2 HB8 cards in one server – first one is not recognized, the second is

Home » Asterisk Users » 2 HB8 cards in one server – first one is not recognized, the second is
Asterisk Users 3 Comments

Hi list,

I installed 2 HB8 cards each of them with a Quad Bri modules in a HP 360
G6 running Debian Squeeze. Here is an output of dmesg wafter server has
booted:

[ 9.784123] wctdm24xxp 0000:0b:08.0: PCI INT A -> GSI 31 (level, low)
-> IRQ 31
[ 11.847073] bnx2: eth0 NIC Copper Link is Up, 1000 Mbps full duplex
[ 11.847600] ADDRCONF(NETDEV_CHANGE): eth0: link becomes ready
[ 11.859589] wctdm24xxp 0000:0b:08.0: Timeout waiting for receive frame.
[ 14.871333] wctdm24xxp 0000:0b:08.0: Timeout waiting for receive frame.
[ 14.871404] wctdm24xxp 0000:0b:08.0: The firmware may be corrupted.
Please completely power off your system, power on, and then reload the
driver with the ‘forceload’ module parameter set to 1 to attempt recovery.
[ 14.893874] wctdm24xxp 0000:0b:08.0: PCI INT A disabled
[ 14.893886] wctdm24xxp: probe of 0000:0b:08.0 failed with error -5
[ 14.893911] alloc irq_desc for 30 on node -1
[ 14.893913] alloc kstat_irqs on node -1
[ 14.893919] alloc irq_2_iommu on node -1
[ 14.893927] wctdm24xxp 0000:0e:08.0: PCI INT A -> GSI 30 (level, low)
-> IRQ 30
[ 16.915156] wctdm24xxp 0000:0e:08.0: Timeout waiting for receive frame.
[ 17.924645] wctdm24xxp 0000:0e:08.0: firmware: requesting
dahdi-fw-hx8.bin
[ 17.953971] wctdm24xxp 0000:0e:08.0: Hx8 firmware version: 2.06
[ 18.840157] wctdm24xxp 0000:0e:08.0: Port 1: Installed — BRI
quad-span module
[ 18.943159] wctdm24xxp 0000:0e:08.0: Port 2: Installed — BRI
quad-span module
[ 19.047162] wctdm24xxp 0000:0e:08.0: Port 3: Installed — BRI
quad-span module
[ 19.151164] wctdm24xxp 0000:0e:08.0: Port 4: Installed — BRI
quad-span module
[ 19.484173] wctdm24xxp 0000:0e:08.0: Port 5: Not installed
[ 20.099191] wctdm24xxp 0000:0e:08.0: Port 6: Not installed
[ 20.407198] wctdm24xxp 0000:0e:08.0: Port 7: Not installed
[ 20.715206] wctdm24xxp 0000:0e:08.0: Port 8: Not installed
[ 20.725444] wctdm24xxp 0000:0e:08.0: Found a Hybrid card: HB8-0000 (4
BRI spans, 0 analog channels)
[ 21.086213] wctdm24xxp 0000:0e:08.0: xhfc: Configuring port 0 span 1
in TE mode with termination resistance DISABLED
[ 21.136213] wctdm24xxp 0000:0e:08.0: xhfc: Configuring port 2 span 2
in TE mode with termination resistance DISABLED
[ 21.185214] wctdm24xxp 0000:0e:08.0: xhfc: Configuring port 1 span 3
in TE mode with termination resistance DISABLED
[ 21.232215] wctdm24xxp 0000:0e:08.0: xhfc: Configuring port 3 span 4
in TE mode with termination resistance DISABLED

You can see that first cerd failed with error -5 and firmware not loaded
as the second card is OK.

I thought about motherboard problem, HP already change it.

In the output of lspci -v I see that there is no kernel driver for the
first card:

0b:08.0 Ethernet controller: Digium, Inc. Hx8 Series 8-port Base Card
(PCI-Express) (rev 11)
Subsystem: Digium, Inc. Hx8 Series 8-port Base Card (PCI-Express)
Flags: medium devsel, IRQ 31
I/O ports at 5000 [size=256]
Memory at fbef0000 (32-bit, non-prefetchable) [size=1K]
[virtual] Expansion ROM at e4200000 [disabled] [size=128K]
Capabilities: [c0] Power Management version 2

0e:08.0 Ethernet controller: Digium, Inc. Hx8 Series 8-port Base Card
(PCI-Express) (rev 11)
Subsystem: Digium, Inc. Hx8 Series 8-port Base Card (PCI-Express)
Flags: bus master, medium devsel, latency 64, IRQ 30
I/O ports at 6000 [size=256]
Memory at fbff0000 (32-bit, non-prefetchable) [size=1K]
[virtual] Expansion ROM at e4100000 [disabled] [size=128K]
Capabilities: [c0] Power Management version 2
Kernel driver in use: wctdm24xxp

before asking RMA for the card, I would like to know what you think
about this matter.

Thanks for any hint

3 thoughts on - 2 HB8 cards in one server – first one is not recognized, the second is

  • First, Digium technical support would be more than happy I’m sure to
    help you trouble shoot this. That being said…

    First thing I would do is update to the current trunk of dahdi-linux.
    Revision 9397 [1]
    http://svn.asterisk.org/view/dahdi?view=revision&revision=9397 was added
    because of some systems that did not provide reliable polling from the
    board side, which could result in erroneous “your firmware may be
    corrupted…” messages. However, since you have one card that works and
    one that doesn’t I give this a low probability of fixing it.

    Next, if updating the driver does not help and if the problem follows
    the card (i.e., you can swap cards and now the second card fails to
    load), I would disable dahdi from starting automatically, power off your
    system, remove the working card, power on, and try “modprobe wctdm24xxp
    forceload=1″ on the chance that the firmware on the board actually is
    corrupted.

    If neither of those things work, you may need to RMA your card.

    Cheers,
    Shaun

  • Le 26/10/2010 14:49, Shaun Ruffell a écrit :
    Didn’t test this yet but
    switching cards gives kernel panic 🙁 on boot
    Will try card by card, then slot per slot

    Thanks for your help

  • Le 26/10/2010 14:49, Shaun Ruffell a écrit :
    Installed trunk from today. Same result.
    [ 689.968684] wctdm24xxp 0000:0b:08.0: PCI INT A -> GSI 31 (level, low)
    -> IRQ 31
    [ 692.011458] wctdm24xxp 0000:0b:08.0: Timeout waiting for receive frame.
    [ 692.015668] wctdm24xxp 0000:0b:08.0: firmware: requesting
    dahdi-fw-hx8.bin
    [ 692.039148] wctdm24xxp 0000:0b:08.0: Reloading firmware. Do not power
    down the system until the process is complete.
    [ 694.055980] wctdm24xxp 0000:0b:08.0: Timeout waiting for receive frame.
    [ 694.056052] wctdm24xxp 0000:0b:08.0: Hx8 firmware version: 1.128
    [ 694.078435] wctdm24xxp 0000:0b:08.0: PCI INT A disabled
    [ 694.078443] wctdm24xxp: probe of 0000:0b:08.0 failed with error -5

    I will do it.

    BTW, does someone use Digium cards -specially HB8- in HP 360 G6 servers?
    I have some doubts about compatibility of those machine and telephony cards.

    Regards