Closed pbdger closed 7 months ago
Hey @pbdger, the CRD for the EventTailer seems to contain image specs: https://github.com/kube-logging/logging-operator/blob/e87e3e2df4f422be21f00360de6d45e9ffb29cfa/charts/logging-operator/crds/logging-extensions.banzaicloud.io_eventtailers.yaml#L314
Could you share your yaml config used?
We set the values for Helm chart via Flux. The code looks like
---
apiVersion: helm.toolkit.fluxcd.io/v2beta1
kind: HelmRelease
metadata:
name: logging-operator
spec:
chart:
spec:
sourceRef:
kind: HelmRepository
name: myhelmrepository
namespace: mynamespace
chart: logging-operator
version: "4.5.1"
interval: 10m
values:
logging:
enabled: true
fluentbit:
image:
repository: myrepo/fluent/fluent-bit
eventTailer:
image:
repository: my/kube-logging/eventrouter
tag: "v0.1.0"
pullPolicy: "Always"
workloadMetaOverrides:
labels:
"gnp.phoenix.local/k8s-api-access": "read-events"
The outcome of the generated CRD contains for EventTailerSpec the workloadMetaOverrides specs but no data for image specs.
The CRD looks like
...
spec:
controlNamespace: logging
workloadMetaOverrides:
labels:
gnp.phoenix.local/k8s-api-access: read-events
So the image specs are not created via logging operator the the event trailer CRD.
In the example shown above a fluentd image spec is defined, too. This is created for CRD logging-operator as expected.
at the moment you can use
containerOverrides:
image: "my/kube-logging/eventrouter:v0.1.0"
pullPolicy: "Always"
but I understand the helm chart is missing the image field, so the issue is valid
Thx @pepov for the workaround.
I've found a fix in the pipeline (https://github.com/kube-logging/logging-operator/pull/1576). I hope it's coming soon.
@pbdger It would need some testing (I don't have the time for it), but let me know if I can help (for example with a dev release if needed)
@pepov If you create a dev release, I will be happy to test it.
@pbdger
helm fetch oci://ghcr.io/kube-logging/helm-charts/logging-operator --version 4.6.0-branch.event-tailer-chart-up.0
https://github.com/kube-logging/logging-operator/actions/runs/7716262509
Great. I'll start the test on tomorrow.
This issue has been automatically marked as stale because it has not had recent activity. It will be closed if no further activity occurs. Thank you for your contributions!
it was solved with https://github.com/kube-logging/logging-operator/pull/1674 but additional changes like the same for hostTailer are in: https://github.com/kube-logging/logging-operator/pull/1576
In the latest documentation of the eventtailer (https://kube-logging.dev/docs/configuration/crds/extensions/eventtailer_types/) you find the specs to define image specs.
Expected behaviour:
The CRD of the eventtailer should contain the image specs, e.g.
A concise description of what you expected to happen.
Steps to reproduce the bug:
Add the spec as shown above in the values.yaml of the Helm chart.
Environment details:
Bug: The CRD contains no image spec information.