Hi, this is the second time this happens for me. First time with vdr-1.3.22 + xine-network-0.7.2, now with vdr-1.3.23 + xine-network-0.7.3. What happens is that VDR gets completely stuck. Restarting xine does not bring picture back, and what's worst, VDR does not process timers (does not do recordings).
All I could find on syslog was that at some point something (=xine?) got stuck and started producing ring buffer overflows (lots and lots of those). Then when it was time for first timer, this is what I could find:
May 11 10:58:49 mewtwovdr vdr[22400]: ERROR: 10860 ring buffer overflows (2041680 bytes dropped) May 11 10:58:55 mewtwovdr vdr[22400]: ERROR: 13267 ring buffer overflows (2494196 bytes dropped) May 11 10:59:03 mewtwovdr vdr[20211]: ERROR: thread 5914642 won't end (waited 3 seconds) - canceling it...
None of the timers were recording until I killed vdr (killall -9 vdr), killing/reconnecting xine did not produce any kind of picture. Event status changes were still processed, but nothing else (like pid-changes etc).
Any ideas what could be the problem? I don't care if my picture gets stuck from time to time, but I wouldn't want to *ever* again see any of my timers failing...
I am afraid but I don't think I'm able to reproduce the error (e.g. suspending xine does not make vdr to get stuck).
Teemu
Hi,
Rantanen Teemu wrote:
Hi, this is the second time this happens for me. First time with vdr-1.3.22 + xine-network-0.7.2, now with vdr-1.3.23 + xine-network-0.7.3. What happens is that VDR gets completely stuck. Restarting xine does not bring picture back, and what's worst, VDR does not process timers (does not do recordings).
All I could find on syslog was that at some point something (=xine?) got stuck and started producing ring buffer overflows (lots and lots of those). Then when it was time for first timer, this is what I could find:
May 11 10:58:49 mewtwovdr vdr[22400]: ERROR: 10860 ring buffer overflows (2041680 bytes dropped) May 11 10:58:55 mewtwovdr vdr[22400]: ERROR: 13267 ring buffer overflows (2494196 bytes dropped) May 11 10:59:03 mewtwovdr vdr[20211]: ERROR: thread 5914642 won't end (waited 3 seconds) - canceling it...
None of the timers were recording until I killed vdr (killall -9 vdr), killing/reconnecting xine did not produce any kind of picture. Event status changes were still processed, but nothing else (like pid-changes etc).
Any ideas what could be the problem? I don't care if my picture gets stuck from time to time, but I wouldn't want to *ever* again see any of my timers failing...
I am afraid but I don't think I'm able to reproduce the error (e.g. suspending xine does not make vdr to get stuck).
Well, before 0.7.3 something like that could happen at a PTS wrap (happens about every 26.5 hours).
For any error report, please include the output drivers used (-V, -A) and the tail of xine's "--verbose=2" output.
Bye.