Topic: TDM400P problem after kernel update

[soprom]

I've got incoherent informations concerning TDM400P on my system.
It is detected in the PCI section, but there is a problem:

loading tuning coefficients
open error on /dev/zap/1: No such file or directory

Starting asterisk:                                         [  OK  ]
[root@pbx ~]# cat /proc/zaptel/1
Span 1: WCTDM/0 "Wildcard TDM400P REV I Board 1" (MASTER)

           1 WCTDM/0/0 FXSKS (In use)
           2 WCTDM/0/1 FXOKS (In use)
           3 WCTDM/0/2 FXOKS (In use)
           4 WCTDM/0/3

Any help?

[SARK devs]

HI Sophie

Which asterisk rpms are you running?

Best

S

[soprom]

Version: sail-2.2.1-625

After applying the update, I followed a post here for a similar problem:
yum install zaptel-kmdl-`uname -r` --enablerepo=atrpms


# uname -r
2.6.9-67.0.7.EL

# rpm -qa  | grep zaptel
zaptel-kmdl-2.6.9-67.0.4.EL-1.4.10-50.el4
zaptel-kmdl-2.6.9-67.0.7.EL-1.4.10-50.el4
zaptel-1.4.10-50.el4

As you can see, zaptel-kmdl is there twice (i removed the old one when I saw this).

[SARK devs]

Hi Sophie,

The multiple kmdl rpms is normal  Each one installs the zaptel load modules into the correct /lib/modules/{kernelrelease}/extra for your current kernel (uname -r).   They are not versions of the same rpm (which is normally the case), each one is a separate rpm in its own right. Thus when you add a new one, it doesn't supercede the previous one, if that makes sense.

Anyhow, back to your problem...

I think that a formal stop/start of asterisk will cure your problem, or a stop followed  by a manual startup... as folows

Code: [Select]

/etc/init.d/sark stop
/etc/init.d/sark start
OR...

Code: [Select]

/etc/init.d/sark stop
modprobe wctdm
ztcfg -vv
ls /dev/zap
asterisk
we have a tdm board on one of our test servers here..  Here is what it should look like

Code: [Select]

[root@switch ~]# uname -r
2.6.9-55.0.9.EL

[root@switch ~]# rpm -q sail
sail-2.2.1-622

[root@switch ~]# /etc/init.d/sark stop
Shutting down asterisk:                                    [  OK  ]

[root@switch ~]# ls /dev/zap
ls: /dev/zap: No such file or directory

[root@switch ~]# ls /proc/zaptel
ls: /proc/zaptel: No such file or directory


OK its down. Now let's start it manually...

Code: [Select]

[root@switch ~]# modprobe wctdm
[root@switch ~]# ztcfg -vv
Notice: Configuration file is /etc/zaptel.conf
line 0: Unable to open master device '/dev/zap/ctl'

1 error(s) detected

[root@switch ~]# ztcfg -vv

Zaptel Version: 1.4.7.1
Echo Canceller: OSLEC
Configuration
======================


Channel map:

Channel 01: FXS Kewlstart (Default) (Slaves: 01)
Channel 02: FXS Kewlstart (Default) (Slaves: 02)

2 channels to configure.

Changing signalling on channel 1 from Unused to FXS Kewlstart
Changing signalling on channel 2 from Unused to FXS Kewlstart

[root@switch ~]# ls /dev/zap
1  2  3  4  channel  ctl  pseudo  timer  transcode
[root@switch ~]# cat /proc/zaptel/1
Span 1: WCTDM/0 "Wildcard TDM400P REV I Board 1"

           1 WCTDM/0/0 FXSKS (In use)
           2 WCTDM/0/1 FXSKS (In use)
           3 WCTDM/0/2
           4 WCTDM/0/3

[root@switch ~]# asterisk
[root@switch ~]#


Notice, I had to issue the ztcfg-vv twice.  This is because it takes udev a good few seconds to build the device trees and I was a bit too quick with the command . 

Kind Regards

S

[soprom]

I followed your steps and I got the trunk back into the list by stopping the pbx, then in PCI Telephony Cards, I clicked initialize. That made the card slot disappeared. Then I probed again and echo tuned the card.

After restarting the pbx, the zap trunk was back in the list. I did not reinstall the  kmdl rpm that I had deleted.

It is now back to normal.

Thanks!