From: Jeff Cleverley (jeff.cleverley@avagotech.com)
Date: Wed Mar 29 2006 - 03:06:40 EST
Greetings,
We have been seeing odd things with our network and nfs in the last
month or so. One of the main side effects is that some hpux
workstations (C3000, hpux 11.00) will mount a file system from a hpux
server (rp5470 hpux 11.11). In all cases, you can do a bdf . and also
string /etc/mnttab and see the file system is mounted. When you do an
ls, you get nfs server not responding messages. In other cases, you
will get a file system with 5 top level directories and the hang occurs
on 3 of them, but the other 2 are fine. You can immediately go to the
same nfs directory from another hpux workstation and do an immediate
ls. It works fine. Rebooting the workstation has no effect.
The number of file locks is low (default 200) on the server and will be
corrected in about 10 days when we take everything down. I'm not seeing
flock errors that indicate the server is out of locks. I've done some
looking around in /var/statmon/sm.bak on both server and workstation and
the systems do not show up there. They are in the /var/statmon/sm
directory. I've turned on debugging (kill -17) of statd, lockd, and
mountd on both ends, and ran some nettl commands. I don't really know
what the output is trying to tell me and if it's normal or not. Below
are a few lines from the output. The 141 address is the workstation,
and the 229 is the server.
14:08:12.898448 IP 130.29.211.141.1020 > 130.29.209.229.2049: [DF] udp
84 nfs call (ver 3) readdirplus
14:08:12.899239 IP 130.29.209.229.17449 > 130.29.211.141.22965: [DF] udp 23a
14:08:12.899361 IP 130.29.209.229.10 > 130.29.211.141.0: [DF] udp 18
14:08:12.899383 IP 130.29.209.229.17440 > 130.29.211.141.15738: [DF] udp
389c
14:08:15.398116 IP 130.29.211.141.1020 > 130.29.209.229.2049: [DF] udp
84 nfs call (ver 3) readdirplus
I've wondered if the servers might have some type of lock being cached
for the workstation, but can't find anything. These are critical
servers, so I cannot readily reboot them and I'm a little concerned
about stopping and restarting all the rpc processes on the server.
Any help on this would be appreciated.
Thanks,
Jeff
--
---> Please post QUESTIONS and SUMMARIES only!! <---
To subscribe/unsubscribe to this list, contact majordomo@dutchworks.nl
Name: hpux-admin@dutchworks.nl Owner: owner-hpux-admin@dutchworks.nl
Archives: ftp.dutchworks.nl:/pub/digests/hpux-admin (FTP, browse only)
http://www.dutchworks.nl/htbin/hpsysadmin (Web, browse & search)
This archive was generated by hypermail 2.1.7 : Sat Apr 12 2008 - 11:02:52 EDT