apache / apache/openwhisk-deploy-kube
Install-packages repeatedly got error status
- Dominant language
- Shell
- Stars
- 309
- Forks
- 231
- PR merge metrics
- No merged PRs in 30d
Description
I tried to deploy openwhisk on kind with 2 virtual worker nodes. After I used helm to install, other pods successfully completed or were running except the install-packages pods:
```
➜ ~ kubectl get po -n openwhisk
NAME READY STATUS RESTARTS AGE
owdev-alarmprovider-5b86cb64ff-kv8s2 1/1 Running 0 3h35m
owdev-apigateway-64fc695f65-9xbz9 1/1 Running 0 3h35m
owdev-controller-0 1/1 Running 0 3h35m
owdev-couchdb-74896bddf6-fxsxb 1/1 Running 0 3h35m
owdev-gen-certs-kk87r 0/1 Completed 0 3h35m
owdev-init-couchdb-knlf2 0/1 Completed 0 3h35m
owdev-install-packages-g6gdk 0/1 Error 0 178m
owdev-install-packages-j8m94 0/1 Error 0 3h35m
owdev-install-packages-m4gv9 0/1 Error 0 170m
owdev-install-packages-wt6zk 0/1 Error 0 3h
owdev-invoker-0 1/1 Running 0 3h35m
owdev-kafka-0 1/1 Running 0 3h35m
owdev-kafkaprovider-5574d4bf5f-wj4wm 1/1 Running 0 3h35m
owdev-nginx-58c75d69b4-bf7kz 1/1 Running 0 3h35m
owdev-redis-756ffc8fb4-72tc8 1/1 Running 0 3h35m
owdev-wskadmin 1/1 Running 0 3h35m
owdev-zookeeper-0 1/1 Running 0 3h35m
wskowdev-invoker-00-18-prewarm-nodejs10 1/1 Running 0 171m
wskowdev-invoker-00-8-prewarm-nodejs10 1/1 Running 0 3h2m
```
All error install-packages pods belong to the same job owdev-install-package, and after 4 pods getting stuck in error status there were no more install-packages pods created.
I tried to use the command `kubectl describe` to check these install-packages pods, but there were no events.Therefore then I used the `kubectl logs` to check these pods, and I got:
```
Cloning into 'openwhisk'...
fatal: unable to access 'https://github.com/apache/openwhisk/': gnutls_handshake() failed: The TLS connection was non-properly terminated.
```
How to solve this problem? Thx.
Contributor guide
Research direction
Start with the owdev-install-packages Job and its pod logs, using the reported kind cluster with two worker nodes and the Helm deployment. Inspect the failed GitHub clone and the surrounding Kubernetes configuration for the TLS handshake failure. Done means the install-packages Job completes successfully without repeatedly creating Error pods.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- docker, helm, kubernetes, shell
- Domain
- cloud, devops, infrastructure
- Issue type
- Bug
- Difficulty
- 4/5
- Estimated time
- 3-5 days
- Activity status
- Stale
- Clarity
- Needs clarification
- Newbie friendliness
- 20/100