Old Fragmentation Warnings/Event 41561

[TJensen]

I'm getting old fragmentation warnings every day.

"Index fragmentation was detected after scan at '2/11/2018 1:00 PM'. Disregard this warning if you have defragmented the listed indexes since."

I've already addressed the fragmentation issue, but every day it alerts based on the detection from 2/11. How do I clear this?

[GertjanA]

Hello T.

Look at this: https://vox.veritas.com/t5/Enterprise-Vault/EV9-Status-Severe-alert-old-date/m-p/654555

That's an old one from me. You should be able to find the row in the table, although it is not for the eventnumber mentioned. You might have to do some additional searching first.

[TJensen]

Following your direction I looked at EnterpriseVaultDirectory\Monitoring. I couldn't fuss out from the other table which MonitoringItemType I was looking for, so I looked at the threshhold column. My warning indicated a 30% threshold and there was only one row that had 30 as the threshhold value.

I turned the enabled bit from 1 to 0 then re-ran all the status checks. The fragmentation warning didn't re-appear. I changed the bit to 1 and re-ran the checks, the fragmentation warning from 2/11 came back.

How do I silence the warning from 2/11 without silencing the warning completely?

[GertjanA]

Hello T.

You'll have to delete the row.

Delete from MonitoringHistory
where MonitoringItemEntryId = 'ID_from_monitoring_table'

Be advised that the 'where' clause has to have a unique row identifier. I'm not sure monitoringitementryid is unique, but as you are able to determine which row it is, you can find it yourself I assume.

[Marcde · Accepted solution]

You could run the computefragmentation stored procedure which will update the relevant entries based on the fragmentationstatus on the time of execution. Last time I verified this was under 11.0.1 but might work as well on EV 12. The best is to execute as the last part of the maintenance or shortly after the maintenance.

Running the 'ComputeFragmentation' stored procedure

https://www.veritas.com/support/en_US/article.000023310

Regards

Marc

[TJensen]

What seems to have resolved this was manually running an index re-org (its in my nightly maint plan anyways) and then running the stored procedure to compute fragmentation.