Hound gracefully shutting down at startup

[MohamedTalhaoui]

I have deployed Hound in my K8S cluster to index my gitlab instance.
Container is running all all index are built successfully, however hound is immediately shutting down whith no reason.

2023/06/26 13:32:02 All indexes built!
2023/06/26 13:32:02 running server at http://localhost:6080
2023/06/26 13:32:02 Graceful shutdown requested..

These are all the logs I have.

[userbradley]

I have the same issue as well. I've not figured out what the issue is yet

[MohamedTalhaoui]

Hi,
it is fixed on my side.
I have deployed Hound on K8S and I had to extend the startup/liveness probes to give some time at startup.

[userbradley]

That would make sense.

If you don't mind me asking @MohamedTalhaoui - How are you dealing with SSH keys?

We've mounted some private keys in to the /root/.ssh/id_rsa but then we have issues with known_hosts being write only.

Interested to see your approach to this

[MohamedTalhaoui]

Relevant part of my deployment yaml file below:

...
      containers:
          env:
          - name: GIT_SSH_COMMAND
            value: "ssh -o StrictHostKeyChecking=no -o UserKnownHostsFile=/dev/null -i /etc/ssh-key/id_rsa "
...
          volumeMounts:
            - name: ssh-key
              mountPath: /etc/ssh-key
              readOnly: true
...
      volumes:
      - name: ssh-key
        secret:
          secretName: hound-ssh-key     
          defaultMode: 0600
...

Assuming existing k8s secret hound-ssh-key holding the ssh private key.
Hope that helps