Today I just discovered that my Journaling and Archiving tasks for Exchange have not been running for several days. Is there anything I can point monitoring tools at to watch to see if these tasks fail and/or stop running?
Currently we have PRTG and Log Rhythm at our disposal for solving this problem potentially
You may use evservice command...it dont check for status but you can use your monitoring tool to start the task and if said something like task is already started then you know it is running....
evservice start | stop |pause | resume <computer name> <task> [<task>...]
There's also the Get-EVTaskState cmdlet. This is how our SCOM management pack checks task states, but there's no reason you can't run it outside of the SCOM context in your own PS script. You just need to import the Symantec.EnterpriseVault.PowerShell.Monitoring module into your PS session.
Import-Module 'C:\Program Files (x86)\Enterprise Vault\Symantec.EnterpriseVault.PowerShell.Monitoring.dll' Get-EVTaskState -TaskEntryId 1A79706974090414FB9E9D1A09C6562521012000evserver1.EV.LOCAL
(You can find a list of the TaskEntryIds for your tasks in the Task table of the EnterpriseVaultDirectory database.)
I have also attached a little script that pulls the TaskEntryIds from the database and loops through them, getting the state for each one. It returns psobjects with the properties TaskEntryId, TaskName, and State, which you can format as you like. Example:
Hope that helps.
Thats a great script @ChrisLangevin
I believe the limitation with this script is it only returns the task state for tasks which are on the local server and not from remote EV servers.
Working on your script I have modified it to query SQL to find all the EV servers in the environment, then use Invoke-Command to run Get-EVTaskState remotely and write the output to a consolidated single text file which can then be emailed as txt or in an HTML report.
The result now is that you can view the status for all tasks across all servers in the Enterprise Vault environment by running the script only on one EV server.
No further modifications are required in the script and have optimised it as much as I could.
As far as I remember it is also possible to fetch Eventlog entries with PRTG.
You can then react to failures.
What I like to monitor also is working backups, and service states (running & backup mode). I send a message to a mailbox when a creteria is failed and automatically create a ticket out of it.