Originally posted by: SystemAdmin

Seems like a rather broad question, but, here we go......

Yesterday, all of the users on a particular system where taking
very long times trying to ssh to the system.

I finally got in and ran topas and vmstat but saw no noticeable issues...

I started to think about what recent changes I had made and realized that
2 days prior, I had used SMIT to start NFS.

(Unfortunately, I did not have the time to start debugging since we were
in production)

I used SMIT again to stop NFS.

Magically, everyone was able to ssh normally.

Has anyone else seen this?

Here's some infor

oslevel: 5300-06-01-0000
model: RS/6000 Model = IBM,9110-51A
2 - 73GB disks mirrored
WebSphere 6.0.2.13 running

Thanks!

davezarn@hotmail.com
#AIX-Forum

Originally posted by: SystemAdmin

I suspect the following for your issue

1.) DNS - issue. As per my practical experience whenever you have the delay with telnet/rlogin/ssh. it may be related with name resolve if there is no load on destination server.

2.) if your(user) home directory is on NFS or if you have PATH variable on your profile which is pointing to NFS share.So this case if nfs hangs, it has to wait for timeout from nfs directories.
Hope it helps.
#AIX-Forum

Originally posted by: SystemAdmin

Thank you very much!

It indeed was a DNS issue....

Turns out, that this system was only configured with one DNS server.

Unknowing to me, the network engineer brought down that system and later
brought it back up.

The termination of NFS coincided with the DNS server being brought up.

I have corrected this situation by adding an additional DNS server

Thanks again!

DaveZ
#AIX-Forum