Erik Fair wrote:
> Check to see that /sbin/init (process 1) is still alive.  If it's dying,
> that would explain the zombies.

init is still alive and in "Is" state.

Curt Sampson wrote:
>> 	foreach i ( machine1 machine2 machine3 )
>> 	? echo '<root password>' | rsh $i "sudo sh /etc/init.d/syslog start"
>> 	? end
>> 
>>    This has always worked fine for me in the past.  But now, it completes
>>    the first iteration, and then it stalls and I find the remote commands
>>    have turned into unreaped Zombies:
> 
> Just a thought, but have you tried rsh -n?

If I close stdin with "-n" then I can't pass the password into the "sudo" on
the remote end.  :-)

I'm really baffled.  I can do the above with, say, "sh ... syslog stop" and
it works.  I can do other NOP stuff (e.g., "sh /etc/init.d/uucp start") and
it works.  But doing a remote "syslog start" as above hangs.  And I know
it didn't happen under 1.2.1 because I've had to do this several times in
the past - for some reason, our AutoClients running Solaris 2.6 have
recurring problems with "syslogd" randomly dying silently (no core dump).

And I'm sure as heck not going to log out of CDE for awhile since I'm not
too fond of Watchdog Resets  :-)

	- Greg