Hi,
Thanks for your help. Here's a summary. Micky's answer
is probably the most accurate since we start several copies of
rpc.netisamd on startup. I guess if the 1st copy died, the
sunisamd (prog 100065) probably haven't cleaned up yet when
the next was started.
My thanks to...
Micky Liu (micky@ejv.com)
Benjamin Cline (benji@hnt.com)
Shane Bouslough (hane.bouslough@peri.com)
>Does anybody have any idea what happened when this messages appears?
>
>
>rpc.netisamd[287]: couldn't register prog 100065
>
>Then the rpc.netisamd did not start. Once rebooted, the message
>did not appear and rpc.netisamd didn't have anymore problems.
>
From: micky@ejv.com (Micky Liu)
>>rpc.netisamd is a daemon process that manages access to Sun's ISAM
>>subsystem which may be used either by Sun apps or some 3rd party apps
>>that you have installed on your systems. Whenever rpc servers exit
>>for any reason they are supposed to 'clean up' after themselves and
>>un-register themselves with the portmapper process. If the rpc server
>>process exits without cleaning up, the portmapper is left with a
>>state that is incorrect.
>>
>>In your case, your rpc.netisamd process died and the portmapper still
>>thought that an active one was registered. So the next time that the
>>rpc.netisamd process was initiated, it failed to register -- since the
>>portmapper thought one was already running. See rpcinfo(8C) to see
>>how you can interrogate the portmapper -- and also see the '-d' option
>>to see how you can intentionally un-register a rpc program.
From: shane.bouslough@peri.com (Shane Bouslough)
>>It was probably already running. If it happens again, try
>>the rpcinfo command and see if it's registered.
From: Benjamin Cline <benji@hnt.com>
>>I would guess that rpcbind died off (or was having serious problems).
>>rpcbind handles the mapping of RPC port numbers to IP port numbers, >>and all rpc based programs need to communicate with it. Take a look
>>at the man page for rpcbind(1M) for more information.
-- Eric Chiang Sun Systems Support CSA Malaysia
This archive was generated by hypermail 2.1.2 : Fri Sep 28 2001 - 23:11:09 CDT