| LinHES Forums http://forum.linhes.org/ |
|
| Stabilityproblems with my mythtv installaton http://forum.linhes.org/viewtopic.php?f=2&t=8961 |
Page 1 of 1 |
| Author: | marc.aronson [ Mon Mar 13, 2006 10:10 pm ] |
| Post subject: | Stabilityproblems with my mythtv installaton |
I recently upgraded to R5A30.2, moved to new hardware and added a hidef tuner (DCT-6200) to my myth setup and I am having some stability problems. Periodically I find that the entire system is hung. I can't even remote log in via "putty" and I am forced to hit the reset button. After a recent hang I looked in /var/log/messages I saw the following: Quote: Mar 12 11:50:49 mythhd -- MARK -- Mar 12 12:10:49 mythhd -- MARK -- Mar 12 12:30:49 mythhd -- MARK -- Mar 12 15:42:53 mythhd syslogd 1.4.1#16: restart. Mar 12 15:42:53 mythhd kernel: klogd 1.4.1#16, log source = /proc/kmsg started. From this I conclude that the system died somewhere around 12:30; I believe the entry at 15:42 was a result of my resetting the system. After doing some more grepping of various logs I have found the following error messages: Quote: kern.log:Mar 12 21:56:46 mythhd kernel: APIC error on CPU0: 40(40)
Mar 13 19:53:29 mythhd kernel: hdb: drive_cmd: status=0x51 { DriveReady SeekComplete Error } Mar 13 19:53:29 mythhd kernel: hdb: drive_cmd: error=0x04 { DriveStatusError } Note: I mis-copied the "drive_cmd" messages in my original posting -- I've updated them to reflect the two different messages I am getting. The APIC error is repeated many times. The drive_cmd error messages each occur only once/boot. Does anyone have an experience with these error messages? Do the APIC or "drive_cmd" error messages indicate a root cause for the hang? I ran a quick test of both hard drives (hda & hdb), but it came up clean. Thanks for any help you can provided. Marc |
|
| Author: | marc.aronson [ Tue Mar 14, 2006 12:43 am ] |
| Post subject: | |
A brief update. I've made two changes to my configuration and the "drive_cmd" message no longer happen during boot: 1. I moved the "hdb" drive so so that it is now "hdd". So my configuration is "hda = boot drive; hdb = unused; hdc = DVD burner; hdd = second drive". 2. I changed my hdparm commands in bootmisc.sh from Code: /sbin/hdparm -c1 -d1 -u1 /dev/hda /sbin/hdparm -c1 -d1 -u1 /dev/hdd to Code: /sbin/hdparm -c1 -d1 -u0 /dev/hda
/sbin/hdparm -c1 -d1 -u0 /dev/hdd ie: I now set the "interup umask" flag to 0. The drive_cmd error messages are no longer being produced on the boot. I will need to run for a few days to see if this has any impact on my stability. Marc |
|
| Author: | marc.aronson [ Sat Mar 18, 2006 1:01 am ] |
| Post subject: | |
Posting the solution in case anyone has the same problem and finds this thread on a search. I did the following: 1, Edited "/etc/lilo.conf" and added "noapic" to the "append" line, 2. Ran "lilo -v" 3. Rebooted. The APIC errors are gone and hyper-threading still works... Marc |
|
| Page 1 of 1 | All times are UTC - 6 hours |
| Powered by phpBB® Forum Software © phpBB Group http://www.phpbb.com/ |
|