1) Try adding this to the 'Remote Access' property of
cluster.foobar.com:
root AT server1.foobar DOT com
root AT server2.foobar DOT com
You may also need to add the hosts to the 'Aliases' property. I had to
do this on an NDMP cluster - Oracle/Sun 7410 - that happens to be
active/passive but with separate names.
2) What does the device activity show on this AFTD when the errors
occur? Does it report that 4 sessions are writing already? You may
need to raise it. And how many filesystems does this client have?
Most of our AFTDs here have device parallelism set pretty high. Usually
between 12 and 30, depending on purpose (we have three SNs, though only
the server and one SN share the main 60 TB array - the other two are
client-specific).
Also, server parallel set to 4 seems a bit low; The base setup for one
Server was 16 I believe. Are you running Business Edition?
--TSK
=====
Tim Kimball http://sungak.net
=====
-----Original Message-----
From: Brian O'Neill [mailto:oneill AT OINC DOT NET]
Sent: Friday, March 18, 2011 9:25 AM
Subject: Issues after upgrade to 7.6.1
So I finally upgraded one of our 7.5.2 servers to 7.6.1. Everything
seems to be fine except for two things - one is an issue, one is an
oddity (at the moment).
1) I have a floating IP as part of a manual cluster setup between two
Linux boxes - call them server1.foobar.com and server2.foobar.com. Both
are backed up independently, but a set of directories are excluded by
.nsr directives.
The floating IP, called "cluster.foobar.com", explicitly backs up the
directories (they are replicated - only need one copy). In the client
config for cluster.foobar.com, "server1.foobar.com" and
"server2.foobar.com" are explicitly named for "Remote access".
This worked in 7.5.2, but now the backup fails, with:
save: File index error: permission denied, `root' on
`server1.foobar.com' must have remote access privilege to client
cluster.foobar.com.
I tried adding an explicit "root@" in front of the host names, but this
didn't work either.
I updated the client to 7.6.1 as well, didn't help.
2) This server has an AFTD volume, which is the Default pool. Despite
the volume being mounted and available, networker still says "Waiting
for 1 writable volume(s) to backup pool 'Default' disk(s) or tape(s)".
Backups still occur, but it looks like it is waiting longer than it
should in some cases. If I try and back up ONLY the problem client
above, it actually sits and waits after giving a bunch of errors,
apparently waiting on a writable volume.
I vaguely recall some change in parallelism in some update. I checked,
and server parallelism is set to 4, with the AFTD device set to have a
target sessions of 4 and max sessions of 512.
-Brian
To sign off this list, send email to listserv AT listserv.temple DOT edu and
type "signoff networker" in the body of the email. Please write to
networker-request AT listserv.temple DOT edu if you have any problems with this
list. You can access the archives at
http://listserv.temple.edu/archives/networker.html or
via RSS at http://listserv.temple.edu/cgi-bin/wa?RSS&L=NETWORKER
To sign off this list, send email to listserv AT listserv.temple DOT edu and
type "signoff networker" in the body of the email. Please write to
networker-request AT listserv.temple DOT edu if you have any problems with this
list. You can access the archives at
http://listserv.temple.edu/archives/networker.html or
via RSS at http://listserv.temple.edu/cgi-bin/wa?RSS&L=NETWORKER
|
