* Wait until failover finishes during master pod shutdown
This improves on #5528 by checking and waiting until the failover is
finished on both the redis and the sentinel container. This completely
eliminates momentary service interruption during rollouts.
As we cannot guarantee the failover will be successful the wait time
is capped by the termination grace period - 10s.
* Separate terminationGracePeriod setings for each pod type
* make the use of REDISCLI_AUTH clear
* [bitnami/redis] Update components versions
Signed-off-by: Bitnami Containers <containers@bitnami.com>
Co-authored-by: Bitnami Containers <containers@bitnami.com>
* Fix for NetworkPolicy blocking sentinel metrics exporter
When sentinel metrics exporter is enabled, that metrics port needs to be added to the network policy
* inc chart version
* Fixed port to not be hard coded
* Add Redis Sentinel Exporter
Add support to get Prometheus metrics via the Redis Sentinel Exporter.
For that the sentinel configuration contains a new section metrics,
which adds an additional sidecar with the Redis Sentinel Exporter, when
enabled.
* Remove trailing spaces
* Apply suggestions from the PR review
* Fix linter error
* Fix component name for Redis Sentinel metrics
* Update PR with bitnami/redis-sentinel-exporter
* Redis Sentinel metrics service port configurable
* Update README.md with default port 9355
* Apply review suggestions
* Apply review suggestions
* Apply review suggestions
* Apply review suggestions
* Apply review suggestions
* Apply review suggestions
Co-authored-by: Miguel Ruiz <miruiz@bitnami.com>
* exporter presents client certs to server only in mtls context
* Allow to set redis hostname to better deal with certificate constrainsts
* set user only if require pwd
* [bitnami/redis] Minor version bump
Co-authored-by: hightoxicity <tony.fouchard@prevision.io>
Co-authored-by: Miguel Ruiz <miruiz@vmware.com>
* fix a typo in the redis readme.MD file for the usage of metrics.extraArgs
* bump version for redis chart
* Update Chart.yaml
In this case it is not needed to bump the chart version
Co-authored-by: Dong, Hujun <Hujun.Dong@IGT.com>
Co-authored-by: Carlos Rodríguez Hernández <carlosrh@vmware.com>
* update redis README.md for the usage of metrics.extraArgs when redis.tls is enabled
* bump redis Chart version
Co-authored-by: Dong, Hujun <Hujun.Dong@IGT.com>
* Fix for slave volumePermissions init container command
The command of the volumePermissions init container which is run by the slave pod should by run with `/bin/bash -cx`.
Previously the slave init container execution always failed. After this fix it started working. Tested on minikube.
* Pump PATCH version
* Pump PATCH version of chart version not appVersion
* Dont print evaluated shell commands in initContainer
This makes it more consistent with the other executions of `/bin/sh`
* Consistently use -ec options for shell command in volumePermissions initContainer
* Pump PATCH version of chart version (again after merging master)
* Used /bin/bash -ec consistently for volumePermissions initContainers
Co-authored-by: David Leonhartsberger <david.leonhartsberger@eventworld.com>
* Added preExecCmds to start-node.sh
* Updated chart version
* Update chart version to 12.5.2
* Added preExecCmds to start-node.sh
* Updated chart version
* Update chart version to 12.5.2
* New chart verseion
Add label and annotations for Statefulset volumeClaimTemplates
Add default label and annotation values.
Add Parameters to README.ME
Bump minor version in chart.
* Add trademark to Redis and Redis Cluster
* add trademark in other charts
* Fix table indentation in READMEs
* Add trademark to yaml, tpl and txt
* Rebase master and bump chart version
* One line disclaimer
* Add trademark to recent changes and add zookeeper and cassandra
Co-authored-by: jotamartos <jotamartos@vmware.com>
* Add annotations/labels for Redis Sentinel
Similar to the configuration for the persistent volume, we are now using
the annotations and labels from the slave section for the StatefulSet.
* Fix the used path and subpath for Redis Sentinel
The Redis Sentinel StatefulSet uses the values from the slave section to
configure the persistent volumes, but a static respectively the master
values in the volume mounts.
This should be fixed by using the configured values from the slave
section.
