I have been using mythwelcome for waking up and shutting down my machine for a looooong time now. It seems that since I upgraded to 7.4 I have intermittently had times that the box locks up essentially immediately after waking up.
Representative example: yesterday it wakes up at 3:45 pm for daily wakeup (not to record anything). Wife gets home and the mythwelcome screen is up but nothing is responding. If it is the same as what I have seen, there would have been no hard drive activity and totally unresponsive (keyboard, ssh, etc.). The time on mythwelcome would have stopped at the time it locked up (likely 3:45 or 3:46). If I check the database in mythweb no tables are showing crashed or left open, so it does not seem to be accessing mysql at the time... I cannot find anything out of the ordinary in the logs.
This probably happens once every week or two. Normally it's not a big deal, but if we were going to be on vacation for a week or something, it could lock up on day one and just do nothing all week then. I have reseated the northbridge heatsink on my motherboard (for the 2nd time). Sometimes it was causing lockups during normal operation, but since then it has been stable except for the occasional lockup on startup. My system is an ASUS M2NPV-VM motherboard with a dual core processor, NVIDIA 9500GT video card (added since this problem began), NVIDIA drivers 295 (not sure whether this has changed since the problem began). I have something like 2.5 GB RAM, I think I added 1GB after this problem began, but the next thing I may try is removing it just to be sure. When I upgraded to 7.4 I also added an HDHR-Prime if that could somehow be related. I removed a PVR500 as well.
I believe that mythwelcome has an option to run a command on startup, where I am running a script that sets the CPU frequency to ondemand, turns on the UPS APC daemon, the LIRCD daemon for an IR blaster (which I no longer used and commented this line out this morning), writes something to allow RTC to be used for vsync instead of usleep (forget the command), and write something to shut up some complaint about a size for alsa. The RTC and alsa commands are new since 7.4, so this morning I just put some comments in between each of the commands in the script to log them to a file so that I can see if any of them are responsible the next time it happens.
Otherwise, I am curious if anybody else has seen this or has any other suggestions on how to track down the problem. Since it is not a completely random lockup, I feel that I should be capable of tracking it down with enough logging. I would just like to get everything in place now so that I can catch it the next time. As always, any suggestions are much appreciated.
|