Mysterious reboot

[Robin Vley]

Robert Huff wrote:

Robert,

 >     I have a -CURRENT machine - P4, 512mb RAM, SCSI disks,
 > de-driver ethernet, all mainstream hardware - that has this problem.
 > Since early last year (maybe longer) it will occasionally reboot.


Actually, I had it on three machines now. First on a P4 I leased in the 
US. Then on the dual Athlon MP with 3ware raid that I own and THEN on 
the SuperMicro machine I currently use. After the first reboot on that 
machine (8 days ago now) I did some things:

- removed all monitoring done by CPanel (except the cpanel monitor itself)
- Upped the maxfiles and vnodes a bit
- Removed an ipfw rule that counted everything that goes out and comes 
in. I used this for MRTG graphing, but now I changed to SNMP polling on 
a seperate cacti machine.

So far it's 8 days (a NEW record for my FreeBSD machines, woopie), knock 
  knock. It's possible it's one of the things above that did the trick, 
or that it's just a few more days untill the next reboot. We'll see.

 > No warning, no panic, no core-dump, just - bang.  If I was lucky it


Right. Same here, on all machines.

 >     After an update in October (??), things slowly got better;
 > several further updates and I haven't seen it in weeks.


I'll keep an eye on it too. Hope it's working like it should. If not 
it's time to start looking into some way of dumping run data or doing 
some serial logging or so.



/Robin