If quotas are enabled, kernel spontaneously reboots after about five hours of uptime. With quotas disabled, I get about a day and a half before it hangs 'til it gets a manual reboot (yes, I know, that sounds like two completely unrelated problems). Fix: The quota problem is worked-around by... well... disabling quotas :-/ I haven't yet worked out what to do about the hangs, other than asking my users to please put up with it 'til we can work something out. Unfortunately, when a system is catatonic, you can't break out to the kernel debugger to find out why... How-To-Repeat: Boot 'er up and leave it running multiuser for a couple of days. Unfortunately, I can't isolate it to a single bit of code: I have binaries here from FreeBSD 1.0, 1.1 and 2.0 being banged on by about 300 users (not all at once :-). Any one of them at any time could be triggering it. The system is also acting as a secondary nameserver, news server for about 3 dozen sites, secondary MX forwarder for about 70 UUCP, dialup SLIP and UUCP sites, NFS server for a client with far too little disk space, proxy caching HTTP server, a MOO, fileserver for a Sun 3/60 acting as an X11R6 X-terminal, and a dialup SLIP/PPP server. I guess what I'm trying to say is that with everything else that's going on, I haven't been able to narrow down the problem at all. I'm not being very helpful, am I? :-)
State Changed From-To: open->closed Most likely fixed by 2.0.5 or 2.1. Even if its not, the information in this PR is out of date and a new PR should be logged aginst -stable or -current if the problem still exists.