Veritas-bu

[Veritas-bu] NetBackup on Windows backups hanging.

2004-01-06 22:30:21
Subject: [Veritas-bu] NetBackup on Windows backups hanging.
From: Dan.Larson AT peters DOT com (Dan Larson)
Date: Tue, 6 Jan 2004 21:30:21 -0600
HBAs are new HP (2g), switch is a Cisco (2g).  The old server had the
old gen (1g) HP/Compaq HBAs and was connected thru a Brocade (1g)
switch.

2 HBAs, separate for tape and disk, tape zone only has the one HBA and
the bridge in it.

--- I don't know about errors on the switch ports, we will check
tomorrow---

The library has two bridges (1g) in it,  each has the 6 drives for a
single OS.  The UNIX backups never see the issue unless we need to power
cycle the library.

The really weird thing is we saw this same behavior with different
server/drivers/HBAs/switch/bridge/fiber/power/

-----Original Message-----
From: Mickey Baker [mailto:mb AT unixusa DOT com] 
Sent: Tuesday, January 06, 2004 17:19
To: Dan Larson; veritas-bu AT mailman.eng.auburn DOT edu
Subject: RE: [Veritas-bu] NetBackup on Windows backups hanging.

What HBA and switch type? 
Separate HBAs for disk and tape storage?

See any errors on the switch ports? The library is segmented for the two
environments?

How are the drives separated? They're in a separate zone from the UNIX
drives? Zoning by switch port?

I've seen scsi errors caused by SAN congestion and arbitration issues
cause similar issues.

Mickey Baker
Safe Data Services, LLC


-----Original Message-----
From: veritas-bu-admin AT mailman.eng.auburn DOT edu
[mailto:veritas-bu-admin AT mailman.eng.auburn DOT edu] On Behalf Of Dan Larson
Sent: Tuesday, January 06, 2004 5:29 PM
To: veritas-bu AT mailman.eng.auburn DOT edu
Subject: [Veritas-bu] NetBackup on Windows backups hanging.


We had 3 policies running, they all stop writing to the tape drives at
about the same time.  No errors that I can find, nothing in the server
event logs, Veritas is looking at the NetBackup logs but I don't know if
they will find anything there.  The only way we have found to get
everything talking again is to down the server, power cycle the library
then boot up the server.  Stopping and starting the services or even
rebooting the server doesn't seem to do it.  The other media servers
don't seem to have these issues, but they are not backing up anywhere
near the data that the main server is.

We started to have this issue with our original master/media server, but
since that hardware was coming off lease we didn't chase it too hard.
It mostly happens on the weekend when we are using all 6 drives for a
really extended time.  First we have trouble with one drive, then it
spreads until nothing is moving.

All new server hardware, Clean load of Windows 2000, SP4, windows
updates applied, new HBA, talking thru a new fiber Switch.  Qualstar
412600 library, we have half of it for Windows, the Unix guys have the
other 6 drives and 300 slots.

Clean install of NetBackup, moved the catalog databases over then
upgraded to FP3, then up to FP5.

It worked pretty well for a while, then the problems reappeared.

Replaced the fiber-scsi bridge in the library so at this point the
hardware and software is all different pretty much end to end and we
have chased the drivers as far as we have been able to determine.




_______________________________________________
Veritas-bu maillist  -  Veritas-bu AT mailman.eng.auburn DOT edu
http://mailman.eng.auburn.edu/mailman/listinfo/veritas-bu





<Prev in Thread] Current Thread [Next in Thread>