AutoDeploy - ZenPack Operation Errors

From Zenoss Wiki
This is the approved revision of this page; it is not the most recent. View the most recent revision.
Jump to: navigation, search

When using the core-autodeploy.sh script, some users have encountered a ZenPack Operation Error that haults the installation. This error is rare and very difficult to reproduce. Below are suggestions you can use to help resolve the issue.

Example Error

There was an error during the ZenPack operation: 
Command 'zenpack  --install /opt/zenoss/packs/ZenPacks.zenoss.ApacheMonitor-2.1.3-py2.7.egg' 
returned non-zero exit status 3
Error: A zenpack failed to install. Fix the zenpack or delete it from 
/opt/zenoss/var/zenpack_actions.txt, then restart zenoss.


Quick Fix

Many users simply create a new Virtual Machine/Instance and start over from scratch. Since the error is intermittent, chances are that the second installation will go well.


Rabbit

See Working with Queues for examples...

  • Restart the rabbit service to verify there aren't any errors
  • Verify the queues are present, recreate them if missing
  • Start the zenrrdcached daemon to see if there are any errors (zenhub.log and syslog or system)
  • Adjust $ZENHOME/etc/global.conf with the correct rabbit user credentials (if you rebuilt the queues, you will most likely need to use zenoss for username/password)
  • Go through the ZenOSS_Tuning#rabbitmq-server


Misc

  • Execute $ZENHOME/bin/zenoss_upgrade_pre as the root user (you will have to add $ZENHOME="/opt/zenoss" to it)
  • Execute $ZENHOME/bin/zenoss_upgrade_post as the root user


core-autodeploy.sh

  • Execute core-autodeploy.sh once more to see if the errors have cleared up


Delete the ZenPack

  • Delete the ZenPack the script is having issues with (Example "rm /opt/zenoss/packs/ZenPacks.zenoss.ApacheMonitor-2.1.3-py2.7.egg")
  • Delete the ZenPack entry from /opt/zenoss/var/zenpack_actions.txt
  • Once zenoss is up and running, install the ZenPack manually.


Feedback

Any feedback or tips are greatly appreciated, we want to eliminate this error so that you can spend more time working on transforms or ZenPacks!