* Yann Golanski <yann@???> [20040308 12:43]: wrote:
> Quoth Woon Wai Keen @ doubleukay.com on Sun, Mar 07, 2004 at 20:24:12 +1100
> > Is anyone using this combination?
>
> I am using 5.2.1-RELEASE-p1 and exim-4.30_1 with no problems.
>
> > I was previously using freebsd 5.1 +
> > exim 4.30 and it worked smoothly. However, after I upgraded (cvsup, make
> > build/installkernel, make build/installworld) to 5.2.1, exim suddenly
> > started barfing.
> >
> > I get lots of segfaults like these:
> >
> > (/var/log/dmesg.today)
> > pid 97038 (exim-4.30-0), uid 26: exited on signal 11
> > pid 97042 (exim-4.30-0), uid 26: exited on signal 11
> > pid 97249 (exim-4.30-0), uid 26: exited on signal 11
> > pid 97323 (exim-4.30-0), uid 26: exited on signal 11
> >
> > And lots of strange 'space check failed' warnings like these:
> >
> > (/var/log/exim/mainlog)
> > 2004-03-07 03:58:21 spool directory space check failed: space=0 inodes=0
> > 2004-03-07 03:58:36 spool directory space check failed: space=0 inodes=0
For the record, I had the same problem on two 5.x servers after I
installed exim from the ports tree. Well, same OS version, same Exim
version. On one such server, I have
mailserver# df -h
Filesystem Size Used Avail Capacity Mounted on
/dev/ad0s1a 34G 2.2G 29G 7% /
devfs 1.0K 1.0K 0B 100% /dev
There is sufficient space, but in both cases, I happened to have put in
two config params in the Exim configure file:
#check_spool_inodes = 100
#check_spool_space = 20M
I had believed those would be useful later, because anyway those two
servers do not handle so much mail!
After I commented those out and gave exim daemon a -HUP signal the
problem disappeared, from both boxes. To be sincere, one box brought the
problem first and I commented out those entries. I used the same info
from my mind cache to solve the other. I am wondering if this folk has
the same entries. Maybe??
> > If anyone's using freebsd 5.2.1 and exim 4.30, could you please reply
> > and tell me how's it going on your setup? Thanks.
You mean like I've mentioned above? Apart from those, I do run Exim 4.30
on several boxes here.
> Do you have a corrupted hard disk? What's the inode table saying? Any
> other messages in the logs? What's your disk usage? What does exim -d
> say?
I do believe sure his disk is OK for all those questions ;)
cheers
- wash
+----------------------------------+-----------------------------------------+
Odhiambo Washington . WANANCHI ONLINE LTD (Nairobi, KE) |
<wash at wananchi dot com> . 1ere Etage, Loita Hse, Loita St., |
GSM: (+254) 722 743 223 . # 10286, 00100 NAIROBI |
GSM: (+254) 733 744 121 . (+254) 020 313 985 - 9 |
+---------------------------------+------------------------------------------+
"Oh My God! They killed init! You Bastards!"
--from a /. post
