Open
Description
Scenario: Successfully created ICP cluster using 3.1.1 release version of the terraform module. Actual ICP version is 3.1.2 (but not material to this issue). Attempting to scale number of workers by adding vms (tested on IBM Cloud).
Issue: Increasing the number of workers triggers the icp-worker-scaler function with the correct set of new worker IPs. Scripts fail with messages like:
module.icpprovision.null_resource.icp-worker-scaler: Provisioning with 'remote-exec'...
module.icpprovision.null_resource.icp-worker-scaler (remote-exec): Connecting to remote host via SSH...
module.icpprovision.null_resource.icp-worker-scaler (remote-exec): Host: ###.###.###.###
module.icpprovision.null_resource.icp-worker-scaler (remote-exec): User: root
module.icpprovision.null_resource.icp-worker-scaler (remote-exec): Password: false
module.icpprovision.null_resource.icp-worker-scaler (remote-exec): Private key: true
module.icpprovision.null_resource.icp-worker-scaler (remote-exec): SSH Agent: true
module.icpprovision.null_resource.icp-worker-scaler (remote-exec): Checking Host Key: false
module.icpprovision.null_resource.icp-worker-scaler (remote-exec): Connected!
module.icpprovision.null_resource.icp-worker-scaler (remote-exec): cat: /workerlist.txt: No such file or directory
module.icpprovision.null_resource.icp-worker-scaler (remote-exec): Couldn't find any entries in old list of workers. Exiting'
Cause:
fails to reference the correct old worker list. Refactoring of parameters for scripts has left out invokingscaleworkers.sh with the common set defined in icp-cluster function. A fix consistent with recent changes would be to update main.tf with the correct parameters.Metadata
Metadata
Assignees
Labels
No labels