Adding an optional deployment to increase AI pickup speed #688
Conversation
Bumps [helm/chart-testing-action](https://github.com/helm/chart-testing-action) from 2.6.1 to 2.7.0. - [Release notes](https://github.com/helm/chart-testing-action/releases) - [Commits](helm/chart-testing-action@v2.6.1...v2.7.0) --- updated-dependencies: - dependency-name: helm/chart-testing-action dependency-type: direct:production update-type: version-update:semver-minor ... Signed-off-by: dependabot[bot] <support@github.com> Signed-off-by: Daniel Hendricken <daniel.hendricken@bedag.ch>
Signed-off-by: Daniel Hendricken <daniel.hendricken@bedag.ch>
Signed-off-by: Daniel Hendricken <daniel.hendricken@bedag.ch>
Signed-off-by: Daniel Hendricken <daniel.hendricken@bedag.ch>
…rker deployment Signed-off-by: Daniel Hendricken <daniel.hendricken@bedag.ch>
Co-authored-by: Kate <26026535+provokateurin@users.noreply.github.com> Signed-off-by: Daniel Hendricken <162698556+dhendricken@users.noreply.github.com>
|
Alright no worries, I add these workers per kustomization myself at the moment. Works like a charm. I made the PR because I thought it might increase the ease of use for others that decide to use the assistant. Ah and I forgot to mention that I tested it with both the apache and fpm-alpine images. |
| app.kubernetes.io/name: {{ include "nextcloud.name" . }}-ai-worker | ||
| app.kubernetes.io/instance: {{ .Release.Name }} | ||
| app.kubernetes.io/component: app |
There was a problem hiding this comment.
You add a new component.
Please adjust the labels like in other deployments here:
e.g.:
| app.kubernetes.io/name: {{ include "nextcloud.name" . }}-ai-worker | |
| app.kubernetes.io/instance: {{ .Release.Name }} | |
| app.kubernetes.io/component: app | |
| app.kubernetes.io/name: {{ include "nextcloud.name" . }} | |
| app.kubernetes.io/instance: {{ .Release.Name }} | |
| app.kubernetes.io/component: ai-worker |
| volumeMounts: | ||
| {{- include "aiWorker.volumeMounts" . | trim | nindent 12 }} | ||
| volumes: | ||
| - name: nextcloud-main |
There was a problem hiding this comment.
I believe we should we move that into _helpers.tpl and reuse it with deployment of nextcloud.
| {{- toYaml .Values.aiWorker.podSecurityContext | nindent 8 }} | ||
| initContainers: | ||
| - name: wait-for-service | ||
| image: alpine/curl |
There was a problem hiding this comment.
Image should be in values.yaml too.
|
|
||
|
|
||
| {{/* | ||
| Create volume mounts for the nextcloud-aiworker container. |
There was a problem hiding this comment.
Why is it needed without the defaultConfigs and configs like in nextcloud.volumeMounts ?
I believe we should use the same setup then in the main deployment.
Description of the change
The aim of this PR is to add the process described in increasing AI pickup speed in a simple but functional way to the helm-chart.
It functions similarly to the cronjob sidecar by mounting the volumes under nextcloud-main.
Benefits
The deployment is completely optional as it only has relevance to those who want to use the nextcloud-assistant or other AI integrations. Additionally this method allows for scaling by increasing the number of replicas of the deployment. (This is why I decided against using a container withing the main pod).
It makes use of the same image and volumes as the main nextcloud pod, so I don't think it will require additional maintenance unless something about how AI Tasks are handled changes in the future. Maybe we won't even need this anymore who knows :)
Possible drawbacks
None that I can think of.
Applicable issues
I didn't find any open issues concerning this functionality.
Additional information
I created a modified version of the nextcloud.volumeMounts helper template to reduce the required volumes that need to be mounted to the pods as much as possible.
Checklist
Chart.yamlaccording to semver.