[Mailman-Users] kernel panic

[wiren@libero.it]

Firstly, many thanks to all the people who took their time to answer me about my mailman + exim question.

Now, the whole thing gets much more interesting...

As a matter of fact, I managed to edit a exim.conf file which is perfectly working with mailman 2.1.5; I followed guidelines into the readme.exim file, made several tries and now it is working.

But there is a little side effect. After half an hour the system is working, it starts getting slower and slower and it ends up with "Aiee, kernel panic"...

I tried a dozen of time. Every time freezes this way, then I try to reboot and it does not. I have to put a floppy, mount the system, re-run LILO and reboot. 

If I do not start mailman, the system goes smoothly. Right now, I am not running mailman and in fact the system is up and running.

BTW, every time I boot the server mailman does not start automatically at boot time. There is a message saying there is a stale master lock and stating I should use mailmanctl with the -s flag. This in fact is the way I use to start mailman, only to get stuck after 30 minutes.

Any idea of what could cause that?

And BTW would you please tell me 

1) where on the disk could I find messages waiting to be processed by mailman? Are there into a single indexed file or are there single files?

2) Can I delete manually lock files when mailman it is not running? I read through the docs and I saw these are files build by the qrunner process to avoid processing at the same time the same mailing list, so I guess while mailman is not running they should not exist... is it correct?

Thank you for your attention.

Wiren