Dear sun managers,
My original posting was:
I recently upgraded our SPARC 10 server from solaris2.2. to solaris2.3.
Ever since we have had several problems.
1) rpc.nisd keeps dying. I also noticed that two rpc.nisd get started at
startup. This prevents users from logging in and I have to restart it.
2) nfs services hang. This prevents other machines from mounting/automounting
from the server
3) Continuous error messages on the console about address already
in use.(I also noticed that it tries to start most of the services two times)
Please see /var/adm/messages file attached. I have called SUN and they have not resolved it yet.
If anyone has experienced the same problem could you let me
know how you fixed it? When I did the upgrade, I did not do it from scratch.
but upgraded over the existing solaris2.2
/var/adm/messages starts
----------------------------------------------------------------------------
May 19 14:25:25 moses unix: MFS Release 1.1.3A (Thu Jul 1 13:53:52 EDT 1993)
May 19 14:25:25 moses unix: Copyright (c) 1990-1993, Atria Software, Inc.
May 19 14:25:03 moses rpcbind: udp: address in use
May 19 14:25:03 moses rpcbind: tcp: address in use
May 19 14:25:03 moses rpcbind: ticlts: address in use
May 19 14:25:03 moses rpcbind: ticotsord: address in use
May 19 14:25:03 moses rpcbind: ticots: address in use
May 19 14:25:03 moses rpcbind: could not find loopback transports
May 19 14:25:03 moses syslog: svc_tli_create: could not bind to requested address: No Error
May 19 14:25:03 moses last message repeated 2 times
May 19 14:25:04 moses nis_cachemgr: WARNING: nis_cachemgr already running
May 19 14:25:04 moses kerbd[155]: svc_tli_create: could not bind to requested address: No Error
May 19 14:25:18 moses inetd[171]: ftp/tcp: bind: Address already in use
May 19 14:25:18 moses inetd[171]: telnet/tcp: bind: Address already in use
May 19 14:25:18 moses inetd[171]: shell/tcp: bind: Address already in useMay 19 14:25:18 moses inetd[171]: login/tcp: bind: Address already in use
May 19 14:25:18 moses inetd[171]: exec/tcp: bind: Address already in use
May 19 14:25:19 moses inetd[171]: uucp/tcp: bind: Address already in use
May 19 14:25:19 moses inetd[171]: finger/tcp: bind: Address already in use
May 19 14:25:19 moses inetd[171]: time/tcp: bind: Address already in use
May 19 14:25:19 moses inetd[171]: echo/tcp: bind: Address already in use
May 19 14:25:19 moses inetd[171]: discard/tcp: bind: Address already in
-----------------------------------------------------------------------
There were several answers:
Look in your /etc/rc2.d/* for extra copies of internet startup
scripts.
I found an extracopy of inetsvc and rpc in my rc2.d. Removing the
extra copies took care of that problem.
Many people asked me to install various patches. But the ones that
I installed were the one list strongly recommended by SUN.
They are
1) 101329-15 (NIS+ PATCH)
2) 101344-05 (NFS JUMBO PATCH)
3) 101318-45 (Kernel JUMBO PATCH)
4) 101316-01
5) 101317-06
6) 101327-05
7) 101331-03
8) 101347-01
It looks like all these 8 patches are mandatory. Most of my
problems have now disappeared.
The following people answered to my question. I really appreciate
all your help. All your comments were useful to resolve the problems.
casper@fwi.uva.nl[Casper Dik]
reggie@rentec.com[Reggie Dugard]
worsham@aer.com[Robert D Worsham]
john@oncology.uthscsa.edu[John Justin Hough]
djm@bioimage.millipore.com[Drew Montag]
Thanks again
-Ramesh
This archive was generated by hypermail 2.1.2 : Fri Sep 28 2001 - 23:09:02 CDT