http://qs1969.pair.com?node_id=1084153


in reply to Re: How to debug failure in HTTP::Daemon::SSL->new()
in thread How to debug failure in HTTP::Daemon::SSL->new()

I guarantee you something changed.

Doubtless.

Unfortunately, it was several weeks before I noticed that it had stopped working, so tracking down exactly what had changed at that precise moment is not entirely straightforward.

Replies are listed 'Best First'.
Re^2: How to debug failure in HTTP::Daemon::SSL->new()
by Anonymous Monk on Apr 28, 2014 at 19:05 UTC

    Oh. Any luck with the debugger?

    HTTP::Daemon::SSL inherits its new() method pretty much directly from IO::Socket::SSL, so the problem should be coming from there. That module had a release just a few days ago, perhaps you could try pulling the latest version of that and opening a socket with that, maybe that'll get you an improvement, or at least an error message?

      Any luck with the debugger?

      Not as yet. The control flow through IO::Socket::SSL's constructor appears to be rather baroque, and I'm having rather a hard time following what's going on. (I'm not very comfortable with debuggers at the best of times. They never seem to want to tell you what you really need to know. In this case, I probably mostly want what Carp would tell me if called from right before the innermost return, but I haven't managed to successfully identify said innermost return.)

      I am seriously considering switching to a different module. Perhaps I might also contemplate moving from HTTPS to SSH while I'm at it. I have never had a great deal of fondness for HTTPS, particularly in terms of how it handles certificates.