Video Screencast Help
Search Video Help Close Back
to help
Not able to make it to Vision this year? Get a sampling in the Best of Vision on Demand group.

Storage unit selection criteria within a group

Updated: 22 May 2010 | 2 comments
Marianne van den Berg's picture
0 0 Votes
Login to vote

We have a Storage Unit group containing 2 STU's with 'Load Balance' as selection criteria. Since the configuration of the 2 media servers are identical, the assumption is that jobs will be equally distrbuted between the two. What we see in practise, is that the 1st STU in the list is assigned twice the amount of jobs in comparison with the 2nd one (about 800 jobs a night to the 1st one and +- 400 jobs to the 2nd one).
Our other choice is maybe Round Robin, but the manual says the following about this selection:
If a storage unit is not available, or if no unit has available space, the job fails and is not queued. NetBackup does not continue to select the same storage unit.

Any idea how we get jobs to be evenly distributed between the 2 STU's?

Comments

dustin_yonak's picture
19
Aug
2009
1 Vote +1
Login to vote

Hello Marianne, This is not

Hello Marianne,

This is not an issue I've come across before but the following TechNote may help:
http://seer.entsupport.symantec.com/docs/273511.htm

The key to this configuration is to understand that each storage unit needs to have the Maximum concurrent drives used for backups parameter configured for fewer tape drives then what is configured for the media server in Media and Device Management. If this step is missed and the setting is configured for all the tape drives in the library, the first media server in the storage unit group will perform the majority of the backups.

Dustin Yonak
Supportability & Release Readiness Team - NBU Global Technical Services
www.symantec.com

Symantec Enterprise Technical Support - 800-342-0652
Or open a case online at https://mysupport.symantec.com/

Marianne van den Berg's picture
19
Aug
2009
1 Vote +1
Login to vote

STU's configured as per TechNote

Thanks Dustin

Pure co-incidence - the configuration is identical to the example described in the TechNote:
6 drives shared between the master and 2 media servers (all drives confgured on all sytems in device management). 2 media server stu's configured with max 2 drives each and grouped into stug (with load balance configured). Mpx values same on both stu's. Policies that generate most jobs are going to stug. Unfortunately a 'balanced load' is not seen in practise...

Supporting Storage Foundation and VCS on Unix and Windows as well as NetBackup on Unix and Windows.
Handy NBU links