I’m using a TE207P board installed in a new system running fedora core 6. The hardware is an Intel D945 GTP with a P4 dual core CPU, 1G RAM, and using software raid mirror with two 150G SATA drives.
About half the time I boot, the system hangs when the /etc/init.d/zaptel script attempts to load the zaptel modules. I cannot see any reason to it - its totally random. I’ve removed the second NIC on the PCI bus and moved the TE207P to another slot and still have the same issues. I’ve tried this with both 2.6.18 and 2.6.19 kernels with the same results.
I am not sure if it is the loading of the modules that causes the problem, or when the card is actually provisioned that things go wrong.
The relevant portions of my zaptel.conf file are (everything else is commented out):
Span 1 is connected to the PSTN
Span 2 is connected to a channel bank
and from zapata.conf
If I comment out these lines shown above, the system boots reliably, which would seem to suggest I am setting this up incorrectly. I have the two DS-1 ports connected together using a cross-over cable. When things do load properly, I can use a SIP client to place an outbound call (e.g. 95551234) and sample dialplan calls out on the 1st DS-1 port and answers on the 2nd DS-1 port. This is just for testing purposes and intend to create a more meaningful configuration for deployment.
Am I doing something really wrong in the zaptel or zapata files? Any thoughts or insight would be appreciated.
********** Update ****************
I have verified that the kernel panic occurs when ztcfg runs. As I mentioned before, if I leave the zaptel.conf and zapata.conf files empty, then the machine boots reliably and the zaptel modules are always loaded. With the machine running, I re-add the lines to the conf files and run ztcf -vv and the machine freezes. I updated the subject line to reflect this new understanding. Any help is appreciated.