Randy

To: <NETWORKER AT LISTSERV.TEMPLE DOT EDU>
Sent: Thursday, September 25, 2008 1:01 PM
Subject: [Networker] Linux to Solaris Networker Server Migration

Hello,

I am working on a linux to solaris 10 migration for Legato networker. Essentially what I am wanting to do is move the indexes, media database, and res directory over to the solaris server along with the gst database.

Source server:
Debian Linux
IBM Server with 400 Gigs of Internal drives
4 AX150 EMC chassis with 18 TB of useable space
1 Scalar 100
Networker 7.3.1


New Sun server:

Sun coolthreads 5220 server with 16 gigs of memory and 650 Gigs of usable internal hd space (will be used for indexes) 1 AX4 with 60 TB of space (not usable, will be a bit less once I get the zfs pools setup)

1 Scalar 50 with 2 LTO-4 Drives

The DR solution will be the same as this Sun setup, except it will not have the AX4

Right now I am testing this migration on a Solaris 10 and Solaris 9 box without much luck.

Steps taken thus far:
1. Moved bootstrap from production Linux server to Solaris servers
2. Moved indexes over to Solaris servers (currently about 360 Gigs)
3. I moved over the gst database

I installed Networker Client, server, storage node, and NMC on the Solaris servers (version 7.3.1 sp4) I could not find the 7.3.1 version on EMC's website, so the difference in software is Linux is running 7.3.1 and Solaris is running 7.3.1 SP4.

I can't get the gst database to pick up, it refuses to let me login. So I use the default database that is provided, and login with administrator/administrator. Once I am in the java network management console, I can see the groups under monitoring. Under configuration, I can see the groups, and clients, and can browse quite a bit. Everything is good except for under local hosts, I get an error stating: "unknown host". The devices tab shows all the devices from the old server, I am not really worried about this tab because the new server will require a re-config anyways. Under the media tab I can see label templates and media pools. Under index, I can see all the hosts, but when I double click one, I get the following error: no savesets were found for xyz.host.com.

Under volumes, I see nothing.


Running an nsrck -L 6 looks ok:
nsrck: checking index for 'hostname.companyname.com'

nsrck: /mountpoint/nsr/index/hostname.companyname.com contains 862511 records occupying 87 MB

Running an nsrck -L 7 bombs:
nsrck: checking index for 'hostname.companyname.com'

nsrck: The file index for client 'hostname.companyname.com' will be recovered.

nsrck: The index recovery for 'hostname.companyname.com' failed.

nsrck: File index error: can't find index backups for 'hostname.companyname.com' on server 'sunserver.companyname.com'

If needed, I can paste the truss output for a nsrck -L 7, but the output is very very long.

I know this is not supported by EMC, but I have seen a lot of folks going from Solaris to linux, so I am assuming that Linux to Solaris has to be possible. Any input/help would be greatly appreciated.

+----------------------------------------------------------------------
|This was sent by tmarshall AT internap 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

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