[Asterisk-Users] Line Noise HELP!

[Damian Funnell]

Hi,

For those that were having the same line noise problem that we were, an 
update:

    * Our TDM400P *was* sharing an IRQ, despite the output from 'cat
      /proc/interrupts' showing that it wasn't.  Running 'lspci -v'
      showed that it was and we had to perform some card juggling to get
      it (and our ISDN card) to sit on different IRQ's.
    * It appears that this change has resulted in a better *average*
      accuracy via zttest (although as zttest does not provide an
      'average' figure, this is anecdotal only).  Certainly the 'best'
      and 'worst' output is no different (best = 99.975586%, worst =
      99.963379%).
    * The users reported a significant improvement on the last full day
      of use - they experienced a couple of 'spikes' of noise, but no
      incidences of the prolonged noise that they were experiencing
      earlier.  This was only over one day, though, so not sure if this
      means the problems licked or if Friday was just a good day.
    * I've asked Digium for further advice, as they have told me that we
      should try and get no less than 99.98% accuracy via zttest.
    * They've recommended disabling hyper threading and we are yet to
      give this a go.  Will be pretty annoyed if we have to leave hyper
      threading turned off in order to get this to work, as performance
      is sure to suffer.


Keen to hear if anyone else has managed to get anywhere with their noise 
problem.  Not sure if we're on the right track or not, but will report 
back in a couple of days to see if the intermittent crackling has returned.


Rich Adamson wrote:

>>Excerpt of email from Digium support:
>>
>>Digium does not support CAPI or BRI, either freely or commercially.
>>
>>We can only provide support for you Digium TDM400P card.
>>
>>Your zttest output is extemely low.  We are looking for nothing less
>>than 99.98%.  If you are receiving 99.96% this will cause major
>>problems.  Have a best output of 99.975% is really low.
>>    
>>
>
>For the record, I ran zttest against a new TDM04b Rev H board and it
>consistently reported 99.975586%. There is no noise, crackling, etc.
>So, not sure what "99.975% is really low" is based on.
>
>  
>
>>If you are running an IDE hard drive please verify that you are using
>>DMA mode with a UDMA setting of no lower than 2 or higher than 3.  UDMA
>>mode 2 is ATA33.  UDMA mode 3 is ATA44.  This can be done using hdparm.
>> We suggest using "hdparm -d 1 -X udma2 -c 3 /dev/[IDE Device]".  You
>>can check the status using "hdparm /dev/[IDE Device]" and "hdparm -i
>>/dev/[IDE Device]".  If you make modifications to your IDE hard drive
>>settings they will only be kept until you reboot.  
>>    
>>
>
>Also, did the above, which had zero impact on the zttest results.
>
>If I try to use spandsp-pre11 for fax reception, it results in far less 
>then usable output (*.tiff), and supposedly that is due to missed 
>frames, missed interrupts, or something like that with the TDM04b 
>card.
>
>Checked and double-checked share interrupts, and that isn't a problem.
>
>This is on a cvs-head RHv9 box with 2.2ghz processor, and nothing else
>running on the system.
>
>So, best guess is that 99.975586% is impacting fax but not voice.
>
>
>_______________________________________________
>Asterisk-Users mailing list
>Asterisk-Users at lists.digium.com
>http://lists.digium.com/mailman/listinfo/asterisk-users
>To UNSUBSCRIBE or update options visit:
>   http://lists.digium.com/mailman/listinfo/asterisk-users
>
>
>
>  
>