SAP backup Groups configured through Split mirror Option are getting into hang state after upgrading Networker server from Solaris10 to Solaris11.Index & bootstrap are not getting updated due to which the Group remains in running mode.

Hi,

Backups confgured through other Storage Node(Old one) work fine but from new Networker server it doesn't work.

After successful backup it doesn't write index & bootstrap.

If anything else works fine, I suspect a problem with the appropriate device (not available) or an unmounted volume.

Keep in mind that NW does not install device drivers, just jukebox drivers.

"inquire" is the command to verify the known device names.

However, in this case, you should see a "Media Waiting" alert as long as an appropriate media is not mounted.

If the volume is on a disk, unmount and remount it carefully - do NOT accidentally re-label it.

What I said ...

Remember that INDEX backups can be directed to a device at a remote storage node.

BOOTSTRAPS however must be stored as a local device at the NW server.

- what kind of upgrade (from what to what)?

- what error/pending messages?

Hi,

As I’ve mentioned that the Index & bootstrap is not getting written or updated after each SAP database backups which are of off host/split mirror backup.

It was working earlier but didn’t work after Master server(Networker server) hardware was replaced with new Operating System was changed to Solaris-11 from Solaris-10.

NOTE- Index for other similar backups are getting updated from Storage Node -1 which is unchanged.

Thanks

It's a hardware & OS upgrade of Master/Networker server.

OS Old  One- Solaris-10

OS New - Solaris-11

Similar backups from an existing SN is OK.

Hi

As I've mentioned Index & Bootstrap doesn't work for offhost backup configured for SAP database backup.

Group goes to hang state in Networker level seen through /nsr/applogs ( it doesn't end)

Normal client backup is ok.

The index update is beeing created while the backup runs (except for NMDP). The info is extracted from the save stream.

Index and bootstrap backups are server related issues. Exclusively. They do not have anything to to with the client data backup at all. That's why it is obvious that it is not SAP related. I am pretty sure that if you configure a 'standard' backup of the NW server client (see below) it will also not work.

However, 'does not work' and 'worked before' are not really helpful without any further feedback and error messages. (Remote) control without any feedback can just make assumptions. Worst case, look at /nsr/logs/daemon.raw

Let me suggest that you

  - create a new group

  - add the NW server's client name only

  - specify the save set to a minimum size (/etc) as mentioned.

  - run "savegrp -l full -G " or "savegrp -O -G " and check the output.

  - While running the group, check the NW server for alerts (use the admin GUI or nsrwatch).

In case 2- We also installed Solaris-11

From: Devdutta Mohanty (WI01 - GIS - ENU)

Sent: Sunday, May 3, 2015 6:55 PM

To: 'jive-735985249-6ujy-2-iwz6@emc-ecn.hosted.jivesoftware.com'

Subject: RE: - SAP backup Groups configured through Split mirror Option are getting into hang state after upgrading Networker server from Solaris10 to Solaris11.Index & bootstrap are not getting updated due to which the Group remains i...

Hi,

We did below things –

Case-1- Carried out OS Kernel patch update in Solaris-10(Old Networker server). No changes to NSR installation.

Case-2- Replaced the Old Networker server/hardware as it’s out of hardware support. Here we copied old “res,mm & Index”.

In both scenario problem is same.

Hi,

We did below things –

Case-1- Carried out OS Kernel patch update in Solaris-10(Old Networker server). No changes to NSR installation.

Case-2- Replaced the Old Networker server/hardware as it’s out of hardware support. Here we copied old “res,mm & Index”.

In both scenario problem is same.

Was this upgrade done in a way that you kept same name?  What about NSR?  Did you copy over resources or you did total new install?  I used PS before and I know of one occassion I had reinstall all components to get things back on proxy (split mirror on DMX at that time) after I changed content of nsrladb.  I suspect in your case something similar might have happened so it might be crucial to know what exactly you did with various nsr databases.

And I assume you use 8.1.x at minimum, right?  Anyway, can you:

a) stop NW server

b) do mv /nsr/res/nsrladb /nsr/res/nsrladb.orig

c) rm -rf /nsr/res/jobsdb

d) start NW server

If above doesn't help, run index&bootstrap backup (savegrp -O) on one of the groups and check where is breaks via debug log collection, but it would be also good to show us also logs (and even daemon.raw rendered at its basic level).

My problem was not related to backup server, but rather remote storage node.  I removed nsrladb at the end and also removed nsrauth entries for storage node (as there would be new ones created once I rename nsrladb).

?Hi,

Pls. find below logs(nsr/applogs) where the backup is not getting completed.