Commit Graph
14 Commits
Author SHA1 Message Date
Sébastien Prud'homme 7be0b80cf0 [stable/prometheus-operator] Resolve scraping controller-manager and scheduler outside out the cluster (#10366)
* [stable/prometheus-operator] Resolve scraping controller-manager and scheduler outside out the cluster

Signed-off-by: Sébastien Prud'homme <sebastien.prudhomme@gmail.com>

* [stable/prometheus-operator] Bump version

Signed-off-by: Sébastien Prud'homme <sebastien.prudhomme@gmail.com>
2019-01-08 08:08:09 -08:00
Denis Iskandarov ef0300e8fd use proper k8s image registry (#10325)
as by https://github.com/kubernetes/kubernetes/pull/54174

Signed-off-by: Denis Iskandarov <d.iskandarov@gmail.com>
2019-01-02 03:30:27 -08:00
Vasily Sliouniaev 295b8a192f [stable/prometheus-operator] Fix typo selecor -> selector (#10053)
* Fix typo

Signed-off-by: Vasily <vasily.sliouniaev@jet.com>

* bump chart version to 1.0.0 since its a stable repo

Signed-off-by: Paul Czarkowski <username.taken@gmail.com>
2018-12-17 12:27:33 -08:00
Vasily Sliouniaev ffa0580f93 [stable/prometheus-operator] Update to v0.26.0, address setting empty selectors (#9647)
* Option to set empty servicemonitor, rules selector

Signed-off-by: Vasily <vasily.sliouniaev@jet.com>

* Update to 0.26 operator

Signed-off-by: Vasily <vasily.sliouniaev@jet.com>

* Remove spaces to fix lint

Signed-off-by: Vasily <vasily.sliouniaev@jet.com>
2018-12-15 00:40:48 -08:00
Harshal Shah eb4180bd15 [stable/prometheus-operator] Upgrading prometheus version to v2.5.0 (#9710)
* Upgrading operator, prometheus and alertmanager versions

- Upgrading prometheus-operator version to 0.26.0
- Upgrading prometheus version to 2.5.0
- Upgrading alertmanager version to 0.15.3
- Tested on GKE by installing prometheus operator with above versions.
- helm lint passed.

Signed-off-by: Harshal Shah <harshal2623@gmail.com>

* Removing operator version upgrade

Keeping prometheus and alertmanager upgrades intact.

Signed-off-by: Harshal Shah <harshal2623@gmail.com>

* Fixing the chart app version

Signed-off-by: Harshal Shah <harshal2623@gmail.com>
2018-12-08 15:03:18 -08:00
Rajat Vig e320d5188c Allow for custom labels and annotations for default rules (#9154)
Signed-off-by: Rajat Vig <rajat.vig@gmail.com>
2018-11-13 01:12:47 -08:00
Kierran McPherson 635d310027 [stable/prometheus-operator] Allow annotations on prometheus service (#8867)
* Allow annotations on prometheus service

Signed-off-by: Kierran McPherson <kierranm@gmail.com>

* Add annotations to alertmanager

Signed-off-by: Kierran McPherson <kierranm@gmail.com>

* service.type and service.nodePort for prometheus

Signed-off-by: Kierran McPherson <kierranm@gmail.com>

* Update Chart.yaml versions

Signed-off-by: Kierran McPherson <kierranm@gmail.com>

* Fix typo

Signed-off-by: Kierran McPherson <kierranm@gmail.com>

* Remove indentation

Signed-off-by: Kierran McPherson <kierranm@gmail.com>

* Remove extra service in values; Add annotations

Signed-off-by: Kierran McPherson <kierranm@gmail.com>
2018-11-11 12:42:06 -08:00
Rajat Vig 75b020f7e8 Allow for custom service annotations for prometheus and alertmanager (#8966)
* Allow for custom service annotations for prometheus and alertmanager

Signed-off-by: Rajat Vig <rvig@etsy.com>

* Add documentation for the values.yaml

Signed-off-by: Rajat Vig <rvig@etsy.com>

* Update the CI file

Signed-off-by: Rajat Vig <rvig@etsy.com>

* Bump the chart version

Signed-off-by: Rajat Vig <rajat.vig@gmail.com>
2018-11-11 10:04:27 -08:00
Vasily Sliouniaev 12b6bbbfb9 Upgrade to 0.25.0 (#8987)
This release (https://github.com/coreos/prometheus-operator/releases/tag/v0.25.0) introduces two new features
* [FEATURE] Allow passing additional alert relabel configs in Prometheus custom resource (coreos/prometheus-operator#2022)
* [FEATURE] Add ability to mount custom ConfigMaps into Alertmanager and Prometheus (coreos/prometheus-operator#2028)

- Skip deploying CRDs on upgrade - the resources already exist and helm
  is not able to modify them using the CRD hook. The operator manages the
  upgrade by itself and it's possible to use the extra features and
  perform the upgrade all in one step. In the future it may be necessary
  to use a job to perform the CRD upgrade at the crd-install point

Signed-off-by: Vasily Sliouniaev <vasily.sliouniaev@jet.com>
2018-11-04 06:09:40 -08:00
Dmitry Verkhoturov 42092b0db2 [stable/prometheus-operator] fix prometheus ruleSelector example (#8888)
* fix prometheus ruleSelector example (#8831)

Signed-off-by: Dmitry Verkhoturov <paskal.07@gmail.com>

* remove outdated comment, unify text in values.yaml and test-values.yaml

Signed-off-by: Dmitry Verkhoturov <paskal.07@gmail.com>

* remove outdated comment, unify text in values.yaml and test-values.yaml

Signed-off-by: Dmitry Verkhoturov <paskal.07@gmail.com>
2018-11-02 06:35:50 -07:00
Gustavo Michels b083549b79 Fix Prometheus and Alertmanager templates to support routePrefix (#8760)
* Fix Prometheus and Alertmanager templates to support routePrefix

Signed-off-by: Gustavo Michels <gustavo.michels@nanthealth.com>

Bump up chart version

Signed-off-by: Gustavo Michels <gustavo.michels@nanthealth.com>

Fix routePrefix reference for Prometheus ingress

Signed-off-by: Gustavo Michels <gustavo.michels@nanthealth.com>

Simplify logic to handle routePrefix in Prometheus and Alertmanager ServiceMonitors

Signed-off-by: Gustavo Michels <gustavo.michels@nanthealth.com>

* Simplify logic for Prometheus routePrefix in Grafana datasource ConfigMap. Bump up chart version.

Signed-off-by: Gustavo Michels <gustavo.michels@nanthealth.com>

* Fix ci/test-values.yaml to match values.yaml except for cleanupCustomResource

Signed-off-by: Gustavo Michels <gustavo.michels@nanthealth.com>
2018-11-02 05:14:37 -07:00
Joe Hohertz f77b473719 Fix default values for tolerations, nodeSelector and resources. (#8849)
In several places, the empty default for tolerations was an object
where it should have been an array, for nodeSelectors the opposite,
should be an object but was specified in an array, and for one
case of resources, the empty object was just missing.

Updated these all to what they should be, which has eliminated
errors like the following when running:

```
2018/10/27 00:46:18 warning: cannot overwrite table with non table for tolerations (map[])
2018/10/27 00:46:18 warning: destination for nodeSelector is a table. Ignoring non-table value []
2018/10/27 00:46:18 warning: destination for nodeSelector is a table. Ignoring non-table value []
2018/10/27 00:46:18 warning: destination for resources is a table. Ignoring non-table value <nil>
2018/10/27 00:46:18 warning: cannot overwrite table with non table for tolerations (map[])
2018/10/27 00:46:18 warning: destination for nodeSelector is a table. Ignoring non-table value []
```

Signed-off-by: Joe Hohertz <joe@viafoura.com>
2018-10-29 11:47:40 -07:00
Dmitry Verkhoturov f432fe893d fix typos in prometheus-operator (#8705)
* fix typos

Signed-off-by: Dmitry Verkhoturov <paskal.07@gmail.com>

* bump version

Signed-off-by: Dmitry Verkhoturov <paskal.07@gmail.com>

* fix typo

Signed-off-by: Dmitry Verkhoturov <paskal.07@gmail.com>
2018-10-24 09:57:40 -07:00
Giancarlo Rubio 7d0450231c Migrate prometheus operator from coreos (#6765)
* Create promethues-operator chart

Signed-off-by: Giancarlo Rubio <gianrubio@gmail.com>

* Fixes for stable/prometheus-operator

Changed:
- Update Prometheus to 2.4.3, alertmanager to 0.15.2
- Change some job labels - should refer to a label appearing on the service. Tried to keep this consistent with what was used in the coreos version
- Add confguration for using CoreDNS
- Add insecureSkipVerify to kubeApiServer
- Add `prometheus.additionalScrapeConfigs`, `prometheus.additionalAlertManagerConfigs`

- Add ability to deploy an operated prometheus without installing anything else
  - Switch `prometheus.serviceAccountName` to be based on `prometheus.fullname` to prevent clashes
- Fix `serviceMonitorSelector` typo
- Add `serviceMonitorNamespaceSelector` to values.yaml
- Fix grafana typo (https://github.com/helm/charts/pull/6765#issuecomment-428653560)

Add CoreDNS dashboard

- Fixed dashboard use of updated name metric

Signed-off-by: Giancarlo Rubio <gianrubio@gmail.com>
Signed-off-by: Vasily Sliouniaev <vsliouniaev@users.noreply.github.com>

Vas prometheus operator (#5)

* Fix lint

* Fix missing selector on kubelet

https://github.com/helm/charts/pull/6765#issuecomment-431035212

* Fix braces in files

* Fix naming

* Move non-exporters out of exporters folder

* Use common value names, add to readme

* Change value names deploy -> enabled

https://github.com/helm/charts/pull/6765#issuecomment-428283768
https://github.com/helm/charts/pull/6765#issuecomment-430070684

* Update etcd rules

https://github.com/etcd-io/etcd/blob/master/Documentation/op-guide/etcd3_alert.rules.yml

* Fix etcd rules to match job name

* Review readme

* Change initial version to 0.1.0

Additional Fixes While Github is down (#6)

* Fix additional service monitor template

* Fix label indentation

https://github.com/helm/charts/pull/6765#issuecomment-431788727

Fix cleanup, requirements.lock file (#7)
2018-10-23 09:34:30 -07:00