Short Summary :
Suggestion was to use
/usr/sbin/shutdown -g0 -y -i6
instead of
init 6
Question :
Machine : Ultra - 1
OS : Solaris 2.5.1
#crontab -l
0 1 * * 0 sync;sync;init 6 > /dev/null 2>&1
At 1:00 AM the machine reboots on all sundays, but on monday when i accessed
the
machine though I could
telnet to the machine many services where not available.
There was a network problem during the weekend and this server has some
directories automounted
df -k output hung after showing the local file systems
I had to "reboot" the machine, i tried a init 6 , but it didn't work.
The following is a edited output which i captured when i first noticed the
problem.
# who -r
. run-level 6 May 23 01:00 6 0 3
#
# ps -ef
UID PID PPID C STIME TTY TIME CMD
root 0 0 0 May 16 ? 0:00 sched
root 1 0 0 May 16 ? 0:25 /etc/init -
root 103 1 0 May 16 ? 1:07 /usr/sbin/rpcbind
root 2690 2678 0 May 23 ? 0:00 /sbin/sh /sbin/umountall -F
nfs
root 2474 1 0 May 23 ? 0:00 /sbin/sh /sbin/rc6
root 2678 2474 0 May 23 ? 0:00 /sbin/sh
/etc/rc0.d/K75nfs.client
stop
# w
7:46am up 8 day(s), 6:44, 4 users, load average: 0.03, 0.02, 0.02
# ps -ef|grep K
root 2678 2474 0 May 23 ? 0:00 /sbin/sh
/etc/rc0.d/K75nfs.client
stop
root 3202 3188 0 07:47:41 pts/2 0:00 grep K
#
# ps -ef|wc -l
424
why is the machine at run level 6 and still running ?
what happens when there is a network problem and at the same time a init 6
is
tried ?
would like to hear from you.
Regards
Aaj
Summary :
Suggestion was to use
/usr/sbin/shutdown -g0 -y -i6
instead of
init 6
This should run the appropriate init scripts to terminate network services
so that they can restart cleanly on the reboot. The "init 6" command is not
guaranteed to do this.
Thanks to
Ronald Loftin
Juan Manuel Del Rio C.
Anthony David
Bruce Zimmer
Bruce Zimmer wrote :
It is waiting for the K75nfs.client script to finish. This will happen even
if you try to do an init 6 from the command line. When most people have
this happen after they have initiated an init 6 and it hangs will do a
stop-A , sync then boot the system. Unfortunatly you started it from a cron
job so you were un-aware that it had hung until you came in Monday. I have
seen this behavior before when there are network problem and unfortunatly I
don't have enough information about the network problem you experianced to
assist you in determining what might have caused it.
This archive was generated by hypermail 2.1.2 : Fri Sep 28 2001 - 23:13:20 CDT