Hi Brian,
I haven't reported this yet, as I don't have an overall picture of what
is happening, but....
A couple of weeks ago I had several machine lockups on the same day
while testing MFC/R2 with a tor2. It hasn't happened any more here. I
have no idea why it suddenly started or stopped. However, now people are
starting to deploy R2, I have reports of occasional lockups with tor2
cards. I have no idea if these lockups have the same cause as mine.
Regards,
Steve
Brian West wrote:
>Anything after these versions:
>
>zaptel.c version 1.95 (known working)
>chan_zap.c version 1.357 (known working)
>
>with a tor2 card... causes kernel panic...
>
>Can anyone else confirm this? I honestly think it's a combo issue with
the
>new zap reload and that zaptel change. But I have spent hours trying to
>narrow it down to those two files and those changes.
>
>Has anyone else seen strange issues when using PRI?
>
>If we have zaptel.c 1.95 and latest chan_zap.c you can place and take calls
>but if you do something like show channels at the CLI you'll deadlock
the
>box. I have no thread apply all bt since the glibc on this box didn't
have
>debug compiled in on it. (will retry this tomorrow)
>
>If you have the latest zaptel.c and the latest chan_zap.c placing any call
>out/in the zap interface will cause a kernel panic and kill the box.
>
>Use the above listed known working files and you have no problems.
>
>I would open a bug report but I would like to find more information before
>doing so.
>
>Thanks,
>Brian
>
>