rgw doesn't respond `livenessProbe` if the number of connection reaches its
limit (by default, 1000). Then rgw is out of service but still live.
Hense the current `livenessProbe` logic is suitiable for `readinessProbe`.
`tcpSocket` is enough for `livenessProbe`.
Closes: #8407
Signed-off-by: Yuzuki Mimura <yuzuki725.m@gmail.com>
Signed-off-by: Satoru Takeuchi <satoru.takeuchi@gmail.com>
Ths NFS spec now supports the CephBlockPool spec which means that it can
take advantage of all the known settings like compression, size, failure
domain etc.
Closes: https://github.com/rook/rook/issues/9034
Signed-off-by: Sébastien Han <seb@redhat.com>
We can now set labels to the crash collector deployment by editing the
CephCluster CR with:
```yaml
spec:
labels:
crashcollector:
```
Closes: https://github.com/rook/rook/issues/9039
Signed-off-by: Sébastien Han <seb@redhat.com>
replaces all occurences of lduo/rduo quotation marks to make
sure that using the snippets in a k8s manifest will work
This fixes an issue with ArgoCD not being able
to apply `common.yaml` because of an encoding issue
Signed-off-by: PixelJonas <jonas@janz.digital>
The rook-ceph-purge-osd job should be run for a particular CephCluster
and should therefore be namespaced to the cluster.
Signed-off-by: Blaine Gardner <blaine.gardner@redhat.com>
When deploying a cluster-wde encrypted cluster we now use the rook
binary to execute some code to fetch the key encryption key.
Using cURL all the time to fetch the key has its limitations. The
incoming integration with Kubernetes Authentication through service
accounts is leading the usage of cURL to its end.
The logic is really complex and prone to errors. Re-implementing the
Go logic into Bash is not a viable option.
Also, using the lib in a binary allows us to keep a consistent behavior
throughout the life cycle of our code.
Signed-off-by: Sébastien Han <seb@redhat.com>
Create a Helm library chart to share content between the Operator Helm
chart (rook-ceph) and the CephCluster (etc.) Helm chart
(rook-ceph-cluster).
This will ensure that when resources related to the CephCluster change,
notably RBAC resources, they will change for both charts.
Signed-off-by: Blaine Gardner <blaine.gardner@redhat.com>
CephMgrIsAbsent
----------------
This alert initially had the following query
absent(up{job="rook-ceph-mgr"})
which will fire when the 'up' query is not present, but had two flows
a. it will not be fired if 'up' provides a result with ZERO value
b. it will not give any fields in the metric, so 'namespace' was missing
when the above query was replaced with the following,
up{job="rook-ceph-mgr"} == 0
query had the following shortage
a. whenever mgr pod is completely down (like 'replicas' set to ZERO
and 'mgr' is not coming up), 'up' query will not give any result.
Thus we had to combine both the queries to get results in both the scenarios.
CephMgrIsMissingReplicas
------------------------
This query previously was,
sum(up{job="rook-ceph-mgr"}) < 1
had the same structure as the above (Absent) query, but it's
intention was to check the no: of 'replicas' count for ceph mgr.
Now it is changed to a kube query which handles the replicas count.
Signed-off-by: Arun Kumar Mohan <amohan@redhat.com>
Using a default value for CompressionMode to none effectively overrides
any values for Parameters. It is deprecated but still takes precedence.
Which means that in its previous form, Parameters was always ignored
since CompressionMode was always set to none when empty.
Signed-off-by: Sébastien Han <seb@redhat.com>
The volume replication CRDs are an external component, not owned by Rook.
Therefore, they should be installed as any other independent component
in case the admin will install other consumers of the volumereplication CRDs
in the future in addition to Rook and the CSI driver.
Signed-off-by: Travis Nielsen <tnielsen@redhat.com>
This commit updates the CephNFS CR to make the RADOS settings optional
for Ceph versions above 16.2.7 due to the NFS module changes in Ceph.
The changes in Ceph make it so the RADOS pool is always ".nfs" and the
RADOS namespace is always the name of the NFS cluster.
This commit also handles the changes in Ceph Pacific versions before 16.2.7
where the default pool name is "nfs-ganesha" instead of ".nfs".
Closes: https://github.com/rook/rook/issues/8450
Signed-off-by: Joseph Sawaya <jsawaya@redhat.com>
As 'CephMonHighNumberOfLeaderChanges' means there is a
multiple monitor election and indicate some communication
issue between the monitors. Increasing interval timing for
this alert is not considered as safe. So reverting this
change back to 5m
Signed-off-by: Arun Kumar Mohan <amohan@redhat.com>
Create a new log level for Rook that is hidden from users. This is the
most verbose log level, and it is the level developers would like to use
to get debug logs that are important for debugging but that could leak
senstivie information like credentials in production use.
If a user sets their debug level to "TRACE", they will merely get
"DEBUG" level logs. Only if they set "TRACE_INSECURE" will they get
trace logs, and those are likely to include insecure information. Rook
tries very hard not to leak sensitive information in logs even with
verbose "DEBUG" logs.
Resolves#8778
Signed-off-by: Blaine Gardner <blaine.gardner@redhat.com>
The following alerts,
CephMonHighNumberOfLeaderChanges
CephOSDDiskNotResponding
CephClusterWarningState
, which are resolved automatically, in most cases,
are causing unnecessary admin events. So we are increasing the
alert delay time to '15m'.
Signed-off-by: Arun Kumar Mohan <amohan@redhat.com>
Instead of using 'absent' query, we are trying to use 'up' which should
provide us with the needed 'namespace' field in the resultant metrics
Signed-off-by: aruniiird <amohan@redhat.com>
While build.all we have race condition where images.txt
file was update with duplicate entries. So, removing `-a`
from 1st tee and also add `sort -h`and `uniq` command to
confirm no duplicate entry is in the file.
Signed-off-by: subhamkrai <srai@redhat.com>
If we call the script with --monitoring-endpoint
10.1.8.29,10.1.8.36,10.1.8.25, we need to parse each comma-separated
address and not the entire block.
Now each address is tested properly.
Signed-off-by: Sébastien Han <seb@redhat.com>
The purge osd job need to read and update persistentvolumeclaims
While the job doesn't failed with osd and deployments well deleted
The associated pvc is not deleted by the job as it failed to read it
Signed-off-by: n.fraison <n.fraison@criteo.com>
We don't need to use tini.
We don't have anything in the rook operator that would
either create zombie processes (no threads) or use
exec (to fork). The Go binary has a really good
signal handling mechanism.
Closes: https://github.com/rook/rook/issues/8794
Signed-off-by: Sébastien Han <seb@redhat.com>
Reduce the RBAC scope for the lib-bucket-provisioner to the minimum
necessary for Rook to operate with lib-bucket-provisioner. Comment the
RBAC so that users can make informed decisions about whether or not they
can remove the permissions for their uses.
Signed-off-by: Blaine Gardner <blaine.gardner@redhat.com>
Travis Nielsen
Arun Kumar Mohan
Yuzuki Mimura
Sébastien Han
subhamkrai
Blaine Gardner
Yug Gupta
PixelJonas
Madhu Rajanna
Gowtham Shanmugasundaram
parth-gr
Joseph Sawaya
Blaine Gardner
Satoru Takeuchi
n.fraison