As the name suggests.
Description
Description
| Status | Subtype | Assigned | Task | ||
|---|---|---|---|---|---|
| Resolved | NicoleLBee | T324507 Develop a web app for editing Toolhub records | |||
| Resolved | NicoleLBee | T330155 [backend] Deploy on Toolforge | |||
| Resolved | NicoleLBee | T331784 [backend] Update files for Toolforge environment | |||
| Resolved | NicoleLBee | T331841 [backend] Get Celery and Redis running |
Event Timeline
Comment Actions
Celery is up and running on Toolforge.
If for some reason the kubernetes pod should go down or need to be restarted for some reason, here are the steps needed to get it back up again.
- kubectl get pods to see if a toolhunt-api.worker pod exists
- if it does, kill it with kubectl delete deploy toolhunt-api.worker
- spin up a new pod with kubectl create --validate=true -f $HOME/www/python/src/deployment.yaml
To check Celery logs (I'll try to add proper logging at some point):
- kubectl get pods to see the name of the pod (currently toolhunt-api.worker-ff8d56c79-5r62g)
- kubectl logs <name>
Comment Actions
Unless you are changing the contents of the deployment itself, kubectl delete pod <pod name> would be a faster way to recreate the toolhunt-api.worker pod. The Deployment maintains a replicaset which will notice that 0/1 desired pods are present and recreate the deleted pod. This can be turned into a 1-liner by using the 'name' label you setup in the Deployment to find the pod's name: kubectl delete $(kubectl get pod -l name=toolhunt-api.worker -o name)