EV server Running out of Space - Help.
Updated: 21 May 2010 | 5 comments
Currently we are running EV 7.5 sp1 on a IBM x series server with over 600 gb of drive space allocated on the D: partition. The C Partition is only 40 GB running the OS only (Win2003).
The D: parttion has all the stores, indexes, EV program files stored on it and currently only has 120 GB left out of the 660GB total space. In estimation it only has 7-8 months left before space melt down.
We are going to set up a IBM n series San and attach it to the Server using ISCSI.
1. Has any one yet virtualised a EV server yet and has there been any problems.
2. Is it better now to setup the SAN and leave all the original stores and indexes etc and create a new drive link ing to the SAN and creating a new store so in theory the only think growing on the original server would be the indexes. ?
3. Or would it be better to disconnect the D: drive after stopping all the services then renaming it E: then copy all the data to the san storage to a drive called D: and then re-starting the services.
4. Just moving all the stores onto the SAN to a drive called E: , but this will mean messing around with the database renaming links etc.
We are planning on backing up the SAN using IBM TSM (Tivoli).
Any help and advice would be appreciated .
Matthew.
Discussion Filed Under:
Comments
Hope this helps.
http://seer.entsupport.symantec.com/docs/273271.htm
How to move Enterprise Vault (EV) indexes that are managed by the same EV server
http://seer.entsupport.symantec.com/docs/273141.htm
I would advise speaking to our Professional Services branch. If you have a Technical Account Manager get in touch with them. They can look at your system, do some tests for analysis of throughput, growth etc and implement/assist.
Thanks,
Mark
Supportability Analysis Engineer for Enterprise Vault products.
**REMEMBER TO VOTE IF THIS HELPS AND MARK IT AS RESOLVED TOO IF IT IS!!**
I moved 2TB+ of data from an EMC SAN to a NetApp SAN due to space constraints. I moved the vault stores, indexes, message queues and temporary file locations. The overview process:
1) Copied all the closed partitions while EV was online
2) Shutdown all EV services
4) Copied the open partitions and indexes
5) Modifed a handful of SQL tables to point to the new drives and folders
6) Restarted the EV services
I can dig up the documentation that I used to change the SQL tables if it would help. The whole process was actually pretty easy but the copies using Robocopy took days due to all the small files. I would definitely suggest doing the open partitions and indexes over a weekend since EV will be offline.
Indexes should be put on your fastest disk. Most likely that will be the SAN, so I'd suggest going that route. Here's a good performance guide from Symantec that should help: http://seer.entsupport.symantec.com/docs/288395.htm
Andy
It's easy enough to close your existing partitions on D: and open new ones on E: (or whatever, you can write directly to CIFS shares, too). With 7.5 you can move the indexes, or add new Index Locations to slow the growth then move them later. None of that requires any sql. That will buy you plenty of time to move the remnants from D: when you need a new server.
good luck
Would you like to reply?
Login or Register to post your comment.