Ever since my last 2 SVN upgrades, auto-expire has not been working, and my disk has been filling up regularly. My last good working was from 8/13 (sorry I don't have an SVN revision; my build system is still set up to use date-name directories, leftover from the CVS days). In my backend logs, I see this any time it tries to auto-expire a program:
2005-10-06 01:17:19.935 AutoExpiring: Blue Collar TV Tue Oct 4 02:00:00 2005 865 MBytes 2005-10-06 01:17:19.937 AutoExpiring: The Bellboy Tue Oct 4 17:45:00 2005 1974 MBytes Cannot find program info for 'AUTO_EXPIRE 1026 2005-10-04T02:00:00', while attempting to Auto-Expire. Cannot find program info for 'AUTO_EXPIRE 1071 2005-10-04T17:45:00', while attempting to Auto-Expire. Also, I know the backend is supposed to check for a certain (configurable) amount of free space before starting a recording... does the same apply to the transcoder? If not, I had a thought... perhaps that configurable amount of space should be considered "per simultaneous job"? In other words, let's say I set my threshold at a minimum of 8 GB free... that works fine as long as I only have 1 tuner recording. If there's a second tuner with a simultaneous recording, the backend needs to take that into account. The same for a transcode (needs space for the .tmp file until it's done). Of course, these situations wouldn't normally present a problem if auto-expire was working properly, but there's always the possibility that the disk is full of recordings set to *not* auto-expire. The workaround, of course, is to set the 'new recording' disk space threshold to account for the maximum space that could be consumed by potential simultaneous jobs. Perhaps that could go into the docs somewhere... Anyway, if anyone has any thoughts on why my auto-expire is broken, I'd appreciate it. Oh, and BTW, normal delete via the frontend or mythweb works just fine. TIA, JAC
_______________________________________________ mythtv-users mailing list [email protected] http://mythtv.org/cgi-bin/mailman/listinfo/mythtv-users
