Kaseya Community

Oscar Alerts

  • I was wondering if anybody else has been receiving Oscar Alerts during nightly maintenance after installing 9.5.0.19 patch?  

    "The Oscar procedure (dbo.dbCleanupScriptLogAgentOrphans) reached it's batch duration limit with additional records remaining to be removed."

  • I got that message last night too. Wasn't sure what it meant or what was going on. Especially since it didn't appear to have our other Kaseya admins on the distro.

  • Upon further investigation it looks like that message was sent to all Master users on our VSA. Anyone know what we're supposed to do about it?



    forgot a word. added it :)
    [edited by: jchindemi@computex.net at 8:34 AM (GMT -7) on Mar 12, 2019]
  • From Kaseya Support :

    "The oscar procedure is a maintenance procedure that runs every night, it basically clears data from tables. It is normal to receive this notification if the procedure couldn't clear out all the records that same night, and may take a few days to clear the records. If there are more records than the batch size the procedure uses to delete these records, it will not fully delete the data and send the notification.

    If you continue to receive these messages for many days then that may indicate a problem, and we can take a closer look and monitor what is going on."

    Our concern is that we have never received this error until post 9.5.0.19 and now we get them nightly...

  • if you keep getting these alerts, which mean there is an issue with nightly DB maintenance job and it need attention. sometime you might have to manually execute stored procedure and fix this

  •  - That would be an indication that something is not working correctly and if it keeps occurring I would think that it should be addressed rather than manual intervention each time.

    FYI - 1. This maintenance process only runs Monday-Friday.

            2. The proposed fix is to run a Database Cleanup  (will see tomorrow if that was successful)

  • My org installed the 9.5.0.18 update before it was pulled in February. We started getting the Oscar Alerts and I opened a ticket with K-Support.  My Oscar Alerts were for:

    dbo.dbCleanupScriptThenElseOrphans
    dbo.dbCleanupScriptLogSubSteps
    dbo.dbCleanupEventInstanceHistEventOrphans

    We first checked the backup settings, and found it to be fine. After checking if I had Master Admin role, he sent me to this (easter-egg?) url:

    https:// {VSA-Server} /vsaPres/support/Tools/SQL.aspx

    We entered a SQL Query and clicked execute.

    DECLARE    @return_value int
    EXEC    @return_value = [dbo].[dbCleanupScriptStatusOrphans]
    @batchSize = 5000
    SELECT    'Return Value' = @return_value

    After you run the query, you'll get a Return Value if your value is "0" you're done. Otherwise click execute over again until it drops to 0. 

    I guess after that  navigate around the "Kaseya Support Tools" portal on your VSA for funzies.  ¯\_(ツ)_/¯ You also might want to consult KSupport being that your situation may be different.

    I thought it was pretty funny when the support tech said the Oscar Alert was a feature. LoL  He meant that it's a new alert that is triggered by the dbCleanup sql job during the scheduled backup time. You just didn't get an alert before.

  • Kaseya support has advice to run separate stored procedure.. no more oscar alerts on past two days ;)

  • Support eventually changed the time that was allowed to run the cleanup which resolved the issue. After the first week the answer was "To resolve this issue please go to the agent module > manage agents and click on clean database."

    Then escalation clarified:

    "Why wouldn't the process run a clean DB script nightly at midnight so that this alarm would not happen?

    This is happening, roughly at 1 am every morning - the reason you get the alert is the process is not able to completely remove all records in the associated amount of time the system has defined for the clean up to run.

    For some back ground we have built the logic into our DB clean up scripts to dynamically allocate a time to run ( abc script is allowed to run for X amount of seconds) , the amount of records to remove with each batch ( on this iteration abc script is allowed to modify X amount of records), and the amount of times it runs ( abc script will run x amount of times, until Z amount of seconds have passed). When any of the above is not able to remove all of the orphaned records an informational message is sent to the VSA administrators stating as such.

    How can we adjust what addresses are being sent to with this alarm?

    at this time we do not have a way to adjust this other then de-escalating privileges of the users that are receiving the alert. If you would like I would be more then happy to submit a feature request on your behalf in regards to this."

    but the issue did not stop....

    "If the process can be fixed by running a DB Cleanup: ...then "Why wouldn't the process run a clean DB script nightly at midnight so that this alarm would not happen?"

    this is not the current design of the product:

    the DBcleanupscripts (oscar) v.s the exposed Database clean up function do different things however there is an overlap in what they aim to accomplish. i.e cleaning orphaned agents records"

    They did not state what the fix was but responded with:

    "I have applied the adjustment to (dbo.dbCleanupAuditRsltAppsAgentOrphans)

    it should now run for 25 seconds per iteration for an overall duration of 20 minutes.

    it use to run for 10 seconds for an overall duration of 6 minutes. The expectation is that this should be long enough to clear out any orphans generated by deleting accounts."

    which still failed:

    "we would like the raise the increment to 30 seconds and the overall duration to 30 minutes"

    and then there was happiness, rainbows and unicorns as the issue has not returned.