[asterisk-users] Kernel panic w/ DAHDI 2.x/Digium TE220B

Chris Brentano chris.brentano at jivesoftware.com
Tue Oct 20 17:25:58 CDT 2009


I've seen this consistently on three systems, with three different  
cards, and multiple versions of DAHDI. At first I thought the issue  
only occurred on newer, Nehalem-based, systems, but I reproduced it on  
a Core 2 Duo box as well. I've tested with dahdi-linux 2.2.0.2, dadhi- 
linux-complete 2.0.0+2.0.0, 2.1.0.2+2.1.0.2, and 2.2.0.2+2.2.0. The  
card is a Digium TE220B which uses the wct4xxp module. This does not  
happen, on the same systems and kernel version, with a TE121 using the  
wcte12xp module nor does it happen with a T100P using wct1xxp. OS is  
CentOS 5.3, and happens with kernel versions 2.6.18-164.el5 and  
2.6.18-128.el5. I'm posting this wondering if anyone else has seen  
similar behavior.

/etc/dahdi/system.conf:
   span=1,1,0,esf,b8xs
   bchan=1-23
   dchan=24
   loadzone=us
   defaultzone=us

/etc/dahdi/modules:
   wct4xxp
   wcte12xp
   wct1xxp

---

When I start dahdi, I see the following:

   # /etc/init.d/dahdi start
   Loading DAHDI hardware modules:
      wct4xxp:          [ OK ]
      wcte12xp:         [ OK ]
      wct1xxp:          [ OK ]

   Running dahdi_cfg:   VPM400: Not Present
   VPM450: Not Present
                        [ OK ]

Syslog output:

   Oct 20 15:20:54 redbox-ast16 kernel: dahdi: Telephony Interface  
Registered on major 196
   Oct 20 15:20:54 redbox-ast16 kernel: dahdi: Version: 2.2.0.2
   Oct 20 15:20:54 redbox-ast16 kernel: ACPI: PCI Interrupt  
0000:03:08.0[A] -> GSI 16 (level, low) -> IRQ 169
   Oct 20 15:20:54 redbox-ast16 kernel: Found TE2XXP at base address  
dfbfff80, remapped to ffffc20000022f80
   Oct 20 15:20:54 redbox-ast16 kernel: TE2XXP version c01a016c, burst  
ON
   Oct 20 15:20:54 redbox-ast16 kernel: Octasic optimized!
   Oct 20 15:20:54 redbox-ast16 kernel: FALC version: 00000005, Board  
ID: 00
   Oct 20 15:20:54 redbox-ast16 kernel: Reg 0: 0x056af400
   Oct 20 15:20:54 redbox-ast16 kernel: Reg 1: 0x056af000
   Oct 20 15:20:54 redbox-ast16 kernel: Reg 2: 0x00000000
   Oct 20 15:20:54 redbox-ast16 kernel: Reg 3: 0x00000000
   Oct 20 15:20:54 redbox-ast16 kernel: Reg 4: 0x0000ff01
   Oct 20 15:20:54 redbox-ast16 kernel: Reg 5: 0x00000000
   Oct 20 15:20:54 redbox-ast16 kernel: Reg 6: 0xc01a016c
   Oct 20 15:20:54 redbox-ast16 kernel: Reg 7: 0x00001000
   Oct 20 15:20:54 redbox-ast16 kernel: Reg 8: 0x00000000
   Oct 20 15:20:54 redbox-ast16 kernel: Reg 9: 0x00ff00ff
   Oct 20 15:20:54 redbox-ast16 kernel: Reg 10: 0x0000004a
   Oct 20 15:20:54 redbox-ast16 kernel: Found a Wildcard: Wildcard  
TE220 (4th Gen)
   Oct 20 15:20:54 redbox-ast16 kernel: TE2XXP: Launching card: 0
   Oct 20 15:20:54 redbox-ast16 kernel: TE2XXP: Setting up global  
serial parameters
   Oct 20 15:20:55 redbox-ast16 kernel: About to enter spanconfig!
   Oct 20 15:20:55 redbox-ast16 kernel: Done with spanconfig!
   Oct 20 15:20:55 redbox-ast16 kernel: dahdi: Registered tone zone 0  
(United States / North America)
   Oct 20 15:20:55 redbox-ast16 kernel: About to enter startup!
   Oct 20 15:20:55 redbox-ast16 kernel: TE2XXP: Span 1 configured for  
ESF/B8ZS
   Oct 20 15:20:55 redbox-ast16 kernel: wct2xxp: Setting yellow alarm  
on span 1
   Oct 20 15:20:55 redbox-ast16 kernel: timing source auto card 0!
   Oct 20 15:20:55 redbox-ast16 kernel: SPAN 1: Primary Sync Source
   Oct 20 15:20:55 redbox-ast16 kernel: VPM400: Not Present
   Oct 20 15:20:55 redbox-ast16 kernel: VPM450: Not Present
   Oct 20 15:20:55 redbox-ast16 kernel: Completed startup!

---

Now if I either start asterisk, or if I stop dahdi, it will panic:

   # /etc/init.d/dahdi stop
   Unloading DAHDI hardware modules:   TE4XXP: Version Syncronization  
Error!
   TE4XXP: Version Syncronization Error!
   TE4XXP: Version Syncronization Error!
   TE4XXP: Version Syncronization Error!



   HARDWARE ERROR
   CPU 1: Machine Check Exception:                      4 Bank 8:  
00000000000000
   TSC 0
   This is not a software problem!
   Run through mcelog --ascii to decode and contact your hardware vendor
   Kernel panic - not syncing: Uncorrected machine check


Syslog output (not much before restart):

   Oct 20 07:11:54 localhost kernel: TE4XXP: Version Synchronization  
Error!
   Oct 20 07:14:24 localhost syslogd 1.4.1: restart.
   ...

---

I only see the machine check exception on the two Nehalem boxes (HP  
ProLiant ML350 G6, Z800 Workstation); on a Core 2 Duo (Dell Optiplex  
745) it just hard freezes after the "Version Syncronization Error!"  
messages. If there's any further details I can provide I'm happy to do  
so. Would like to figure out what's happening here if anyone can help  
shed any light as this is completely holding up migration to Asterisk  
1.6 and DAHDI. Thanks.

- Chris




More information about the asterisk-users mailing list