Just putting this out there to see if anyone else has seen any issues.
May cross-post to asterisk-dev if it's indeed a bug (and not my own
stupidity).
I've got a Digium TE220 (2xT1 interface w/Echo canceller) that in two
separate Nehalem-based (Xeon E5520 "Gainestown") boxes (HP ProLiant
ML350 G6; HP Z800 Workstation) has caused numerous kernel panics. This
is only when the dahdi service is running with a very simple config
(I've defined the first span, the bchans and the dchan, and that's
about it). If dahdi is stopped, or the card is removed, everything's
fine. I instead installed a Digium TE122P in the ML350 and haven't had
any issues. I also haven't seen this in a pre-Nehalem Xeon server.
I'm using Asterisk 1.6.1.6, Dahdi 2.2.0 and LibPRI 1.4.10.1, running
on CentOS 5.3 (2.6.18-164.el5).
Has anyone seen anything similar?
- Chris
