[Home]

Summary:ASTERISK-02407: trunk not always used
Reporter:zoa (zoa)Labels:
Date Opened:2004-09-15 03:29:26Date Closed:2011-06-07 14:10:46
Priority:MinorRegression?No
Status:Closed/CompleteComponents:Channels/chan_iax2
Versions:Frequency of
Occurrence
Related
Issues:
Environment:Attachments:
Description:When doing some trunking benchmarks for my astricon tutorial, i noticed that:

1) There should be a warning if you want to do trunking (for each call) when you have no zaptel timing device - otherwise you will have one way trunking.

2) When doing > 20 trunked calls, for some reason 'iax2 trunk debug' gives different results on both ends.




****** ADDITIONAL INFORMATION ******

Processed trunk peer (10.3.3.6:4569) with 24 call(s)
Processed trunk peer (10.5.0.7:4569) with 171 call(s)

I don't know if its just the output, or also the calls itself not being trunked correctly.


Comments:By: zoa (zoa) 2004-09-15 03:29:59

oh, i'm on latest cvs of course. (didnt try any other versions, too much benchmarks to do.)
By: zoa (zoa) 2004-09-15 03:52:00

if you do twice in a row: iax2 trunk debug and now calls are being dropped between those moments, chances are big, the values will be different:

e.g.

halo*CLI> iax2 trunk debug
IAX2 Trunk Debug Requested
Beginning trunk processing
Processed trunk peer (10.5.0.7:4569) with 296 call(s)
Ending trunk processing with 1 peers and 296 calls processed
halo*CLI> iax2 trunk debug
IAX2 Trunk Debug Requested
Beginning trunk processing
Processed trunk peer (10.5.0.7:4569) with 284 call(s)
Ending trunk processing with 1 peers and 284 calls processed

I have a small suspicion this might also cause small audio gaps.
By: Mark Spencer (markster) 2004-09-15 11:43:26

Look on the wire and see the packets, you should be able to determine from that.
By: Mark Spencer (markster) 2004-09-17 00:07:22

Hello, anybody home?
By: zoa (zoa) 2004-09-17 07:15:33

Wow wow, hold on... its on my todo list, i also spoke to steve about it, looks like the iax2 show channels claims them to be trunked.

I already have some measurements for the data itself, but still have to interpret it and don't have the time to do so now.

Will update as soon as i find the time.
(Bug is minor and only when doing a lot of calls.)
By: zoa (zoa) 2004-09-17 07:15:53

if it even is a bug at all...
By: Mark Spencer (markster) 2004-10-01 23:17:33

Did you ever determine what was on th ewire?
By: zoa (zoa) 2004-10-02 07:37:30

From what i see in the graphs, the bandwidth used looks normal, it doesnt seem to be increasing faster than normal with a new call, so i suppose its only the counter doing strange things ?
By: twisted (twisted) 2004-10-27 16:25:17

Is this still an issue you can see with latest code?  

Update?

--Housekeeping
By: Russell Bryant (russell) 2004-11-07 19:49:25.000-0600

I'm going to close this since there doesn't seem to be much interest.