Cron silently dies. Why?

Cron silently dies. Why?

Post by Youri N. Podchos » Sat, 29 Apr 1995 04:00:00

Did anybody encounter the following problem before:

Occasionally, at random times cron daemon bombs out without
any visible reason: no resource bottlenecks registered, no
messages in /usr/adm logs, nothing special incron's own log,
very simple set of crontabs.  I don't see any dependencies
between this cron's deaths and what is going on in the system
at the time it happens.

I understand that there might be a zillion of reasons for
that, but I just hope that somebody might face things like
that before, or maybe this is some "well-known" problem
under some special circumstances.

Any clue or advice is appreciated, 'cause I just have never
seen anything similar before, and even don't have a slightest
idea about where to start investigation.

SCO UNIX 3.2.v4.2 (with a number of third-party products
installed) is running on Compaq Prosignia VS.

| Youri N. Podchosov (ynp)  ***  Davidsohn & Son, Inc. NYC  ***  718-234-4140 |



1. Why does my /etc/cron die?

I have a RS/6000-950 with AIX 3.1.5.   I am running hourly news processing
jobs (plus every 3 hours, and at 1:00 AM) under a normal account, besides the
standard root jobs.  Recently we ran out of /tmp space, and I found masses
of jobs named "croutUNIQUESTUFF" (you get the idea), many with 0 length
(OK, the job didn't produce output), some non-zero (gee, I haven't gotten
many cron mail messages recently, have I?).

I looked at /usr/adm/cron/log and saw this:

! *** cron started ***   pid = 28790 Thu Sep 12 03:00:08 1991
! *** cron started ***   pid = 77122 Thu Sep 12 03:07:48 1991
! *** cron started ***   pid = 33047 Thu Sep 12 05:00:22 1991
! *** cron started ***   pid = 36754 Thu Sep 12 06:00:12 1991
! *** cron started ***   pid = 30289 Thu Sep 12 07:00:10 1991
! *** cron started ***   pid = 30422 Thu Sep 12 08:04:20 1991
! *** cron started ***   pid = 30208 Thu Sep 12 09:00:03 1991
! *** cron started ***   pid = 12037 Thu Sep 12 10:00:02 1991

Showing cron seems to die and get restarted on almost every job!  That
explains the /tmp files not getting mailed and deleted, but why is cron dying?
Any ideas where to look?

By the way, a nearly identical setup running on a 320 for several months has
no problems at all!  cron stays up as long as the system.  The 950 has a
newer version of cnews, but no, it can't...

Thanks for any advice,

Steve Roseman
Lehigh University Computing Center

2. File corruption when running VMware.

3. Cron keeps dying- why?

4. HELP: Keyboard Lockup

5. Why does my cron daemon often die?

6. linux hangs during boot

7. Why cron dies all the time?

8. Error compiling dosemu

9. The named daemon silently dies!!

10. lpsched dies silently

11. 2.4.2-ax2 / ac3 die silently

12. ...and the pppd died silently

13. icc dies silently