After reading your message, I turned logging to "debug", and T.38 on, and sent the FAX again to reproduce the problem.

This time it worked. Sorry.

I did get some white lines and got this in the log:

c5c78d1f-8b8a-48a6-a2a3-a94fcaffde47 2016-05-17 12:21:50.708159 [DEBUG] mod_spandsp_fax.c:435 Bad rows = 55

c5c78d1f-8b8a-48a6-a2a3-a94fcaffde47 2016-05-17 12:21:50.708159 [DEBUG] mod_spandsp_fax.c:436 Longest bad row run = 4

I don't think there's enough info there for debugging.

Next, I'm going to try hooking the FAX machine to the POTS line to see if I can reproduce the problem that way.

Steve

On Tue, May 17, 2016 at 11:41 AM, Michael Jerris <mike@jerris.com> wrote:

This has turned into a giant bug report. Bug reports go in jira. Putting details about bug reports here will absolutely guarantee that they will not get looked at. As i've already said once this week on the list about this issue, we need someone who can map out how to reproduce this issue and get it onto jira so we can track down the issue. Without that, without a doubt, it will not get fixed.

As for downgrading to 1.4 to fix the issue, I understand, but know that we are not putting other bugfixes into 1.4 anymore and eventually you will hit a bug you need to upgrade for, and you will be left without working faxes if we don't get this fixed. Please step up and help us track this down so we can get it fixed properly.

Thanks
Mike