The task in SolusVM 2 goes to Queued state and hangs

Symptoms

• The task in SolusVM 2 goes to Queued state and hangs.
• After cancelling the stuck task in SolusVM 2 > Tasks it remains stuck inside the RabbitMQ container on the Management Server:
  

  # docker exec -ti $(docker ps --format '{{ .ID }}' -f name=solus_rabbitmq) sh
  # rabbitmqctl list_queues | grep <queued_task>
  ...
  vm-<queued_task>-1-v1 12

  Note: Replace <queued_task> with the actual value. It can be seen in SolusVM 2 > Tasks > Action.

Cause

SolusVM 2 bug with ID #SIO-3662 which has been fixed in version 1.1.20949.

Resolution

Update SolusVM 2 by following the article.

For already affected tasks use the following workaround:

Note: There should not be any running tasks in SolusVM 2.

1. Log in to the Management Server via SSH.
2. Backup SolusVM 2 Management node.

3. Enter the rabbitmq container on the Management Server:

   # docker exec -ti $(docker ps --format '{{ .ID }}' -f name=solus_rabbitmq) sh

4. Fetch the list of all queues:

   # rabbitmqctl list_queues | grep <queued_task>
   vm-<queued_task>-1-v1 12
   vm-<queued_task>-4-v1 0
   vm-<queued_task>-5-v1 0

   Note: the affected queue is vm-<queued_task>-1-v1 where the value is not zero.

5. Delete all messages from this queue like below:

   # rabbitmqctl delete_queue vm-<queued_task>-1-v1

   For example,
   

   # rabbitmqctl delete_queue vm-backup-create-1-v1

6. Log in to the Compute Resource via SSH where the VPS with the stuck backup task is hosted.

7. Restart the agent service:

   # systemctl restart solus-agent