[cisco-voip] vg224 sanity check

[Ed Leatherman]

Some more notes on this one to close it up for posterity.

I asked them to lug a fax machine into their server room and attach it
directly to a FXS port on the PRI gateway (cisco 2900 router iirc). Fax
issue was still present to the PSTN from this location, voice call is OK
still.

We pushed service provider to test out the PRI's. They came in and ran some
tests, no problem found but as soon as they did this faxes magically
started working again. Perhaps just a hardware reset on their side of the
circuit is all I can guess.




On Thu, Jun 13, 2013 at 4:14 PM, Countryman, Edward <
Edward.Countryman at presencehealth.org> wrote:

> We just went through something similar at one of our sites. ****
>
> ** **
>
> This site has a large amount of analog so in order to allow feature access
> through a VG224 the ports have to be configured as SCCP.  What we found the
> hard way is the SCCP and fax machines do not play nice.    ****
>
> ** **
>
> The most reliable way to assure faxing works with Cisco is to use t38
> protocol.  Trying to fax through a G711 connection is a nightmare.  ****
>
> ** **
>
> However, Cisco doesn’t really support T38 on sccp (there are some
> contrived config’s that tell you to use cisco NSE’s etc but we never could
> get that to work.)****
>
> ** **
>
> Ultimately, our solution was to configure the ports used for faxing as
> MGCP.  Also as important, configure the T38 protocol the same on both the
> (mgcp) VG224 and your router that has the PRI on it.  ****
>
> ** **
>
> If done correctly, you will actually see the codec change from G711 once
> the call connects and the fax machines start to negotiate.****
>
> ** **
>
> Good luck****
>
> ** **
>
> *From:* cisco-voip [mailto:cisco-voip-bounces at puck.nether.net] *On Behalf
> Of *Ed Leatherman
> *Sent:* Thursday, June 13, 2013 1:37 PM
>
> *To:* Carlo Calabrese
> *Cc:* Cisco VOIP
> *Subject:* Re: [cisco-voip] vg224 sanity check****
>
> ** **
>
> Thanks for the suggestions everyone!****
>
> ** **
>
> They are able to make voice calls out OK with their fax machines so RTP is
> at least working both ways. ****
>
> ** **
>
> Story changes the more questions I ask... Faxes between any VoIP endpoints
> (VG and ATA) work OK, it seems like just faxes over their PRI's now are the
> problem (but its any faxes, VG or ATA both). Their PSTN gateway is H.323,
> pretty basic config but that's my next point to start looking at. The VG224
> and the PRI gateway are literally right next to each other off the same L3
> switch, no firewalls and ethernet interfaces are error/drop free. I asked
> them to check with the local phone company to check their PRI circuits,
> error counters are relatively clean on their local side though.****
>
> ** **
>
> All this has been working up till a day ago or so, everyone swears to me
> nothing has been changed :)****
>
> ** **
>
> Time to start reading up on my Fax, Modem, and Text for IP Telephony book.
> ****
>
> ** **
>
> On Thu, Jun 13, 2013 at 11:54 AM, Carlo Calabrese <
> carlo_calabrese2006 at yahoo.com> wrote:****
>
> Can you check your mgcp config. Do you have****
>
> Mgcp default-package fxr-package****
>
>  ****
>
> This will cause problems with faxes training up.****
>
>  ****
>
> Carlo****
>
>  ****
>
> *From:* cisco-voip [mailto:cisco-voip-bounces at puck.nether.net] *On Behalf
> Of *Ed Leatherman
> *Sent:* Thursday, June 13, 2013 5:57 AM
> *To:* Cisco VOIP
> *Subject:* [cisco-voip] vg224 sanity check****
>
>  ****
>
> Good morning!****
>
>  ****
>
> I have a remote campus with 1 VG224 (SCCP controlled) and a handful of
> ATA's for analog service to fax machines etc.****
>
>  ****
>
> IT staff on-site is complaining now that faxes on the VG can no longer
> send faxes (they fail usually during the first page), whereas faxes via ATA
> are ok oddly enough.****
>
>  ****
>
> Kicker to me is they've also stated that even faxes between two devices on
> the VG224 are having the same issue.****
>
>  ****
>
> To me this along with the ATA's actually working OK seems to make a PRI or
> network issue less likely although not ruling it out in case they are just
> reporting things wrong to me. They also say no changes were made to gateway
> (which I believe cause they prob don't have the login for it) or on their
> CUCM.****
>
>  ****
>
> My question though for anyone familiar with these is, if two devices on
> the gateway are talking to each other via FXS ports, is this signal just
> getting switched natively on the gateway? Is any signal conversion
> happening or DSP's engaged at this point, assuming there aren't any
> conferencing or such going on?****
>
>  ****
>
> Thanks!****
>
>  ****
>
> --
> Ed Leatherman****
>
>
>
> ****
>
> ** **
>
> --
> Ed Leatherman****
>



-- 
Ed Leatherman
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://puck.nether.net/pipermail/cisco-voip/attachments/20130620/4953183c/attachment.html>