OT – Is T.38 Possible On SPA8800 FXO Port ?
Hello,
If I’m not mistaken, it is not possible to get T.38 on a SPA3102 FXO
port (it is possible with the FXS port). Do you know, by experience preferably, if this is possible with an SPA8800 FXO port ?
Regards
8 thoughts on - OT – Is T.38 Possible On SPA8800 FXO Port ?
Perhaps page 50 (Using a Fax Machine) of this document http://www.cisco.com/c/dam/en/us/td/docs/voice_ip_comm/csbpvga/ata/administration/guide/ATA_AG_v3_NC-WEB.pdf will answer your question.
There is also another document you may want to reference.
https://supportforums.cisco.com/sites/default/files/legacy/8/1/0/42018-SPA8800_Asterisk_101909.pdf
In your sip.conf configuration you will need;
canreinvite=no
For newer version of Asterisk I expect you would set;
directmedia=no
You may want to change the SPA8800 option ‘FAX T38 Redundancy’ to the value you have set for redundancy in udptl.conf or 3 if using the default settings.
It’s been a while since I played with my SPA8800, with the test I
performed it did work, that was inbound and outbound calls.
Larry.
Reading your question again, no the FXO port will not perform T.38, you are limited to G711 pass-through, from memory my SPA8800 was on the LAN
the Asterisk PBX was attached and faxes were pretty good though I would have preferred to see T.38.
Grandstream ATA’s support T.38 on the FXO port.
Larry.
2014-11-17 12:55 GMT+01:00 Larry Moore:
Thanks for sharing this here:
while it’s not always easy to discover what a product can do, it’s harder to know what it CANNOT do.
Thanks
I had a lot of trouble to localize SPA3102 settings for Caller ID
presentation, for instance.
How does Grandstream localization works ?
Do you also have hundreds of parameters for this or do you simply select a Country within a pre-configured list (like Patton smartnodes)
?
T.38 is a relay protocol for transmitting Group 3 facsimiles over a packet network. It has nothing to do with the analog FXS or FXO signaling, but it should not be too difficult to use Asterisk as a gateway. I never needed to do this myself, so I can’t tell whether there are any subtle points.
jg
I was fortunate that there is plenty of information for configuring an SPA3000 & SPA3102 for my country.
Perhaps page 53 of this document http://www.cisco.com/c/dam/en/us/td/docs/voice_ip_comm/csbpvga/ata/administration/guide/ATA_AG_v3_NC-WEB.pdf may help, the other parameter I needed to adjust to receive Caller ID
was the ‘PSTN Answer Delay’ setting.
I can’t recall any great pain setting up an HT-503, I suspect I based settings upon an SPA3102/SPA8800, the T.38 faxing on the FXO was successful after a Firmware Upgrade, it was a couple of years ago when I
was setting this up.
Larry.
2014-11-17 12:16 GMT+01:00 jg:
I know all this.
My question came from the fact that as strange as it may seem, SPA3102
and similar products do not offer the SIP features depending on terminating/originating port. More precisely, when a SIP fax call comes in through an FXS port, it triggers T.38 while it doesn’t trigger T.38 when an FXO port is used instead.
Yes, sure.
So, if i understand the question correctly, you have:
Asterisk SIP-> SPA3102 FXS -> Analog fax machine and the PBX to SPA3102 communications are T.38 before converting to analog to go to the fax machine. Then in the other situation you would have:
Analog line -> SPA3102 FXO port -> Asterisk SIP
and the SPA3102 to Asterisk communication isn’t doing T.38?
If that is the case, the only thing I can think of is that maybe they were not thinking that many people would want to do the second situation with a low end device? I imagine the main use case it to keep the old analog device around after switching to SIP delivery. Probably didn’t expect to see analog delivery that gets translated to a sip fax endpoint.
Though I do agree that once you have the transcoding option, I would think it would be trivial to apply it to both ports.
2014-11-18 15:19 GMT+01:00 Kevin Larsen:
I fully agree.
Maybe a simple 3102 has memory limitations, but I would expect an
8-ports device (SPA8800) to alsa support T.38 with communications passing through FXO ports. Thanks to Larry’s testimony, I’m aware this is not the case.