NFS serives running via udp / tcp part 2

[Aslam Ghumra_2]

Hi, if you could interpret that would be great.

the first two files are from the master server. cappsm (147.188.128.153 / 192.168.1.1)
The files are calles cappsm.TRC000 and cappsm.TRC001.

the thrid file is from a successful mount on the server capps1 ( 147.188.128.158 / 192.168.1.2 ) The file is called capps1.TRC000

All the files are encapsulated within a winzip file.

Cheer, once again

[Dave Olker]

Hi Aslam,

I looked at both traces. In the working trace I see all the packets I expected to see. In the failing trace I don't see any MOUNT request at all.

It looks like either the trace was started after the filesystem was mounted, or the trace was left running so long that the MOUNT stuff was dropped from the kernel trace buffers.

So the failing trace is not usable. I'd need to see a trace that actually captures the failed MOUNT request that results in the filesystem being mounted with UDP.

Regards,

Dave

[Aslam Ghumra_2]

Hi

What I did was the following :

unmounted the filesystem /home/den
started the trace ( as per above posts )

# mount -o proto=tcp /home/den
nfs mount: getaddr_nfs: nas9000_private: NFS service not responding(retry delay=5s)
nfs mount: retry: retrying(1) for: /home/den after 5 seconds
nfs mount: retry: giving up on: /home/den
stop the trace

nas9000_private IP address is 192.168.1.80

Am I doing this wrong ?

[Aslam Ghumra_2]

Hi All

We had a power failure which resulted in having to re-start all the affected systems. This is the master and 5 nodes and three nas servers.

By starting the NAS servers first then the master node and then the nodes themselves , seem to have 'clicked' everything together.

So we now have NFS using the TCP protocol.

Cheers, All