NetBackup v6 Administration Console performance issues
Hi all,
We are running NetBackup v6.0 MP7, and over a relatively long period of time (months to a year..) the performance of the Administration Console has deteriorated quite significantly. By this I mean it is very slow to launch the console and slow to navigate within the console (browse between media and devices etc, and different views).
The console works fine, it's just that the speed has deteriorated so much it now takes over 3 minutes from launching the application until the console is up and ready to use.
Is there anything that can be done to improve things?
Platform:
Win Server 2003
Dual Xeon 3.66GHz
3GB RAM
NetBackup v6.0 MP7
Ideally I'd like to rebuild the server and start from scratch, but we would need to build one alongside the existing one and then do a swap, but lack of available similar spec hardware rules this out.
Thanks in advance.
Chris
Comments
Try this, in the %install path%\Veritas\Java\ there should be a file called setconf.bat
Change the vaule of SET FORCE_IPADDR_LOOKUP from 1 to 0. so the new entry should look as follows
SET FORCE_IPADDR_LOOKUP=0
After this, restart the admin console
Let me know.
Cheers!
Manoj
------------------
Time isn't running out, but life is...
Hi Manoj,
Thanks for your response, but we don't appear to have this directory nor this file 'setconf.bat'. I've done a search of the local drive for this file in case it appears in another directory, but found nothing.
I'm not sure if this would affect the presence of the file, but I believe our installation is an upgrade from NetBackup v5.
Thanks,
Chris
Did you apply the Java Admin console patches?
You can get them from the symantec site.
Cheers!
Manoj
------------------
Time isn't running out, but life is...
Excuse my ignorance, but since we access the admin console through remote desktop on Windows, is the Java admin console required?
Since you are running straight of the Netbackup Master server you do not require Java admin console.
I guess, we do not have a easy way to analyze what is causing the delay without enabling verbose logging.
Cheers!
Manoj
------------------
Time isn't running out, but life is...
Chris,
Not sure if you are aware, but there are in fact four different admin consoles:
- Windows native "NetBackup Administration Console" - which you appear to be running
- Windows Java "NetBackup - Java Version 6.x"
- Native Unix Java console, which is typically run on a Unix host and displayed back over X11 to an XServer on your desktop PC or workstattion, typically started with a "jnbSA" command.
- And "bpadm", which is a character cell based menu interface for VT100 type sessions.
Anyway...
In the Windows admin console, if you select the activity monitor, then right-click "View" on the tool bar and select "Options...", then you'll have a dialog where you can select a similar config that Manoj has suggested - i.e. you could try selecting/ticking the "Disable ping connection checking" option.
Also, I'm wondering whether you have tens of thousands of jobs in the activity monitor?
Are any of your master and/or medias clustered? Do all NICs in all master server nodes have entries in your DNS server for both normal and reverse lookup? I ask this because I have seen very slow console behaviour when some master server cluster NIC aliases were not in DNS - but this was for a VCS Solaris 10 clustered master server - which may or may not relevant/similar to your problem.
Also, try looking in the "<install path>\VERITAS\NetBackup\logs\admin" folder and checking through the latest log for anything that looks odd.
One last thing, have you decommissioned or renamed any media servers since the initial build of your environment?
Within the Windows Admin console, on the master server you define which systems are allowed to run it. If you have systems defined there which no longer exist, you will experience the type of slowdown you describe.
NetBackup Management - Host properties - Master Servers - "master server" - properties - Servers - additional servers that can access this host
I had this issue and removed several systems no longer in use, viola!
NBU 7.0.1 on Solaris 10
writing to EMC 4206 VTL
duplicating to LTO2 in SL8500
(Soon to be LTO5)
using ACSLS 7.3.1
Make sure there are no hosts configured in the NBU environment that does no longer exist. In fact, make sure there are no hosts that are not resolvable by hosts file / DNS at all in the environment.
Would you like to reply?
Login or Register to post your comment.