On 5/29/07, Douglas Allan Tutty <[EMAIL PROTECTED]> wrote:
On Tue, May 29, 2007 at 09:42:30AM -0700, Jason Dunsmore wrote:
>
> Whenever updatedb runs via cron, it goes to sleep and never finishes
> the job, leaving all these processes running:
>
> 29519 ? S 0:00 /USR/SBIN/CRON
> 29520 ? Ss 0:00 /bin/sh -c test -x /usr/sbin/anacron || ( cd /
> && run
> 29521 ? S 0:00 /bin/sh -c test -x /usr/sbin/anacron || ( cd /
> && run
> 29522 ? S 0:00 run-parts --report /etc/cron.daily
> 29559 ? Ss 0:00 /bin/sh /etc/cron.daily/find
> 29561 ? SN 0:00 /bin/sh /usr/bin/updatedb
> 29569 ? SN 0:00 /bin/sh /usr/bin/updatedb
> 29572 ? SN 0:00 /usr/bin/sort -z -f
> 29573 ? SN 0:00 /usr/lib/locate/frcode -0
> 29577 ? SN 0:00 su nobody -s /bin/sh -c /usr/bin/find /
> -ignore_readd
> 29578 ? SN 0:01 /usr/bin/find / -ignore_readdir_race ( -fstype
> NFS -o
>
> What could be causing this? It happens on Etch for x86 and AMD64.
>
What happens if you run the command in /etc/cron.daily manually as root?
As always, I'll suggest doing it with X not runing and run it from the
command line. In this case, I'd suggest you run top in another vt and
watch things from there, especially idle% and wa% (waiting for io).
Hmm, when I run it manually, it finishes fine. I don't think it gets
hung up every time it runs via cron. I'm going to try using this
modification to /etc/cron.daily/find to log messages when it gets hung
up.
15c15
< cd / && nice -n ${NICE:-10} updatedb 2>/dev/null
---
cd / && nice -n ${NICE:-10} updatedb >> /var/log/updatedb.log
Thanks for the help.
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
