It happened again at 9:00pm. Exim was the top 6 processes. The load was at 17. At this point Exim stopped exepting connections. The contents of ended with: handling mail from mx1.doc.state.vt.us. And it just hung there. At this point I stopped exim. Added: smtp_max_connections = 600(is this
peasable?). After I started it a couple of seconds later it got the same problem. So after stopping it and starting it many times I eventually recovered it(which is good, I didn't want to drive an hour to work). Is there some way I can invoke -d9 after the daemon is already started?
Is there a way I can put the output to a file rather that my screen? I tried exim -bd -q1h > /var/log/exim_debug
That didn't seem to work. Is the text stderr? Maybe exim -bd -q1h &2 > /var/log/exim_debug or something....I have no idea. This problem is like a snowball effect. Once one thread goes to the top of my process list, a short while after another, then another. Yet if none get caught there in the
first place Exim seems to run fine fr quite a long time.
Philip Hazel wrote:
>
> On Wed, 5 Dec 2001, Joseph Kezar wrote:
>
> > When I start exim my load jumps to 2 to 2.5
> > This in not warrented. We dont have many users. Running tail against /var/log/exim_mainlog shows nothing unusual. When I run top, the top 2 proccesses are exim both using 99% CPU. When I restart exim it corrects the problem for a couple of seconds(almost a minute). What could be wrong?
>
> What's the output from "exiwhat"? (Run it as root.)
> Which of those Exim proceses are using the cpu?
>
> If it's a delivery that's causing the problem, try running it with -d9
> to see what it is doing.
>
> --
> Philip Hazel University of Cambridge Computing Service,
> ph10@??? Cambridge, England. Phone: +44 1223 334714.
>
--
Joseph Kezar
