On Tue, Apr 14, 2009 at 10:46 AM, Lee Howard <[email protected]> wrote: > David Backeberg wrote: >> It may be possible to use hylafax, but >> I don't know how or why you would. > > The reason *why* is generally due to support issues.
What I was specifically getting at in the context of that response was a comparison of dynamic modem pool versus fixed-size modem pool. When faced with the choice between a fixed-size modem pool or one that would grow or shrink dynamically with demand, I think the dynamic is a better choice. That is my opinion, based on my particular style of usage, and I am certainly grateful for the existence of Hylafax. The other thing I was getting at was complexity of a multi-layer software stack in loopback mode when it comes to trying to figure out where the problem could be. One of my suggestions was to separate the sender and receiver to separate machines, and my other suggestion was to reduce the layers of software involved. As you know from having written Hylafax, it's designed for hardware modems and does things like letting you take your fax device online/offline, as well as taking it down for five seconds post-fax to do a reset on the line. These things are all documented, they all have debugging tools, and ways to view modem state. But for a beginning person trying to set up a fax for the first time I feel like it adds complexity to what can already be a complex task. Thus I suggested trying without. As I've been debugging faxes it seems more and more like fax failures are due to poorly-thought-through enterprise voip implementations, or intermittent line noise, or poor implementations of fax protocols in proprietary fax machine hardware. I think the Hylafax versus app_fax thing is really six of one versus a half-dozen of the other, but both are useful data points in debugging a failing fax configuration. _______________________________________________ -- Bandwidth and Colocation Provided by http://www.api-digital.com -- asterisk-users mailing list To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
