deploy.md

How to deploy and troubleshoot

Deployment process

1. Assuming we want to work on process Channel 202, we'll have deployment name as ch202-env2a

2. Prepare ch202-env2a-values.yaml from the example values.yaml

3. Prepare a Secret resource from aws-credential file (the file name must be exact). It should contain:

   [default]
   aws_access_key_id =
   aws_secret_access_key =
   

   Then run:

   kubectl create namespace ch202-env2a
   kubectl create secret generic aws-credential --from-file aws-credential -n ch202-env2a
   

   NOTE: please remember to delete the aws-credential after the Secret resource has been created.

4. (Optional, if not available) Add helm repo

   helm repo add infobloxopen https://raw.githubusercontent.com/infobloxopen/farsight.collector.colo/master/colo/charts
   

5. Run:

   helm repo update
   
   helm search infobloxopen/farsight-collector-colo
   
   helm install infobloxopen/farsight-collector-colo \
       --version <put-version-here> \
       --name ch202-env2a \
       --namespace ch202-env2a \
       -f ch202-env2a-values.yaml
   

Troubleshooting commands

Soft clean up

helm delete --purge colo-process

Hard clean up

After running this command, please refer to One-time setup to recover.

k3s-killall.sh && docker stop $(docker ps -aq) && docker rm $(docker ps -aq)

Ramdisk clean up

rm -rf /dev/shm/*

Restart k3s:

systemcl restart k3s

References

https://k3s.io/