Stuart Henderson wrote: >> (em1) as up and few seconds later I mark the same interface down. My >> network usage drops significantly, currently I am looking it shows 75%. > > Do you mean interrupt%? bsd.mp will probably drop that *way* down.
No, memory management routines (netstat -m). Are you saying I should try with bsd.mp kernel with a single core cpu (opteron 146)? > You may need to also disable USB (may have been one of the things fixed > by moving to either acpi or a newer bios, I don't recall). Damn....I was speaking too fast, it just locked up again. This time after 38 hours, it is an improvement, though. > Most of mine are on original bios, -current from around the time 4.1 was It is a production router so I can't update to current just now, at least until I get a backup. I'll try to update bios later today. > tagged, with acpi on, USB off. They don't crash any more, but I have > recently noticed some odd problem with packet loss with the onboard > bge(4) though; if I ping *from* the h8ssl to another box there's no > trouble, if I ping from elsewhere to the h8ssl I get periods of a few > (1-10) minutes at a time with usually an hour or two between them > where there's 0.5 to 1% loss, a couple of RTM_LOSING, and sometimes > bad enough to cause bgp hold timers to expire on the ibgp sessions > to other h8ssl boxes. I could try that on my box... > the diff to if_em.c is 1000 lines. Some of it's PCIE support, but > there's plenty else. I will made an update as soon as possible. > The one I have with em(4) works absolutely fine now. Good to know. Thank you for you hints. Mitja