same 5.1B?

[ericfjchen]

We have many Alpha servers. This was upgraded to 5.1B by HP CE. However, when we check as below, the output are different? Why? Is the same OS version? What's different?
---------
node05
------
# uname -a
OSF1 node05 P5.1 1017 alpha
# sizer -v
Compaq Tru64 UNIX P5.1B (Rev. 1017); Wed Jun 16 04:41:52 EDT 2004
-----
node01
-----
# uname -a
OSF1 node01 V5.1 2650 alpha
# sizer -v
Compaq Tru64 UNIX V5.1B (Rev. 2650); Sun Dec 5 04:19:00 CST 2004

[Hein van den Heuvel]

The "P" kernel is a 'generic' one (genvmunix).

According to Han (in an HP internal document: note 12381.1 in Digital_unix) this usually this happens when you haven't finished a rolling upgrade.

hth,
Hein.

[ericfjchen]

Sorry, I don't fully understand. Can you pls explain more detail? Where can I down the white paper about this?

Thanks so much

Eric

[Hein van den Heuvel · Accepted solution]

Well, I have not experienced this myself either. So it is a bit of the blind leading the blind. It seems an upgrade on the P5.1 node did not complete succesfully. You may want to retry the upgrade completely. It might be that you are 'simply' booting the wrong kernel.
Carefully look at: ls -l /*vmunix*
Maybe you need a different P00>>> boot -fi xxx

If it was my box, then I would try: doconfig -c node05
(read up on 'man doconfig', do not forget to move the new kernel to /vmunix)

[ Han may get online from Europe in an hour or two and provide better help]

Hein.

[Han Pilmeyer]

Yes Hein. Just woke up. :-)

Node05 appears to be running from a generic kernel. Can you check whether /vmunix and /genvmunix are identical?

If the nodes are part of a cluster, then you could be in the middle of a rolling upgrade. You can do a "clu_upgrade -v status" on the cluster to find out.

[Johan Brusche]

E,

On the node with P5.1, run doconfig -c NODENAME, then "cp /vmunix /vmunix.preupgr"
and "cp /sys/NODENAME/vmunix /vmunix", then
immediately reboot.

Johan.