Hey KV, the KEEPALIVE params on a unix/linux system are variables set in
the networker start scripts, but on windows these need to be done at the
registry level, below is an excerpt from an actual emc article..i hope
this helps.
Windows NetWorker server:
1. On the NetWorker server in the registry, set the KeepAliveTime
registry key to 3420000 ms (57 minutes).
Please refer toMicrosoft KB Article 140325 for further details regarding
how to create this registry key.
2. Re-boot the NetWorker server OS in order for the change to become
effective.
To use NSR_KEEPALIVE_WAIT (to prevent the nsrexecd <-> nsrexec
connection dropping)
On the NetWorker client:
1. Right mouse click on My Computer and select Properties
2. Click on the Environment Variables button.
3. Create a new SYSTEM variable
4. In the variable name field enter NSR_KEEPALIVE_WAIT
5. In the variable value field enter a value in seconds which is less
than the timeout value on the firewall for the TCP Start Timeout. (ie if
TCP Start Timeout = 60 seconds, set NSR_KEEPALIVE_WAIT value = 25 secs)
-----Original Message-----
From: EMC NetWorker discussion [mailto:NETWORKER AT LISTSERV.TEMPLE DOT EDU] On
Behalf Of kurianv
Sent: Monday, 6 April 2009 10:30 PM
To: NETWORKER AT LISTSERV.TEMPLE DOT EDU
Subject: [Networker] Backup of a storage node fails
Preston de Guise wrote:
> On 06/04/2009, at 15:11 , kurianv wrote:
>
>
> > Hi,
> > We have Networker 7.2 running on windows. We have a storage node
> > which is located sumwat away from our site where there are 3 servers
> > to backup. Usually the network connection between our site and the
> > storage node site is really slow or sometimes even breaks . The
> > problem what we notice is that when the backup is initiated from
> > backup server from our site and when the backup is working normally
> > for the storage node, when the network connection breaks or becomes
> > slow the backup fails marking the tape full in the storage node. I
> > dont know why this happens often. I know there is no problem with
> > the tape library.
> >
> > I dont know the exact logic behind this , because once the backup is
> > initiated from our backup server then why does the backup which is
> > taking place in the storage node fails.....Any ideas?
> >
> > The errors what i get is listed below :
> >
> > 1) Lost network connection to host
> >
> >
> > 2) RPC Network connection not available
> >
> > and as a result it becomes media verification failes....
> >
> > Your suggestions wud be valuable...
> >
>
> First, I'd suggest it's time you upgrade away from 7.2. Yes, the 7.2.x
> tree was a good tree, but it's too old now. It's been unsupported by
> EMC since mid 2008, meaning you're trusting your data protection to a
> now untrustable configuration. Do you really want to do that?
>
> Regardless of whether backups are running to a local storage node, you
> have meta-data always running between the clients and the server -
> control communications plus index related data. Additionally, the
> server maintains a heart beat (albeit a slow one) with the storage
> node nsrmmd processes.
>
> You could start by tweaking the following settings on the backup
> server resource itself:
>
> nsrmmd polling interval - number of minutes between checks done by the
> server to make sure nsrmmd is still running
> nsrmmd restart interval - number of minutes NetWorker waits between
> restart attempts of a failed nsrmmd
> nsrmmd control timeout - number of minutes NetWorker waits for storage
> node requests/updates
>
> If you have a slow link, double these as a starting point (from 3, 2
> and 5 respectively to 6, 4, and 10).
>
> However, if you have a link that outright fails, there's only so much
> that can be done - if it fails while NetWorker needs it to not fail,
> it's going to cause you a problem no matter what. I'd suggest though
> that by increasing those nsrmmd intervals, you may be able to at least
> somewhat reduce the frequency of media being marked as full. If the
> backup is becoming slow, then unless you're using the NetWare v4.x
> client, it's likely to be because the link is slow enough that even
> index (meta-data) comms is being affected. If that's the case, you'll
> need to either increase link performance and stability or look at re-
> architecting the configuration.
>
> Cheers,
>
> Preston.
>
>
> --
> Preston de Guise
>
>
> "Enterprise Systems Backup and Recovery: A Corporate Insurance
Policy":
>
>
http://www.amazon.com/Enterprise-Systems-Backup-Recovery-Corporate/dp/14
20076396
>
> http://www.enterprisesystemsbackup.com
>
> NetWorker blog: http://nsrd.wordpress.com
>
>
>
> via RSS at http://listserv.temple.edu/cgi-bin/wa?RSS&L=NETWORKER
Hi,
Preston thank you very much for your recommendations....I do consider
the options of upgrade to 7.4.X. Infact we have already purchased the
licenses and is planning. But what my question is that will this issue
be solved if i upgrade it to 7.4.3.....
My second question is that how do we look for the nsrmmd polling
interval and all other parmaeters which you mentioned in your post
through Networker. What i can see from the devices configurations are
these :
1) Device Polling Interval = <NIL>
2) Device min Load tries = <NIL>
...and etc
Could you tell me the procedure to set these parameters?
Craig,
Thanks also for you r recommendations , but please can you tell me how
to set the parmateers what you have told? Will it affect the other
backups which iam running locally...
Regards,
kv
+----------------------------------------------------------------------
|This was sent by kurianji AT gmail DOT com via Backup Central.
|Forward SPAM to abuse AT backupcentral DOT com.
+----------------------------------------------------------------------
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
No virus found in this incoming message.
Checked by AVG - www.avg.com
Version: 8.0.238 / Virus Database: 270.11.44/2044 - Release Date:
04/06/09 18:59:00
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
|
