post deploy reference data job #81
Conversation
| name: {{ $.Chart.Name }}-update-all-reference-data | ||
| {{- include "seqr.labels" $ | nindent 4 }} | ||
| annotations: | ||
| "helm.sh/hook": post-install, post-upgrade |
There was a problem hiding this comment.
I do not think this should be done post-upgrade, just post-install. I think longer term I will add a better job for just updating the updated resources as a cron job/ pot update task
|
Just realized that we probably need a flag to disable this job for the unit tests... thinking about the right way to do it. |
charts/seqr/values.yaml
Outdated
| @@ -99,6 +99,12 @@ cronJobs: | |||
| - name: check-new-samples-job | |||
| schedule: "*/10 * * * *" | |||
| command: "python manage.py check_for_new_samples_from_pipeline" | |||
| - name: update-all-reference-data-job | |||
| schedule: "0 4 * * 0" # 4am on sunday | |||
| command: "python manage.py update_all_reference_data --use-cached-omim" | |||
There was a problem hiding this comment.
This is not a good idea, it deletes all of genocode and would functionally cause like a half hour of degraded performance and incorrect searches. We do not even run this as a cron job, we run a series of more targeted updates. I will do a post MVP audit of our cron jobs and adapt the ones that are useful for local installs but for MVP we should just maintain current behavior by running this on startup only
There was a problem hiding this comment.
ah ok! I had it written down that we should do this but I must have misunderstood.
There was a problem hiding this comment.
yeah sorry I was musing that we should do it in the future and I guess I was not being clear
No description provided.