
ceph cluster.

Signed-off-by: Rakshith R <rar@redhat.com>
2023-11-29 19:34:28 +05:30
parth-gr df7c401d35 external: add support for rados namespace for external cluster
change the json output of storageclass to also inclusde rados namespace,
and also added the changes in the csi secret

Signed-off-by: parth-gr <paarora@redhat.com>
2023-11-28 11:52:30 +05:30
Ryotaro Banno 0ad88e9d68 core: add pgHealthyRegex to DisruptionManagementSpec
This patch adds `pgHealthyRegex` field to DisruptionManagementSpec.
`pgHealthyRegex` is a regular expression that is used to determine which
PG states should be considered healthy. The default value of
`pgHealthyRegex` is:

        ^(active\+clean|active\+clean\+scrubbing|active\+clean\+scrubbing\+deep)$

which is effectively the same as before.

Signed-off-by: Ryotaro Banno <ryotaro.banno@gmail.com>
2023-11-22 00:19:04 +00:00
travisn 03d077aa6b core: remove support for ceph pacific
Pacific is end of life and no longer necessary to
support in Rook with v1.13.

Signed-off-by: travisn <tnielsen@redhat.com>
2023-11-14 17:07:03 -07:00
Shachar Sharon 7c3e91390e nfs: add optional LivenessProbe to GaneshaServerSpec
Allow end-user to have full control on NFS-Ganesha liveness probe:

  - Disabled
  - Enabled, using user-provided definition
  - Enabled, using Rook's default liveness probe for Ganesha.

Signed-off-by: Shachar Sharon <ssharon@redhat.com>
2023-11-13 16:21:24 +02:00
Alexander Gil d6edc3bba4 docs: fix user capability name in object store user
Signed-off-by: Alexander Gil <pando855@gmail.com>
2023-11-02 11:58:39 +01:00
Satoru Takeuchi bab6e01d37 doc: add the table to know supported osd configuration
It's convenient to visualize the supported OSD configurations.
In addition, remove the reference to the common issue section
about OSD on LV-backed PVC because this issue doesn't happen anymore
for new OSDs.

Closes: https://github.com/rook/rook/issues/10859

Signed-off-by: Satoru Takeuchi <satoru.takeuchi@gmail.com>
2023-10-25 20:10:35 +00:00
sp98 ffd69291f1 osd: use bash script if restart interval is > 0
If `FlappingRestartIntervalHours` for OSD is > 0 in the cephCluster CR,
then start OSD inside a bash script.

Signed-off-by: sp98 <sapillai@redhat.com>
2023-10-11 11:38:26 +05:30
Travis Nielsen b7eaba318c Merge pull request #12941 from subhamkrai/update-caps
external: update healthchecker caps for rbd command
2023-09-27 22:55:31 -06:00
subhamkrai c78bd1cbc6 external: update healthchecker caps for rbd command
when creating networkFence CR, it requires IP's to block
which we get from running `rbd status ...` command. But,
the client.healthchecker user didn't had the right caps
to run hence it was giving error. Now, adding the required
caps `profile rbd-read-only` to osd so that rbd command can
be executed.

Signed-off-by: subhamkrai <srai@redhat.com>
2023-09-28 08:21:21 +05:30
Redouane Kachach c763935436 mgr: adding support for prometheus endpoint configuration
Using the new configuration parameters prometheusEndpoint and
prometheusEndpointSSLVerify users can configure dashboard to
point to their prometheus instance.

closes: #12876

Signed-off-by: Redouane Kachach <rkachach@redhat.com>
2023-09-27 11:07:18 +02:00
Travis Nielsen a3dafbe9cf Merge pull request #12715 from sp98/osd-sleep
osd: make osd pod to sleep when osds are flapping
2023-09-15 12:24:08 -06:00
sp98 4eb9f62205 osd: make osd pod to sleep when osds are flapping
When OSDs flap, ceph stops the OSD daemon if its marked down greater than
5 times in 600 seconds. But OSD pod restarts and marks the OSD `up` again.
This causes the PGs mapped to these OSDs to peer. While the PGs are peering,
IO to these PGs are blocked.

So we need to ensure that if ceph is marking OSD `down` due to flapping, OSD pod
should not restart to mark the OSDs `up` again.

This PR adds a sleep to the OSD pod if the container returned with a 0 exit code
Default behavior is to sleep for 6 hrs. But user can configure it from the
ceph cluster spec.

Signed-off-by: sp98 <sapillai@redhat.com>
2023-09-15 22:59:46 +05:30
travisn 66547d27be mgr: allow more than two mgrs
While two mgrs is considered sufficient, more mgr daemons are
now allowed in case the admin wants even more fault tolerance
to the active mgr going down, in case multiple mgrs go down.
Up to five mgr pods will be allowed. All mgrs will be in
standby mode except one active mgr. All mgr pods have a sidecar
that will update their respective pod specs with the active
or passive label.

Signed-off-by: travisn <tnielsen@redhat.com>
2023-09-14 13:29:11 -06:00
Micah Nagel 0ff224d453 helm: update docs to reflect reef support
some helm comments incorrectly indicate ceph v18 is not supported

Signed-off-by: Micah Nagel <micah.nagel@defenseunicorns.com>
2023-09-08 12:11:58 -06:00
Blaine Gardner 17f0072d9d Merge pull request #12778 from BlaineEXE/multus-allow-cidr-spec
multus: allow using NADs without inspectable CIDRs
2023-09-07 13:42:41 -06:00
Blaine Gardner 3c43268d0a multus: detect network CIDRs via canary
Change how Rook detects network CIDRs for Multus networks. The IPAM
configuration is only defined as an arbitrary string JSON blob with a
"type" field and nothing more. Rook's detection of CIDRs for whereabouts
had already grown out of date since the initial implementation.
Additionally, Rook did not support DHCP IPAM, which is a reasonable
choice for users. And more, Rook did not support CNI plugin chaining,
which further complicates NADs. Based on the CNI spec, network chaning
can result in any changes to network CIDRs from the first-given plugin.

All these problems make it more and more difficult for Rook to support
Multus by inspecting the NAD itself to predict network CIDRs. Instead,
it is better for Rook to treat the CNI process as a black box. To
preserve legacy functionality of auto-detecting networks and to make
that as robust as possible, change to a canary-style architecture like
that used for Ceph mons, from which Rook will detect the network CIDRs
if possible.

Also allow users to specify overrides for CIDR ranges. This allows Rook
to still support esoteric and unexpected NAD or network configurations
where a CIDR range is not detectable or where the range detected would
be incomplete. Because it may be impossible for Rook to understand the
network CIDRs wholistically while residing only on a portion of the
network, this feature should have been present from Multus's inception.

Improving CIDR auto-detection and allowing users to specify overrides
for auto-detected CIDRs rounds out Rook's Multus support for CephCluster
(core/RADOS) installations. No further architectural changes should be
needed for CephClusters as regards application of public/cluster network
CIDRs for Multus networks.

Signed-off-by: Blaine Gardner <blaine.gardner@ibm.com>
2023-09-07 10:12:55 -06:00
Travis Nielsen a06fd8a54d Merge pull request #12811 from parth-gr/external-name
external: change cluster_name to k8s_cluster_name
2023-08-31 08:14:00 -06:00
parth-gr 98c4567219 external: change cluster_name to k8s_cluster_name
it was confusing if we are calling a ceph cluster name
or k8s cluster name, so re-named

Signed-off-by: parth-gr <paarora@redhat.com>
2023-08-30 14:07:58 +05:30
Travis Nielsen 12023cf72e Merge pull request #12598 from batrick/cephsqlite-doc-fix
docs: correct some libcephsqlite statements
2023-08-25 14:09:51 -06:00
travisn 2b1cc4c7b8 object: allow creating an object user in different namespace
The object user was previously required to be created in the
same namespace as the object store and the cluster. Now,
the object user can be reconciled even in a different namespace
from the cluster and object store. The namespace would be specified
in the object user CR.

Signed-off-by: travisn <tnielsen@redhat.com>
2023-08-25 11:54:57 -06:00
Travis Nielsen 0e3117bd6e Merge pull request #12507 from sp98/osd-migration
Replace OSD to use new backend store
2023-08-17 12:40:49 -06:00
Travis Nielsen afaadd78c4 Merge pull request #12384 from ideepika/feature/ksd-174
mon: add mon storage class support
2023-08-17 11:52:20 -06:00
sp98 886bb357ef osd: replace osd to use new backend store
If osd store is updated in the ceph cluster, then
delete OSDs one by one, cleanup disks and provision a new OSD on
the same disk

Signed-off-by: sp98 <sapillai@redhat.com>
2023-08-17 21:40:55 +05:30
Blaine Gardner 853bab398a Merge pull request #12495 from Nikhil-Ladha/issue12172/multus-validation-job
multus: Add sample job manifest for multus config validation
2023-08-11 13:28:07 -06:00