Implement the first step of `design/ceph/resource-dependencies.md` to
add dependency checking when deleting a CephCluster.
Signed-off-by: Blaine Gardner <blaine.gardner@redhat.com>
This commit changes CSI_ENABLE_HOST_NETWORK default value to true
since it was observed that cephfsvolume gets blocked when csi cephfs
nodeplugin is restarted and nodeunpublish call also hangs when
cephfs nodeplugin is using pod networking.
Updates: #8085
Signed-off-by: Rakshith R <rar@redhat.com>
Currently we only check the Major, Minor, Extra and Build number in the ceph version to determine if all daemons should be upgraded or not. This PR also checks the commit hash for upgrading all the daemons.
This will be useful in cases where build number has not changed due to hotfix. For example:
existing ceph version - ceph version 14.2.11-139.el8cp (b8e1f91c99491fb2e5ede748a1c0738ed158d0f5) nautilus (stable)
new ceph version with hotfix - ceph version 14.2.11-139.0.hotfix.bz1959254.el8cp (5c0dc966af809fd1d429ec7bac48962a746af243) nautilus (stable)
Signed-off-by: Santosh Pillai <sapillai@redhat.com>
We now create and delete pools in parallel.
Before this patch, the deletion:
```
2021-06-08 13:48:26.557328 I | cephclient: no images/snapshosts present in pool "my-store.rgw.control"
2021-06-08 13:48:26.557383 I | cephclient: purging pool "my-store.rgw.control" (id=1)
2021-06-08 13:48:27.030021 I | ceph-object-controller: done disabling the dashboard api secret key
2021-06-08 13:48:28.800661 I | cephclient: purge completed for pool "my-store.rgw.control"
2021-06-08 13:48:29.085744 I | cephclient: no images/snapshosts present in pool "my-store.rgw.meta"
2021-06-08 13:48:29.085762 I | cephclient: purging pool "my-store.rgw.meta" (id=3)
2021-06-08 13:48:30.835487 I | cephclient: purge completed for pool "my-store.rgw.meta"
2021-06-08 13:48:31.128349 I | cephclient: no images/snapshosts present in pool "my-store.rgw.log"
2021-06-08 13:48:31.128368 I | cephclient: purging pool "my-store.rgw.log" (id=4)
2021-06-08 13:48:32.864376 I | cephclient: purge completed for pool "my-store.rgw.log"
2021-06-08 13:48:33.180923 I | cephclient: no images/snapshosts present in pool "my-store.rgw.buckets.index"
2021-06-08 13:48:33.181049 I | cephclient: purging pool "my-store.rgw.buckets.index" (id=5)
2021-06-08 13:48:34.892163 I | cephclient: purge completed for pool "my-store.rgw.buckets.index"
2021-06-08 13:48:35.179952 I | cephclient: no images/snapshosts present in pool "my-store.rgw.buckets.non-ec"
2021-06-08 13:48:35.179994 I | cephclient: purging pool "my-store.rgw.buckets.non-ec" (id=6)
2021-06-08 13:48:37.376067 I | cephclient: purge completed for pool "my-store.rgw.buckets.non-ec"
2021-06-08 13:48:37.665201 I | cephclient: no images/snapshosts present in pool "my-store.rgw.buckets.data"
2021-06-08 13:48:37.665218 I | cephclient: purging pool "my-store.rgw.buckets.data" (id=8)
2021-06-08 13:48:39.394717 I | cephclient: purge completed for pool "my-store.rgw.buckets.data"
2021-06-08 13:48:39.725528 I | cephclient: no images/snapshosts present in pool ".rgw.root"
2021-06-08 13:48:39.725567 I | cephclient: purging pool ".rgw.root" (id=7)
2021-06-08 13:48:41.424628 I | cephclient: purge completed for pool ".rgw.root"
```
It took 15sec to cleanup...
Now with this patch:
```
2021-06-08 14:15:25.621472 I | cephclient: no images/snapshosts present in pool "my-store.rgw.buckets.index"
2021-06-08 14:15:25.621570 I | cephclient: purging pool "my-store.rgw.buckets.index" (id=12)
2021-06-08 14:15:25.668708 I | cephclient: no images/snapshosts present in pool "my-store.rgw.log"
2021-06-08 14:15:25.668729 I | cephclient: purging pool "my-store.rgw.log" (id=11)
2021-06-08 14:15:25.693002 I | cephclient: no images/snapshosts present in pool "my-store.rgw.meta"
2021-06-08 14:15:25.693050 I | cephclient: purging pool "my-store.rgw.meta" (id=10)
2021-06-08 14:15:25.698830 I | cephclient: no images/snapshosts present in pool "my-store.rgw.control"
2021-06-08 14:15:25.698854 I | cephclient: purging pool "my-store.rgw.control" (id=9)
2021-06-08 14:15:25.701732 I | cephclient: no images/snapshosts present in pool "my-store.rgw.buckets.non-ec"
2021-06-08 14:15:25.701758 I | cephclient: purging pool "my-store.rgw.buckets.non-ec" (id=13)
2021-06-08 14:15:25.702816 I | cephclient: no images/snapshosts present in pool ".rgw.root"
2021-06-08 14:15:25.702836 I | cephclient: purging pool ".rgw.root" (id=14)
2021-06-08 14:15:25.717144 I | cephclient: no images/snapshosts present in pool "my-store.rgw.buckets.data"
2021-06-08 14:15:25.717212 I | cephclient: purging pool "my-store.rgw.buckets.data" (id=15)
2021-06-08 14:15:26.393246 I | ceph-object-controller: done disabling the dashboard api secret key
```
It tool around 1sec.
The creation before this patch:
```
2021-06-08 16:47:10.669484 I | ceph-spec: adding finalizer "cephobjectstore.ceph.rook.io" on "my-store"
2021-06-08 16:47:10.677253 E | ceph-object-controller: failed to set object store "rook-ceph/my-store" status to "Progressing". failed to update object "my-store" status: Operation cannot be fulfilled on cephobjectstores.ceph.rook.io "my-store": the object has been modified; please apply your changes to the latest version and try again
2021-06-08 16:47:10.682231 I | op-mon: parsing mon endpoints: b=10.111.63.108:6789,c=10.108.123.222:6789,a=10.100.223.211:6789
2021-06-08 16:47:10.985002 I | ceph-object-controller: reconciling object store deployments
2021-06-08 16:47:11.002364 I | ceph-object-controller: ceph object store gateway service running at 10.99.31.5
2021-06-08 16:47:11.002384 I | ceph-object-controller: reconciling object store pools
2021-06-08 16:47:14.336086 I | cephclient: setting pool property "compression_mode" to "none" on pool "my-store.rgw.control"
2021-06-08 16:47:16.115072 E | ceph-crashcollector-controller: node reconcile failed on op "unchanged": Operation cannot be fulfilled on deployments.apps "rook-ceph-crashcollector-minikube": the object has been modified; please apply your changes to the latest version and try again
2021-06-08 16:47:16.138224 E | ceph-crashcollector-controller: node reconcile failed on op "unchanged": Operation cannot be fulfilled on deployments.apps "rook-ceph-crashcollector-minikube": the object has been modified; please apply your changes to the latest version and try again
2021-06-08 16:47:16.391914 I | cephclient: creating replicated pool my-store.rgw.control succeeded
2021-06-08 16:47:16.391943 I | cephclient: setting pool property "pg_num_min" to "8" on pool "my-store.rgw.control"
2021-06-08 16:47:20.390562 I | cephclient: setting pool property "compression_mode" to "none" on pool "my-store.rgw.meta"
2021-06-08 16:47:22.423269 I | cephclient: creating replicated pool my-store.rgw.meta succeeded
2021-06-08 16:47:22.423294 I | cephclient: setting pool property "pg_num_min" to "8" on pool "my-store.rgw.meta"
2021-06-08 16:47:26.502874 I | cephclient: setting pool property "compression_mode" to "none" on pool "my-store.rgw.log"
2021-06-08 16:47:28.555252 I | cephclient: creating replicated pool my-store.rgw.log succeeded
2021-06-08 16:47:28.555308 I | cephclient: setting pool property "pg_num_min" to "8" on pool "my-store.rgw.log"
2021-06-08 16:47:32.619546 I | cephclient: setting pool property "compression_mode" to "none" on pool "my-store.rgw.buckets.index"
2021-06-08 16:47:34.663020 I | cephclient: creating replicated pool my-store.rgw.buckets.index succeeded
2021-06-08 16:47:34.663050 I | cephclient: setting pool property "pg_num_min" to "8" on pool "my-store.rgw.buckets.index"
2021-06-08 16:47:38.712886 I | cephclient: setting pool property "compression_mode" to "none" on pool "my-store.rgw.buckets.non-ec"
2021-06-08 16:47:40.748430 I | cephclient: creating replicated pool my-store.rgw.buckets.non-ec succeeded
2021-06-08 16:47:40.748467 I | cephclient: setting pool property "pg_num_min" to "8" on pool "my-store.rgw.buckets.non-ec"
2021-06-08 16:47:44.830659 I | cephclient: setting pool property "compression_mode" to "none" on pool ".rgw.root"
2021-06-08 16:47:46.875834 I | cephclient: creating replicated pool .rgw.root succeeded
2021-06-08 16:47:46.875863 I | cephclient: setting pool property "pg_num_min" to "8" on pool ".rgw.root"
2021-06-08 16:47:50.933717 I | cephclient: setting pool property "compression_mode" to "none" on pool "my-store.rgw.buckets.data"
2021-06-08 16:47:52.969461 I | cephclient: creating replicated pool my-store.rgw.buckets.data succeeded
2021-06-08 16:47:52.969549 I | ceph-object-controller: setting multisite settings for object store "my-store"
2021-06-08 16:47:53.585737 I | ceph-object-controller: Multisite for object-store: realm=my-store, zonegroup=my-store, zone=my-store
2021-06-08 16:47:53.585777 I | ceph-object-controller: multisite configuration for object-store my-store is complete
2021-06-08 16:47:53.585787 I | ceph-object-controller: creating object store "my-store" in namespace "rook-ceph"
2021-06-08 16:47:53.585802 I | cephclient: getting or creating ceph auth key "client.rgw.my.store.a"
2021-06-08 16:47:53.930416 I | ceph-object-controller: setting rgw config flags
2021-06-08 16:47:53.931363 I | op-config: setting "client.rgw.my.store.a"="rgw_enable_usage_log"="true" option to the mon configuration database
2021-06-08 16:47:54.201200 I | op-config: successfully set "client.rgw.my.store.a"="rgw_enable_usage_log"="true" option to the mon configuration database
2021-06-08 16:47:54.201225 I | op-config: setting "client.rgw.my.store.a"="rgw_zone"="my-store" option to the mon configuration database
2021-06-08 16:47:54.454589 I | op-config: successfully set "client.rgw.my.store.a"="rgw_zone"="my-store" option to the mon configuration database
2021-06-08 16:47:54.454614 I | op-config: setting "client.rgw.my.store.a"="rgw_zonegroup"="my-store" option to the mon configuration database
2021-06-08 16:47:54.710218 I | op-config: successfully set "client.rgw.my.store.a"="rgw_zonegroup"="my-store" option to the mon configuration database
2021-06-08 16:47:54.710237 I | op-config: setting "client.rgw.my.store.a"="rgw_log_nonexistent_bucket"="true" option to the mon configuration database
2021-06-08 16:47:54.969049 I | op-config: successfully set "client.rgw.my.store.a"="rgw_log_nonexistent_bucket"="true" option to the mon configuration database
2021-06-08 16:47:54.969069 I | op-config: setting "client.rgw.my.store.a"="rgw_log_object_name_utc"="true" option to the mon configuration database
2021-06-08 16:47:55.255491 I | op-config: successfully set "client.rgw.my.store.a"="rgw_log_object_name_utc"="true" option to the mon configuration database
2021-06-08 16:47:55.255631 I | ceph-object-controller: object store "my-store" deployment "rook-ceph-rgw-my-store-a" started
2021-06-08 16:47:55.279316 I | ceph-object-controller: enabling rgw dashboard
2021-06-08 16:47:55.361208 E | ceph-crashcollector-controller: node reconcile failed on op "unchanged": Operation cannot be fulfilled on deployments.apps "rook-ceph-crashcollector-minikube": the object has been modified; please apply your changes to the latest version and try again
2021-06-08 16:47:56.225904 I | ceph-object-controller: setting the dashboard api secret key
2021-06-08 16:47:56.225984 I | ceph-object-controller: created object store "my-store" in namespace "rook-ceph"
2021-06-08 16:47:56.589452 I | ceph-object-controller: starting rgw healthcheck
2021-06-08 16:47:56.650045 I | ceph-object-controller: done setting the dashboard api secret key
```
It took 46sec, and I've seen it taking almost a 1min sometimes.
Now with this patch:
```
2021-06-08 16:51:35.259558 I | ceph-spec: adding finalizer "cephobjectstore.ceph.rook.io" on "my-store"
2021-06-08 16:51:35.270524 E | ceph-object-controller: failed to set object store "rook-ceph/my-store" status to "Progressing". failed to update object "my-store" status: Operation cannot be fulfilled on cephobjectstores.ceph.rook.io "my-store": the object has been modified; please apply your changes to the latest version and try again
2021-06-08 16:51:35.274387 I | op-mon: parsing mon endpoints: b=10.111.63.108:6789,c=10.108.123.222:6789,a=10.100.223.211:6789
2021-06-08 16:51:35.599023 I | ceph-object-controller: reconciling object store deployments
2021-06-08 16:51:35.607256 I | ceph-object-controller: ceph object store gateway service running at 10.104.254.110
2021-06-08 16:51:35.607315 I | ceph-object-controller: reconciling object store pools
2021-06-08 16:51:39.337735 I | cephclient: setting pool property "compression_mode" to "none" on pool "my-store.rgw.buckets.non-ec"
2021-06-08 16:51:40.343290 I | cephclient: setting pool property "compression_mode" to "none" on pool ".rgw.root"
2021-06-08 16:51:40.346335 I | cephclient: setting pool property "compression_mode" to "none" on pool "my-store.rgw.buckets.index"
2021-06-08 16:51:40.362670 I | cephclient: setting pool property "compression_mode" to "none" on pool "my-store.rgw.meta"
2021-06-08 16:51:40.363445 I | cephclient: setting pool property "compression_mode" to "none" on pool "my-store.rgw.log"
2021-06-08 16:51:40.364321 I | cephclient: setting pool property "compression_mode" to "none" on pool "my-store.rgw.control"
2021-06-08 16:51:41.412870 I | cephclient: creating replicated pool my-store.rgw.buckets.non-ec succeeded
2021-06-08 16:51:41.412904 I | cephclient: setting pool property "pg_num_min" to "8" on pool "my-store.rgw.buckets.non-ec"
2021-06-08 16:51:42.460439 I | cephclient: creating replicated pool my-store.rgw.control succeeded
2021-06-08 16:51:42.460503 I | cephclient: setting pool property "pg_num_min" to "8" on pool "my-store.rgw.control"
2021-06-08 16:51:42.470666 I | cephclient: creating replicated pool my-store.rgw.meta succeeded
2021-06-08 16:51:42.470727 I | cephclient: setting pool property "pg_num_min" to "8" on pool "my-store.rgw.meta"
2021-06-08 16:51:42.473029 I | cephclient: creating replicated pool .rgw.root succeeded
2021-06-08 16:51:42.473064 I | cephclient: setting pool property "pg_num_min" to "8" on pool ".rgw.root"
2021-06-08 16:51:42.473895 I | cephclient: creating replicated pool my-store.rgw.buckets.index succeeded
2021-06-08 16:51:42.473923 I | cephclient: setting pool property "pg_num_min" to "8" on pool "my-store.rgw.buckets.index"
2021-06-08 16:51:42.478625 I | cephclient: creating replicated pool my-store.rgw.log succeeded
2021-06-08 16:51:42.478671 I | cephclient: setting pool property "pg_num_min" to "8" on pool "my-store.rgw.log"
2021-06-08 16:51:46.606905 I | cephclient: setting pool property "compression_mode" to "none" on pool "my-store.rgw.buckets.data"
2021-06-08 16:51:48.627320 I | cephclient: creating replicated pool my-store.rgw.buckets.data succeeded
2021-06-08 16:51:48.627368 I | ceph-object-controller: setting multisite settings for object store "my-store"
2021-06-08 16:51:49.325086 I | ceph-object-controller: Multisite for object-store: realm=my-store, zonegroup=my-store, zone=my-store
2021-06-08 16:51:49.325108 I | ceph-object-controller: multisite configuration for object-store my-store is complete
2021-06-08 16:51:49.325121 I | ceph-object-controller: creating object store "my-store" in namespace "rook-ceph"
2021-06-08 16:51:49.325134 I | cephclient: getting or creating ceph auth key "client.rgw.my.store.a"
2021-06-08 16:51:49.657898 I | ceph-object-controller: setting rgw config flags
2021-06-08 16:51:49.657920 I | op-config: setting "client.rgw.my.store.a"="rgw_log_object_name_utc"="true" option to the mon configuration database
2021-06-08 16:51:49.917898 I | op-config: successfully set "client.rgw.my.store.a"="rgw_log_object_name_utc"="true" option to the mon configuration database
2021-06-08 16:51:49.917917 I | op-config: setting "client.rgw.my.store.a"="rgw_enable_usage_log"="true" option to the mon configuration database
2021-06-08 16:51:50.184939 I | op-config: successfully set "client.rgw.my.store.a"="rgw_enable_usage_log"="true" option to the mon configuration database
2021-06-08 16:51:50.184970 I | op-config: setting "client.rgw.my.store.a"="rgw_zone"="my-store" option to the mon configuration database
2021-06-08 16:51:50.463799 I | op-config: successfully set "client.rgw.my.store.a"="rgw_zone"="my-store" option to the mon configuration database
2021-06-08 16:51:50.463821 I | op-config: setting "client.rgw.my.store.a"="rgw_zonegroup"="my-store" option to the mon configuration database
2021-06-08 16:51:50.720953 I | op-config: successfully set "client.rgw.my.store.a"="rgw_zonegroup"="my-store" option to the mon configuration database
2021-06-08 16:51:50.720999 I | op-config: setting "client.rgw.my.store.a"="rgw_log_nonexistent_bucket"="true" option to the mon configuration database
2021-06-08 16:51:50.974875 I | op-config: successfully set "client.rgw.my.store.a"="rgw_log_nonexistent_bucket"="true" option to the mon configuration database
2021-06-08 16:51:50.975024 I | ceph-object-controller: object store "my-store" deployment "rook-ceph-rgw-my-store-a" started
2021-06-08 16:51:51.019820 I | ceph-object-controller: enabling rgw dashboard
2021-06-08 16:51:51.980117 I | ceph-object-controller: created object store "my-store" in namespace "rook-ceph"
2021-06-08 16:51:51.980199 I | ceph-object-controller: setting the dashboard api secret key
2021-06-08 16:51:52.399997 I | ceph-object-controller: done setting the dashboard api secret key
2021-06-08 16:51:52.435936 I | ceph-object-controller: starting rgw healthcheck
```
It took 17sec.
Signed-off-by: Sébastien Han <seb@redhat.com>
We have been having many issues with external mode with Ceph version
mismatching. The operator would have a Ceph version different than the
external cluster. The `radosgw-admin` was used to interact with S3
users, even a small version delta would cause the command to coredump.
After checking with the rgw core team it appears Rook was misusing the
CLI and the admin ops API should be used instead.
So this patch is the first introduction of go-ceph in Rook to consume
the rgw admin ops API instead of the `radosgw-admin` CLI, **only** for
user management in this initial commit.
Later we can do more such as bucket operation, zone management etc.
Closes: https://github.com/rook/rook/issues/7924
Signed-off-by: Sébastien Han <seb@redhat.com>
Fix a bug where delete events on Rook-Ceph CRs would be detected
multiple times if deletion was blocked. Old method used pointer
comparison instead of using the (metav1.Time) Equal() method.
Signed-off-by: Blaine Gardner <blaine.gardner@redhat.com>
Currently it is not possible to set resource limits based on device classes
for different OSDs. Now adding an ability to use predefined keys for main
cluster spec Resource section to reflect resource limits for different OSDs
with different device classes.
Closes: https://github.com/rook/rook/issues/8007
Signed-off-by: Denis Egorenko <degorenko@mirantis.com>
enableCSIHostNetwork was missing in the helm charts
values.yaml example. adding the same in this commit.
Signed-off-by: Madhu Rajanna <madhupr007@gmail.com>
The github actions are fully covering the build needs in master
so we can go ahead and disable Jenkins by default. In case
there is still a desired to run Jenkins for a particular PR,
the [test full] text can still trigger it.
Signed-off-by: Travis Nielsen <tnielsen@redhat.com>
Given the improvements with github actions and other improvements,
we now clean up the build instructions for developers joining the
rook project. There is no need to understand about Jenkins pipelines
anymore since we will soon be fully on github actions.
Signed-off-by: Travis Nielsen <tnielsen@redhat.com>
Changing the crds.enabled setting in a running cluster will cause the
CRs to be deleted, which could be disastrous for a cluster. Add
and clarify the warning in the docs to prevent it and link
to recovery steps if it does happen.
Signed-off-by: Travis Nielsen <tnielsen@redhat.com>
If the CephCluster CR or other CRs are mistakenly deleted,
the cluster can remain intact with steps to remove the owner
references to the cluster and re-creating the CRs.
Signed-off-by: Travis Nielsen <tnielsen@redhat.com>
Both MonSpec.Count and CephVersionSpec.Image should be optional for the
external cluster. So marking them as so.
Signed-off-by: Sébastien Han <seb@redhat.com>
This commit removes obsolete statefulset create, delete,
addlabel, templateToStatefulset functions which were part of
deprecated csi support for k8s 1.13 and should have been
removed as part of https://github.com/rook/rook/pull/5982.
It also adds Unit test for templateToDeployment func.
Signed-off-by: Rakshith R <rar@redhat.com>
update ceph teardown doc to include cleanupPolicy which enables automatic clean up of the dataDirHostPath and the drives
Signed-off-by: Santosh Pillai <sapillai@redhat.com>
Saw this today in the logs:
```
2021-06-02 13:02:54.787838 I | ceph-block-pool-controller: deleting pool "testpool"
2021-06-02 13:02:56.178094 I | cephclient: no images/snapshosts present in pool "testpool"
2021-06-02 13:02:56.178125 I | cephclient: purging pool "testpool" (id=17)
panic: runtime error: invalid memory address or nil pointer dereference
[signal SIGSEGV: segmentation violation code=0x1 addr=0x0 pc=0x1cef49f]
goroutine 1607 [running]:
github.com/rook/rook/pkg/operator/ceph/pool.(*mirrorChecker).checkMirroringHealth(0xc001a98ba0, 0x8, 0xc000d34c60)
/home/runner/work/rook/rook/pkg/operator/ceph/pool/health.go:115 +0xdf
github.com/rook/rook/pkg/operator/ceph/pool.(*mirrorChecker).checkMirroring(0xc001a98ba0, 0xc0016fd500)
/home/runner/work/rook/rook/pkg/operator/ceph/pool/health.go:82 +0x147
created by github.com/rook/rook/pkg/operator/ceph/pool.(*ReconcileCephBlockPool).reconcile
/home/runner/work/rook/rook/pkg/operator/ceph/pool/controller.go:298 +0xee5
```
Essentially, it's intermittent but when deleting the pool the
healthcheck kicked in and fetched the mirroring status, which returned
empty. THe subsequent code tried to access content of a nil pointer,
hence the error.
So now we stop monitoring first, then we proceed with the deletion.
Signed-off-by: Sébastien Han <seb@redhat.com>
CSI drivers failed to start if rook encountered error
while reading configmap with no retry.
This commit introduces retry for starting CSI driver
if there is a failure in reading configmap upto 3 times.
Fixes: #7950
Signed-off-by: Rakshith R <rar@redhat.com>
This commit adds support for adding tolerations and affinities for
cephfs and rbd (provisioner & nodeplugin) pods separately.
Closes: #7060
Signed-off-by: Rakshith R <rar@redhat.com>
Blaine Gardner
Blaine Gardner
Sébastien Han
Travis Nielsen
Rakshith R
Santosh Pillai
subhamkrai
Arun Kumar Mohan
Denis Egorenko
Madhu Rajanna
rohan47