appliance

A200 dropping calls

tjthorson
Posts: 79
Member Since:
2007-01-15

We had a problem with a PIII 800 Mhx box, 512MB RAM, and a Sangoma A200 with EC and 4 FXO ports connected directly to POTS.

On trunk 1 - 1 out of every 3 or 4 calls outbound would not go. No static, no noise, after the send button was pressed, just a click - and silence. I even had users hang on (on speaker) for over 5 minutes - no telco warning, no timeout. To troubleshoot, they switched the wires on lines 1 and 2 - still had the trunk 1 problem. If someone tied up trunk one (with a dead call) - all other lines 2-10 would go out fine.

Had the onsite person switch the FXO boards (figuring the problem should move to trunk 3 if my theory is correct.).

Now, all calls are going through fine - including trunk 3. But now, random calls are being dropped. During the call, users hear static, a few popping noises and the call disconnects. Logs in Asterisk show "hangup".

Not sure what to do here - I suspect a hardware issue - but this will be my 3rd Sangoma A200 card issue out of 5 we have bought. Are they THAT bad? Is it something I can change (maybe the Telco voltage is off a little?) - Any ideas?

This is a Trixbox 2.2 original - with all the updates I could install remotely.

I am in Illinois, this box is in Ohio - so I can only do SSH, Webmin, and TB GUI to it.

--

Todd J Thorson
IT Manager
todd@toddware.com



charles05663
Posts: 69
Member Since:
2007-07-31
I was having some issues

I was having some issues with my A200DE and found out that it was an interrupt issue. The system was using the "old" style of interrupts and when when I switched it to the newer style everything worked fine.



jpatel
Posts: 44
Member Since:
2007-05-24
Confirm

Noise,static or random hang up issue confirm following :

1) Run "ifconfig" few times and confirm "tx" and "rx" counters are incrementing on wanpipe interface (e.g. w1g1). Also confirm there are no overruns on wanpipe interface, if you have any check how fast they are incrementing

2) run "cat /proc/interrupts" and confirm that wanpipe is generating roughly 1000 interrupts per second

--

Jignesh Patel - Sangoma Technologies Inc.



tjthorson
Posts: 79
Member Since:
2007-01-15
I checked both... RX and TX

I checked both... RX and TX are climbing together each second - interrupts is at least 1000 per second.

w1g1 Link encap:Point-to-Point Protocol
UP POINTOPOINT RUNNING NOARP MTU:8 Metric:1
RX packets:497488148 errors:0 dropped:0 overruns:0 frame:0
TX packets:497488148 errors:0 dropped:0 overruns:0 carrier:0
collisions:0 txqueuelen:100
RX bytes:3979905184 (3.7 GiB) TX bytes:3979905184 (3.7 GiB)
Interrupt:7 Memory:e0a60000-e0a61fff
***************************************************************************
CPU0
0: 498054931 XT-PIC timer
1: 8 XT-PIC i8042
2: 0 XT-PIC cascade
7: 497733344 XT-PIC wanpipe1
8: 1 XT-PIC rtc
9: 0 XT-PIC uhci_hcd
10: 9934188 XT-PIC eth0
11: 452442 XT-PIC libata
15: 1133 XT-PIC ide1
NMI: 0
ERR: 0
**************************************************************************

-------------------------------
| Wanpipe Hardware Probe Info |
-------------------------------
1 . AFT-A200-SH : SLOT=15 : BUS=0 : IRQ=7 : CPU=A : PORT=PRI : HWEC=32 : V=08

Card Cnt: S508=0 S514X=0 S518=0 A101-2=0 A104=0 A300=0 A200=1 A108=0 A056=0
***************************************************************************
I did upgrade the drivers to the latest ones in the package manager - that seems to have helped....

Todd J Thorson
IT Manager
todd@toddware.com

--

Todd J Thorson
IT Manager
todd@toddware.com



charles05663
Posts: 69
Member Since:
2007-07-31
You are running the old

You are running the old interrupts. This was the same problem I had. You can edit your:
/boot/grub/menu.lst
and replace apci=off with pci=nommconf

I know I did this and it corrected any issues that I had with the card.



tjthorson
Posts: 79
Member Since:
2007-01-15
Thanks for the help - this

Thanks for the help - this is my line in my menu.lst:

kernel /vmlinuz-2.6.9-34.0.2.EL ro root=LABEL=/1 acpi=off noapic

do I also remove the 'noapic" ???

So - it should be:
kernel /vmlinuz-2.6.9-34.0.2.EL ro root=LABEL=/1 pci=nommconf

Now - I have to do this remotely - the server is out of state - is there any chance this thing wouldn't boot to where i couldnt get back to undo the change if something goes wrong??? I cant risk that.... It seems to be better now with the newer drivers - only 1 bad call in 2 days so far.... But, if what you wrote is it for sure - and the machine will boot up for sure (at least to get SSH going again) - then I will try it.

Todd J Thorson
IT Manager
todd@toddware.com

--

Todd J Thorson
IT Manager
todd@toddware.com



charles05663
Posts: 69
Member Since:
2007-07-31
I made a backup of my

I made a backup of my grub.conf first.

Here was the original file:
kernel /vmlinuz-2.6.9-34.0.2.ELsmp ro root=LABEL=/1 acpi=off

here is the modified file:
kernel /vmlinuz-2.6.9-34.0.2.ELsmp ro root=LABEL=/1 pci=nommconf

After this I rebooted and have had no problems.

This link might help:
http://www.trixbox.org/forums/trixbox-forums/open-discussion/if-y...



tjthorson
Posts: 79
Member Since:
2007-01-15
Thanks for the help. Again,

Thanks for the help. Again, I am not a Linux expert AT ALL.

Is there ANY chance that the system wont boot enough for me to SSH back in to undo this change if it doesnt work? I have no one onsite - it has to be all remotely.

And.... what is the "noapic" - do I remove that too?

Thanks again.... I just cant have the whole box not boot - I need to be sure.

Todd J Thorson
IT Manager
todd@toddware.com

--

Todd J Thorson
IT Manager
todd@toddware.com



charles05663
Posts: 69
Member Since:
2007-07-31
Did you install the system?

Did you install the system?

I am not a Linux person either (I am an OpenBSD user) but there is really only one way to learn and that is under fire.

Also, there is never a sure thing except aging, death and judgment.



tjthorson
Posts: 79
Member Since:
2007-01-15
Yes, I installed it.

Yes, I installed it. Normally, trial under fire is fine, except I am two states away and cant touch the box to reload it, or boot a CD to get back to the partition to undo the change. Its one thing if Asterisk doesnt load, or something else like that - but if it doesnt boot to the point where I can SSH, I cant take that risk.... I am going to ship a Knoppix CD to the office so they can have that onhand first.....

Thanks.

Todd J Thorson
IT Manager
todd@toddware.com

--

Todd J Thorson
IT Manager
todd@toddware.com



tjthorson
Posts: 79
Member Since:
2007-01-15
Update: I left the noapic

Update: I left the noapic in and changed the line as instructed. It now looks like:

kernel /vmlinuz-2.6.9-34.0.2.EL ro root=LABEL=/1 pci=nommconf noapic

So far - no complaints, dropped calls or static. Thanks again for the help!

--

Todd J Thorson
IT Manager
todd@toddware.com



daman
Posts: 14
Member Since:
2006-06-02
IT Freak'n WORKED!

My problem was this. Every time I talked through the phone I would hear a little crackling sound trailing my voice. Then there was a lot of static. I added the line just like you said above (pci=nommconf) and now it is crystal clear. Thank you so much. I owe someone a beer.

Tech Specs:
Trixbox 2.6.0.7
Songoma A200 (6 FXO's)
Polycom 601
Linksys new 10/100 NIC
ACER Inter Core 2 Duo



josen
Posts: 4
Member Since:
2008-04-14
With Zaptel trunk using Sangoma card, I can't dial IVR options

Hi everyone,

This is to kindly request a help for the problem I have:
When calling from outside using the Zaptel Trunk I can reach the IVR but can't dial the IVR options using the GSM cellular phone, but with CDMA cellular phone it works ! I think that this has something to do with DTMF.

So,how I can get the GSM cellular phone working as it should with the IVR is still a mystery to me.
The other problem I'm having even with CDMA cellular phone is that instead of going straight to the IVR when I call, it first rings for a few seconds and then the IVR announcements are heard.

I'm running:
Trixbox2.2.5
FreePBX 2.3.1.0
with a Sangoma A200-BRMD card.

Thanking you in advance



blockparty
Posts: 61
Member Since:
2007-01-07
Add to zapata.conf

add

relaxdtmf=yes

to your zapata.conf file and this took care of it for me. Hopes this helps.



marc.sangoma
Posts: 64
Member Since:
2008-04-08
Hi,

Hi,

For the DTMF issue you can go to http://wiki.sangoma.com/wanpipe-linux-asterisk-debugging#dtmfReco... for steps on how to debug it. Also another thing that you can try is bringing the call into Asterisk and then go into the Asterisk CLI and set the verbosity to 10 with the command "set verbose 10". With the verbosity up this high you can see the DTMF digits that Asterisk is getting, and this will tell you if the digits are getting to the system and if they are coming in correct. If this does not help you please send an email to techdesk@sangoma.com with a summary of your issue and i will assist you from there.

Marc
Sangoma Technologies



Comment viewing options

Select your preferred way to display the comments and click "Save settings" to activate your changes.