Oracle Installation

[Ash_4]

Hi,
I have mounted Oracle CD using,

1. nohup /usr/sbin/pfs_mountd &
2. nohup /usr/sbin/pfsd &
3. /usr/sbin/pfs_mount -t rrip -x unix /dev/dsk/c4t2d0 /SD_CDROM

But while running "orainst" it hangs and gives following mesg.

NFS server f1fa not responding still trying

Any idea to come out.

[Alan Casey]

for pfs_mount you need to have NFS services started.

You can do this through SAM

[Alan Casey]

To manually start NFS:

nfs must be running on the system for PFS to work, even if the
system is standalone. To check to see if nfs is running,
use the command:

rpcinfo -p

and look for nfs and mountd in the last column.

If you do not find them, then make sure the following variables are
set in the /etc/rc.config.d/nfsconf file:

NFS_CLIENT=1
NFS_SERVER=1
NUM_NFSD=4
START_MOUNTD=1

If you modified nfsconf, issue the following two commands to
restart the nfs server:

/sbin/init.d/nfs.server start
/sbin/init.d/nfs.client start

[Pete Randall]

Go back and check Stefan's comments in your previous thread:

http://forums.itrc.hp.com/cm/QuestionAnswer/1,,0x278193e260b0d611abdb0090277a778c,00.html

If you apply the Rockridge Extension patches, then you don't have to mess with pfs_mount - which, judging from the hundreds of questions posted about it, is a real pain to use. With the patches applied, you just use a regular mount/umount - much simpler.

Pete

[Ash_4]

Hi Alan Casey,
As you have suggested, nfs is already running.

[James Beamish-White]

Hiya,

I assume the name of your server is f1fa?

You need to make sure NFS is running - make sure rpcbind is running (ps -ef | grep rpcbind) and if it isn't, start it using /sbin/init.d/ scripts. Then make sure NFS itself is running by changing the /etc/rc.config.d/nfsconf to make everything the various configs to =1 from =0, then start NFS using /sbin/init.d/nfs* scripts.

Hope this helps.

James

[Ash_4]

hi everybody,
Hello James ,
I have checked as u suggested n found
rpcbind is not running.

how can i start it.

[James Beamish-White]

Just run /usr/sbin/rpcbind

If you have done hardening as suggested in the bastion host document, this file may have been renamed and you'll need to rename it back.

Cheers!
James

[James Beamish-White]

... you may need to restart your NFS services after you've got rpcbind running. Not sure there though.

Cheers!
James

PS: You may want to start awarding points too, you have some helpful answers above which you have not thanked with point awards.

[Ash_4]

hi james,
/usr/sbin/rcpbind doesnot exists.

[James Beamish-White]

?!?!?!? What version of HP are you running?

James

[James Beamish-White]

..and can you do "ls -ld /usr/sbin/*rpc*" and post the output please?

Cheers,
James

[Ash_4]

Hp-UX 10.20

[Ash_4]

-r-xr-xr-x 1 bin bin 12288 Jun 10 1996 grpck
-r-xr-xr-x 1 bin bin 407435 May 27 1998 pfs_mountd.rpc
-r-xr-xr-x 1 bin bin 366282 May 27 1998 pfsd.rpc
-r-xr-xr-x 1 bin bin 24576 Jun 10 1996 rarpc
-r-xr--r-- 1 bin bin 106496 Oct 7 1998 rpc.lockd
-r-xr--r-- 1 bin bin 36864 Oct 7 1998 rpc.mountd
-r-xr--r-- 1 bin bin 53248 Oct 7 1998 rpc.pcnfsd
-r-xr--r-- 1 bin bin 86016 Apr 14 1998 rpc.rexd
-r-xr-xr-x 1 bin bin 20480 Apr 14 1998 rpc.rquotad
-r-xr--r-- 1 bin bin 49152 Oct 7 1998 rpc.statd

[James Beamish-White]

Ah! Sorry, assumptions of being on 11...

rpcbind does not seem to exist on 10.20. Can you try to run "rpcinfo -p" and see if gets a valid response?

Cheers!
James

[Ash_4]

# out put of rpcinfo -p
program vers proto port
100000 2 tcp 111 portmapper
100000 2 udp 111 portmapper
100024 1 udp 604 status
100024 1 tcp 606 status
100021 1 tcp 610 nlockmgr
100021 1 udp 1031 nlockmgr
100021 3 tcp 614 nlockmgr
100021 3 udp 1032 nlockmgr
100020 1 udp 1033 llockmgr
100020 1 tcp 619 llockmgr
100021 2 tcp 622 nlockmgr
100068 2 udp 1037 cmsd
100068 3 udp 1037 cmsd
100068 4 udp 1037 cmsd
100068 5 udp 1037 cmsd
100083 1 tcp 1036 ttdbserver
536870913 1 udp 750
536870913 1 tcp 752
100005 1 udp 784 mountd
100005 1 tcp 786 mountd
100003 2 udp 2049 nfs
1342177279 3 tcp 1124
1342177279 1 tcp 1124

[James Beamish-White]

This shows that RPC is running and communicating. It may be you need a patch - I suggest taking a look at PHNE_22288 and seeing if it, or it's latest superceding patch fixes your problem.

It seems like your configuration is fine for RPC on 10.20, so I think I've come to the end of my knowledge on this. Anyone else got any ideas?

Cheers!
James

[Ash_4]

Thanks Alan,
As you said above info. is provided by HP May I know Where can i found this doc. on net.

[Alan Casey]

Hi Ash,

I found it on the Oracle website at:
http://metalink.oracle.com/

Doc ID: Note:1012147.7

You need a login for this site.


Alan