InterLinked1

> This is why I suggested actually digging in more to truly understand the scenario and what is going on to provide a better and more helpful message in the...

cherry-pick-to: 20 cherry-pick-to: 21

> Doesn't cherry pick to 18 so the PR will need to be split. Done, linked PR: https://github.com/asterisk/asterisk/pull/663

cherry-pick-to: 18

> Tests need to be written to exercise the current behavior before this can be merged. Good idea, but how would this work exactly? I don't think the Asterisk test...

> It caused an actual failure in dialplan execution for @creslin287 which is testable through the testsuite, so if that situation can be identified then a test can be written...

> An effort should be put forth to include test coverage for 21 and master. If not possible then an explanation of why not and what would be required to...

> Packet capture? `iax2 set debug on` output? Yeah, I don't know what I was thinking. It was late last night and I didn't think to do so. I've tried...

I think this is already a known issue. I seem to recall seeing it on JIRA, or independently reported by multiple people. > What does the RFC state for multiple...

> current beheavior: > > > As seen on the image is wrong as content type wont change and the other variable is adding to the header. > > I...