ISDN Cause Code 47 Errors

Home » Asterisk Users » ISDN Cause Code 47 Errors
Asterisk Users 3 Comments

We fairly recently switched service providers for our 4 PRI circuits. Since that time, we started to notice some failed inbound calls. These calls terminate with an ISDN cause code 47 ‘resource unavailble’. Most of the time I see this error on the first or second channel on the second span in a trunk group (This is the providers trunk group for hunting, not an Asterisk trunk group). All the PRIs are setup as individual spans, we are not using NFAS. If the provider sets the hunt method to ‘least recently used channel’, then I can receive calls on other channels on the secondary span, it is just the first 2 that consistently fail.

We have had occasion where the error occurs on first span. If enough calls come in at the same time, callers who happen to land on channel 3 or above are OK. When the problem happens on the first span, if we physically disconnect the first span(RED alarm), the calls hunt to the second span and all calls seem to process properly. The only way to clear the cause 47
errors from the first span is a power cycle on the provider equipment. Power cycling my equipment does not solve the problem, only when I cycle their equipment.

The provider says the cause 47 is coming from my equipment, yet the ‘core set debug on’ log, unless I am reading it wrong, says it is coming from their side.

I have a second server as a backup. Both servers have identical hardware and software. When switching to the backup server, the problem remains.

I had the same setup on the previous provider, except using NFAS, and did not have this problem.

Am I reading the log correctly?
Do I have something setup incorrectly?
Is there any way to get even lower level debugging on the PRI?
Has anyone ever had this problem and if so what was done to resolve it?

Thanks in advance for any insight, Dale

============general config info
============There are two trunk groups for hunting by the provider TG1 = spans 2 and
4, TG2 = spans 1 and 3
I have two dual port cards in the server. Each card has 1 span from each TG
so the lost of a card will keep the other span operational. The provider does the same across VWIC cards. The spans do NOT use NFAS.

=================My equipment/software:
=================2x Digium TE210 dual T1/E1 with hw echo canceler and timing cable installed. Asterisk 1.8.12
libpri 1.4.12

=============Provider equipment:
=============Cisco 2821 with 2 VWIC2-@MFT-T1/E1

======================Log sample with ISDN debug on
======================[Jan 14 12:56:04] VERBOSE[13262] chan_dahdi.c: [Jan 14 12:56:04] PRI Span:
[Jan 14 12:56:04] VERBOSE[13262] chan_dahdi.c: [Jan 14 12:56:04] PRI Span:
4 < Protocol Discriminator: Q.931 (8) len

3 thoughts on - ISDN Cause Code 47 Errors

  • Yes. Asterisk has accepted the selected channel and CONNECTed the call. It is the peer that is disconnecting the call with cause 47. This really appears to be a problem in the providers equipment.

    I don’t see anything wrong.

    Not really. The problem is at layer 3 (Q.931) not layer 2 (Q.921). Turning on intense PRI debug will add nothing but noise to the debug log.

    ======================> Log sample with ISDN debug on


  • My guess is your provider did not have a free voice channel to pass audio at some leg in the call. There could be multiple legs in the call and one of them had ‘Network Congestion’.

  • Thanks Richard and Andres.

    I had come to the same conclusion, however the provider was fairly snarky in saying is was my equipment.

    We were able to replace the Cisco 2800 with a Cisco 2900 series and the problem appears to have been resolved.

    Thanks again, I always appreciate another set of eyes just in case I missed something.