Debugging

Edit online

Ansible logs can be found in the /var/SevOne/ansible-upgrade/v7.0.<x>-v7.1.0/<timestamp>/127.0.0.1.log on the Cluster Leader.
If there are failures, resolve them on the relevant peers, and re-run the upgrade to start playbook execution again and the upgrade will start over.
- Failures on individual peers get recorded in the retry_files folder (in the parent ansible directory with the IP address of all the failed hosts)
- There are tags for each step of the upgrade, so in case there is a failure and you would like to kick off the upgrade for tasks remaining after a certain section, you can do that.
- For example, if you know ALL the peers have successfully completed the package upgrade phase, and the upgrade failed in the database section; you can kickoff the upgrade again like shown below. When upgrading from 7.0.<x> to 7.1.0,
  startingNMSversion = 7.0.<x>
  
  upgradeNMSversion = 7.1.0
  
  The command will run the playbook for all tasks except the packages tag aka the RPM upgrade section.
```
ansible-playbook -i hosts.ini main.yml \
--extra-vars "startingVersion={startingNMSversion} forwardVersion={upgradeNMSversion}" \
--tags="database, systempatches, cleanup" -v;
```
- If you know a certain subsection of peers have finished the upgrade completely, you can also remove them from the hosts inventory before restarting the upgrade script. This will result in kicking off all the tasks for all the hosts except the ones you removed from the inventory.