<br><br><div class="gmail_quote">On Wed, Jun 24, 2009 at 10:02 AM, Paul Gallaway <span dir="ltr"><<a href="mailto:pgallaway@gmail.com">pgallaway@gmail.com</a>></span> wrote:<br><blockquote class="gmail_quote" style="border-left: 1px solid rgb(204, 204, 204); margin: 0pt 0pt 0pt 0.8ex; padding-left: 1ex;">
<div class="im">> Any suggestions for increased log levels that might shed light on this?<br>
<br>
</div>Don't have a 2250 but have some suggestions.<br>
<br>
First, make sure you're not running the frontend or backend as root.</blockquote><div><snip><br><br>nope, I double checked and both BE&FE are running as mythtv <br></div><div> </div><blockquote class="gmail_quote" style="border-left: 1px solid rgb(204, 204, 204); margin: 0pt 0pt 0pt 0.8ex; padding-left: 1ex;">
Installing from source </blockquote><div><br>I try to limit this and stick with debian packages otherwise I forget what I did (and how) and things get really ugly...<br></div><div> </div><blockquote class="gmail_quote" style="border-left: 1px solid rgb(204, 204, 204); margin: 0pt 0pt 0pt 0.8ex; padding-left: 1ex;">
<snip></blockquote><div> </div><blockquote class="gmail_quote" style="border-left: 1px solid rgb(204, 204, 204); margin: 0pt 0pt 0pt 0.8ex; padding-left: 1ex;">A hard lock indicates that the problem exists outside<br>
of MythTV so I'd suggest that MythTV is not directly the problem -<br>
although perhaps indirectly. Perhaps the new 2250 drivers require some<br>
additional code put into Myth in the past year plus to properly<br>
communicate with the driver. The driver, if in kernel space, could<br>
bring down the system; just a theory. At the very least look at<br>
getting a newer build from a repository.</blockquote><div><br>agreed, but I was hoping that others were using HVR-2250 with myth and could say if they'd seen and/or fixed similar issues.<br><br>I plan on keeping this current version until I see another hang (then I'll grab tip again). So far I've been up for 3 days through many recordings so it might be the hangs were fixed by some other fiddling I've been doing. See bottom.<br>
<br></div><blockquote class="gmail_quote" style="border-left: 1px solid rgb(204, 204, 204); margin: 0pt 0pt 0pt 0.8ex; padding-left: 1ex;"> The debian-multimedia testing<br>
repo has a build from April '09 </blockquote><div><br>I had a thread a while ago about updating to a newer 0.21-fixes (<a href="http://www.gossamer-threads.com/lists/mythtv/users/384013#384013">http://www.gossamer-threads.com/lists/mythtv/users/384013#384013</a>) and I decided it wasn't worth while, but if these hangs continue I'll revisit that also (after trying newer driver).<br>
<br></div><blockquote class="gmail_quote" style="border-left: 1px solid rgb(204, 204, 204); margin: 0pt 0pt 0pt 0.8ex; padding-left: 1ex;">
My experience: I had similar hard locks with Lenny (but no<br>
HVR-1600/2250 here) and a new hardware build on AMD64 although my<br>
crashes were daily. I tried everything I could think of. My problem<br>
went away when I converted my XFS file systems to EXT4. </blockquote><div><br>I'm using EXT3 for everything with the "delete slowly" flag.<br> <br></div><blockquote class="gmail_quote" style="border-left: 1px solid rgb(204, 204, 204); margin: 0pt 0pt 0pt 0.8ex; padding-left: 1ex;">
My system has<br>
been running without crash for >two weeks now. As a result of my debug<br>
attempts and subsequent upgrades in pursuit of a solution, I am<br>
currently on Squeeze with the 2.6.29 kernel but the problem first<br>
manifested with 2.6.26 and Lenny. I used XFS without problem for a<br>
year or so in Etch and/or Lenny and i386 2.6.18 kernel. Not implying<br>
this is your problem but merely giving you other avenues to<br>
investigate. The second link with the response to my original post<br>
gives some sage advice with respect to hardware troubleshooting.<br>
<br>
<a href="http://www.gossamer-threads.com/lists/mythtv/users/384525#384525" target="_blank">http://www.gossamer-threads.com/lists/mythtv/users/384525#384525</a><br>
<a href="http://www.gossamer-threads.com/lists/mythtv/users/384538#384538" target="_blank">http://www.gossamer-threads.com/lists/mythtv/users/384538#384538</a><br>
</blockquote><div> <br>Thanks for info/suggestions Paul. At the moment I'm more or less in a holding pattern waiting to see if it continues to hang.<br><br>The things I've changed recently that probably don't but might have affected these hangs:<br>
<br>1) added labels to my partitions and changed fstab to use the labels instead of /dev/sd* - This had the surprising side effect that nmbd will now startup properly. Previously smbd would start on reboot but nmbd would not start automatically even though /etc/init.d/samba restart would get both smbd/nmbd up and running. This one has me stumped but maybe it changed the order that partitions are mounted in such a way that the shares weren't mounted when nmbd tried to run before I added the labels.<br>
2) I purged horde3 because it was dumping tons of stuff into my system logs. I wanted to try IMP but gave up trying to configure it since it was so complicated and all I wanted was nicer looking webmail than squirrelmail.<br>
3) I unplugged my USB flash card reader because it too was dumping tons of junk into my system logs (~150 lines per minute in /var/log/messages!). I don't expect to use it often and when I tried it the first time it didn't work OOTB so I unplugged it until I have time to figure out why it's so "chatty".<br>
4) I changed from Normal playback profile to using OpenGL on my GeForce 9300 because I thought it might improve the jaggies I saw on recordings. They were actually caused by bad SNR so I changed back to Normal last weekend in case OpenGL was contributing to my hangs.<br>
5) Plugged in my custom made internal SPDIF header to RCA jack cable so I could use digital audio out from my myth FE instead of the headphone jack. (I should have bought a mobo with digital audio on the back panel!)<br>
<br>That's all I can remember at the moment, and yes I know most of this stuff should have no impact on hard hangs but ... :)<br></div></div>