WARNING[549] chan_iax2.c: Max retries exceeded to host x.x.x

I’ve an * box up and running on an ADSL line. The purpose with the * box is to provide a conferance-service. However these errors occur randomly when people are connected to the conferance. This happens once in 3-4 hours. Is my connection to poor? Is meetme buggy?

When these errors occur, the call is terminated.
My IAX.conf is (slightly censored):

Any suggestions?
Many thanx in advance!
/Johan

I’ve browsed this forum reading about meetme. Many spoke of how poor ztdummy is (that I am using) and that ztdummy cause “voice choppines”. This is also true, not disturbingly much, but it is noticeable.

Is it maybe ztdummy that cause these errors? Any suggestions about the topic?

My system has an X110P clone card in and I get no problems with conferencing or MOH. Perhaps you could obtain one from Ebay, they are very cheap.
Just install it and configure it but do not assign it to be used by any calls.

That is assuming that this is your problem, I am not experienced enough to know for sure.

Mike.

The message you included comes from the IAX2 channel driver. The message is telling you that your Asterisk machine has tried sending data to the IP address listed and it couldn’t not deliver the data. The IAX2 protocol specifies a maximum number of times to try resending the data before the connection is terminated.

This is likely a network issue. Examine the connection between your Asterisk machine and the machine listed in the error message. Look at the codecs you are using, the bandwidth available, latency, etc…

Dan

Thanks for helping!

As I said, I’ve the server on a ADSL-connection. Is it normal to get these error then? Do you think this will resolve if I try colocation with ethernet connection to the Internet?

I read somewhere that you could increase the number of retries. Any comments on this issue?

And thanks again :smile:

These messages are not really “normal”, in the sense that they definitely indicate a problem. However, they are “normal”, in that under certain conditions you can get them, and they can be corrected.

I would suggest attempting to replicate the problem in a controlled situation. If you can do that, look at things like bandwidth usage on your ADSL line, CPU usage on your server, and latency between hosts and the server.

You can use ‘top’ to check CPU usage, and ‘iftop’ to check network usage on the server. Possibly look at using ‘mrtg’ to monitor your ADSL router’s usage.

Dan

Hello, the thing that illudes me is the frequency of the error. It seems to be totaly sporadic, which makes it very hard to monitor and identify.

This for example is the statistics of a trial tonight. The session lasted about 2 hours and this is the statistics. No droped packets at all.

andromeda*CLI> iax2 show netstats
                                -------- LOCAL ---------------------  -------- REMOTE --------------------
Channel                    RTT  Jit  Del  Lost   %  Drop  OOO  Kpkts  Jit  Del  Lost   %  Drop  OOO  Kpkts
IAX2/teleman_out-5          60   -1    0    -1  -1     0   -1      3    0    0     0   0     0    0      0
IAX2/teleman_out-6          41   -1    0    -1  -1     0   -1      2    0    0     0   0     0    0      0
2 active IAX channels

Can the ADSL-line be affected by circumstances that makes it go down for a breif moment, and that isn’t tolerated by the iax2-channeldriver?
I have planed to try to replicate the error in a LAN, setting up SIP-phones just to eliminate the possibility of a bug in the meetme-application.

For the moment I have patched chan_iax2.c in this way:

--- channels/chan_iax2.c.orig   Mon Oct 30 03:51:58 2006
+++ channels/chan_iax2.c        Mon Oct 30 03:52:22 2006
@@ -148,7 +148,7 @@
 static char regcontext[AST_MAX_CONTEXT] = "";

 static int maxauthreq = 0;
-static int max_retries = 4;
+static int max_retries = 10;
 static int ping_time = 20;
 static int lagrq_time = 10;
 static int maxtrunkcall = TRUNK_CALL_START;

I have adjusted the max_retries value from 4 to 10.
I will report back any progress made.
Please feel free to suggest solutions, and many thanks in advance!

I’ve been running this setup since the previous posts and this works. Now Asterisk 1.4 have been released, and I have no idea if the behaviour will change in this version.

I will post updates in this thread about this topic if needed.