05-12-2020 07:02 AM
Hi,
I am after some help on Enterprise Vault 10.0.2.112; I have inherited this from a colleague that has left & know it is no longer supported.
After running low on space in a partition I added a new disk to the VM server running Enterprise vault & followed the article below.
https://www.veritas.com/support/en_US/article.100017745
After the wizard completed I can now longer get into the "vault store group" in the MMC to see the partitions without it crashing & throwing me out. I can get to all other elements in the console but as soon as I drill into this area it crashes.
The drive that I pointed to for the new partition has the folders created by EV (as specified in the wizard).
Any help resolving this would be greatly appreciated as I am now unsure if a partition has been configured or not.
Thanks in advance
Andy
05-13-2020 06:36 AM
Hi Andy,
This is quite strange. Have you tried restarting the EV services?
05-13-2020 07:45 AM
Hi Virgil,
Thanks for the response. I have rebooted the SEV server a few times & also logged on as a few different users (with admin perms).
I can see that the new partition has the folder structure created by EV & 2 XML files (enterprisevaultpartitionroot.xml & partitionsecurednotificaton.xml). I can also see the new partition in the appropriate SQL table.
The rest of the admin console seems to function as normal. I can open the quick tasks "new partition" & that opens.
Andy
05-13-2020 07:46 AM
Andy,
This is a tough situation, because whatever is wrong will probably require correction in the SQL database, which we really don't like to see happen without Support being involved. But you're on EV 10, so Support can't be involved...
The first thing I would do to troubleshoot would be to start a dtrace on the "mmc" process, then reproduce the problem (drill into the Vault Store container and crash the console). That will at least give an idea of what is the last operation the console is attempting before it crashes. If you post the dtrace log here, I'll have a look and see if the fix is something obvious.
--Chris
05-13-2020 08:31 AM
Hi Chris,
I have ran the dtrace against MMC. Am I ok to post the whole output on here or are there certain parts you need?
sorry for my ignorance.
Andy
05-13-2020 08:33 AM
Whole thing is fine, unless you have specific security requirements to redact anything (server names, IP addresses, etc.).
--Chris
05-13-2020 09:04 AM
Dtrace log attached as it was too big. Thanks for taking the time to respond.
Andy
05-13-2020 10:03 AM
Hey, this seems like a known issue back in EV 10. You can either upgrade to EV 10.0.4 (where it was fixed) or go to the EnterpriseVaultDirectory database's PartitionEntry table and update the ScanStartDateTime column. Specifically, any entry where ScanStartDateTime is set to NULL should be changed to a valid date instead. For example:
USE EnterpriseVaultDirectory
GO
UPDATE PartitionEntry
SET ScanStartDateTime = '1900-01-01 00:00:00.000'
WHERE ScanStartDateTime IS NULL
Make sure you take a backup of the database prior to modifying it, in case something goes wrong.
--Chris
05-14-2020 04:44 AM
Chris, thank you so much for providing the solutions.
I am unable to get hold of the the EV 10.0.4 patch so the sql table update looks like the way forward.
Excuse my ignorance but from the below do i need to change the date entry to today date or is any date acceptable (sorry SQL is not a strong point)? & on that note can I just run the below as a query against the db after i have backed it up?
USE EnterpriseVaultDirectory
GO
UPDATE PartitionEntry
SET ScanStartDateTime = '1900-01-01 00:00:00.000'
WHERE ScanStartDateTime IS NULL
Much appreciation for you help on this matter.
Andy
05-14-2020 04:57 AM
Hello Andy,
Chris writes "should be changed to a valid date". 1900-01-01 is a valid date :)