Open wrenix opened 5 months ago
PS: Current the pipeline should failed, till #521 is merged.
But maybe you like to run it, to see the errors :kissing:
I believe we should merge that yet already for other and me ;) here related discussion:
Failing works now: https://github.com/nextcloud/helm/actions/runs/7774014966
==> Linting charts/nextcloud
Error: templates/metrics/deployment.yaml: unable to parse YAML: error converting YAML to JSON: yaml: line 49: did not find expected '-' indicator
Error: failed linting charts: failed processing charts
Error: 1 chart(s) linted, 1 chart(s) failed
------------------------------------------------------------------------------------------------------------------------
✖︎ nextcloud => (version: "4.5.17", path: "charts/nextcloud") > failed waiting for process: exit status 1
------------------------------------------------------------------------------------------------------------------------
failed linting charts: failed processing charts
Error: Process completed with exit code 1.
i rebase it to #521 / main
first look at: https://github.com/nextcloud/helm/pull/517#issuecomment-1923611887
i found option to set indent of sequence here: https://github.com/helm/chart-testing/blob/afa25e92904181aad0674f9483780bf7a76b100c/etc/lintconf.yaml#L33
but on test i only has effect on the Chart.yaml
-> maybe lets create an issue for later (and just merge that).
Still broken?
Oh, thispipeline failed, because it try to install it on an k8s without an prometheus-operator CRDs are preinstalled: https://github.com/nextcloud/helm/actions/runs/7775024318/job/21200687969?pr=523
two options (in my head):
if (.Capabilities.APIVersions.Has "monitoring.coreos.com/v1/ServiceMonitor")
(where where lint/test does not check the ServiceMonitor)Preinstalling sounds like it will test more cases, so let's go with that.
No, that is expacted behaviour. (it should failed, that hrlm try to install something, what the current k8s-cluster does not support) (PS: I thing you know it - i just answer for history/ other reader).
okay lets check out, how it is possible to tune the github actions (i need some more time therefore - i am working more with gitlab-ci).
i split it into a smaller PR #526 - so that the first part could already be merged till i solve an ServiceMonitor / Prometheus-Operator CRDs
Pull Request
Description of the change
Here some value files for chart-testing. it enables "all" features of the helmchart to find bugs like in #522
Benefits
it enables "all" features of the helm-chart during the chart-testing to find bugs like in #522
Possible drawbacks
Longer CI runtime
Additional information
TODO / Checklist (maybe moved to other PRs)
{{- ... | nindent }}
instatt of{{ ... | indent }}
Checklist
Chart.yaml
according to semver.