Currently, custom metrics for "metrics" image of postgresql pod are
broken. This is because the volume is specified incorrectly. Thus, the
deployment ends up creating `emptyDir: {}` volume instead of `configMap`
one.
This commit fixes it.
Signed-off-by: Igor Zibarev <zibarev.i@gmail.com>
* postgresql: Remove myself from maintainers
I have not been responsive to requests in the past year. In order avoid frustrations with future PRs I withdraw myself from the list.
* bump chart version
* [postgresql] Set update strategy to Recreate.
We NEVER want two different postgres instances to be up at the same time using the same database files, or corruption can occur.
At the cost of having a few seconds of downtime at upgrade time.
* Update Chart.yaml
* Adding values for liveness & readiness probe thresholds
* Postgresql: Upping chart version
* Removing use of default as it is redundant
* Incrementing chart version as it seems someone beat me to it
* Update Chart.yaml
Add the ability to specify pod/deployment annotations through
`values.yml`.
I tested this works without any custom annotations, and also tested
specifying custom annotations and verifying they are included in the
existing pod/deployment.
* PostgreSQL metrics need the right username
If we use `POSTGERS_USER` = `sonar`, metrics go boom.
* Bump version
* Bumping PostgreSQL version
* Bumping PostgreSQL version
* PostgreSQL metrics need the right username
If we use `POSTGERS_USER` = `sonar`, metrics go boom.
* Bumping PostgreSQL version
* [stable/postgresql] standardize labels
* use "postgresql.name" as app label instead of "postgresql.fullname"
* match pods by two labels, release and app
* [stable/postgresql] bump version to 0.9.6
* [stable/postgresql] update notes
* [stable/postgresql] bump version to 0.11.0
* Fix for version comparison from strings to semver
See #3002 for more detail
* Include pre-releases in the semver ranges
This is important when testing against alpha and beta builds of
Kubernetes along with environments that use pre-releases to denote
things other than pre-releases (e.g., gke denotes the environment
with a pre-releases)
Add support for a fullnameOverride paramater to the PostgreSQL chart, matching
the pattern used by several other charts, as well as by the `helm create`
template chart.
* Increased probe timeout
When database is stoped brutally database can start with "FATAL: the database system is starting up".
Depending on the storage it may take more than 60 seconds to autofix and the POD will restart indefinitely.
* Increased version
* Update Chart.yaml
* Add persistence.mountPath value
This allows to mount a volume in /var/lib/postgresql/data instead of
/var/lib/postgresql/data/pgdata. This is mandatory if the user starting
the postgres image is not the owner of /var/lib/postgresql/data/pgdata
but has only write access to it.
With this patch, the entrypoint will create /var/lib/postgresql/data
with the user that launched the container as owner and initdb will be
happy.
* Add documentation
* bump chart version
* command to run postgres client should not use rs
Modify the printed command to connect to the postgresql server pod to use a `--restart=Never` so that the pod instantiated that hosts a `psql` client is not backed by a ReplicaSet.
* Increment patch version number.
* [stable/postgresql] Add image pull secrets to deployment template
* update deployment template
* add new parameter to README
* [stable/postgresql] bump chart version
update version to 0.8.5
* [stable/postgresql] add an example of imagePullSecrets to values.yaml
* [stable/postgresql] lower initialDelaySeconds of readiness probe
updated health probes to test postgres connection with POD_IP
* postgresql: adjust probe timeouts
* add persistence.existingClaim to postgresql
* update docs
* make template inline
* nope, that was not right
* up version
* clean up templating
* bump minor version
* Fixes#415
* Use volumeMount subPath feature to accomplish same goal. Removed obseleted init-container to remove lost-found directory.
* Bumped version to 1.0.0. Made persistence.subPath configurable. Added a note in README for users wanting to upgrade.
* Updated persistence readme wording