On Mon, 13 Dec 2004 19:42:09 -0500, Isaac Richards <[EMAIL PROTECTED]> wrote: > Run it in GDB, hit ctl-c to break, and can then get a backtrace as normal.
When debugging mythfrontend I have to tell gdb to handle SIGPIPE and also SIG33 (realtime signal? not listed in kill -l...). Does debugging mythbackend need the same? Is this a stupid question? Also, if anyone wants to examine my perl script for detecting this condition (it's mostly automagic) please let me know. In a nutshell, swatch watches the backend for recording announcements, and when it sees one it launches my script which checks for this condition we're talking about, nuvs not mentioned in the DB, transcoders running against deleted nuvs, nuv.tmp files which have no associated .nuv, and DB entries without nuvs and uses logger to complain. It's not a good script stylistically, but it's nice functionally for what I need. YMMV.
_______________________________________________ mythtv-dev mailing list [EMAIL PROTECTED] http://mythtv.org/cgi-bin/mailman/listinfo/mythtv-dev
