We've found the problem with a friend of mine.
/etc/crontab has wrong syntax in defining crons -
I've used spaces instead of tab between the * * * * * and the username....
really my bad, maybe copied the string from the other machine it was
running on and pasted without cheching but...
I don't think it's normal an incorrect crontb syntax to crash the
crontab with coredump?
is it?
the other strange thing,
the line below the incorrect syntax there is a line with a correct one,
that wasn't working too,
and even if the cron don't core, the correct line wasn't working....
maybe the cron stops parsing if see incorrect line?
is this a suggested behaviour? maybe it's a bug, along with the crashing
with core when no tab used between the * * * * * and the user...
dunno,
pls someone that's familiar to tell...
10x! :-)
Anton - Valqk wrote:
Hi there,
have anyone faced a cron crashing in a jail when changing the
/etc/crontab?
I'm running a jail with nsswitch using pgsql backend and I've noticed
a strange behaviour
of my crontab.
Every time I change the /etc/crontab file I get (or make crontab -e
and save it)
/usr/sbin/cron[32864]: (*system*) RELOAD (/etc/crontab)
in /var/log/cron
and immediately after that the cron dies with a core (core file is in
/var/cron/cron.core),
I've made a ktrace but nothing strange can be seen in it,
heres the link to the kdump file:
http://valqk.ath.cx/ktrace.out.txt
any ideas appriciated!
cheers,
valqk.
--
This message has been scanned for viruses and
dangerous content by MailScanner, and is
believed to be clean.
_______________________________________________
[email protected] mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-stable
To unsubscribe, send any mail to "[EMAIL PROTECTED]"
