Published February 15th, 2004 by Jim O'Halloran
hda: lost interrupt
I’ve been building up a new machine to run jimohalloran.com recently. Well, its not really a new machine, its actually an ancient Pentium 200 box I’ve had for a long time. Anyway, I tried both Debian 3.0 and Fedora Core 1 (eventually settled on Fedora) on the machine, and both had the same problem. Under moderate disk load, both distro’s would hang, and start thowing “hda: lost interrupt” messages onto the console. After a few minutes both tended to come good again.
A bit of googling turned up an email on the Linux Kernel mailing list which suggested that the problem
might be caused by a Seagate ST434* hard disk.
We had same problems with two absolutely different machines, both equipped with a single Seagate ST34311A drive. Different mainboards, different kernels (2.2 series), different eth hardware (RealTek and 3c905B), both non-SMP.
I was running a 2.4 series kernel, but a “dmesg | grep hda” showed I was running a Seagate drive. Another posting also hinted at DMA problems, and I was getting a DMA timeout error message before the lost interrupts. The poster of that message suggested “hdparm -d 0 /dev/hda; while : ; do hdparm -t /dev/hda ; done” which would first disable DMA, then continually exercise the drive. That seemed to do the trick, so I’ve now added “hdparm -d 0 /dev/hda” to /etc/rc.local so that its always run on startup.
Jeremy Zawodny Says
And performance probably sucks that way too…
Feb 16th, 2004 at 4:03 am
[b] Says
I had this same problem just before my powersupply died on a redhat 9 2.4.24-6 kernel. As soon as i replaced my powersupply it stopped reporting this error, so this error may be caused by insufficient power!
Not sure if that helps, but it DEFENATELY fixed my problem =)
Nov 5th, 2004 at 4:35 am
Frick 'n' Frack Says
When I try to install Redhat Linux 6.2, The computer seems to stall on these packages:
ImageMagick, and glibc 2.1.3-15. I can find ImageMagick and uncheck it in custom install, but I cannot find glibc 2.1.3-15. Plus, when I do a control+alt+F5, processing halts, though the cursor is still blinking, and with control+alt+F4 there is the message; “Lost Interrupts”. The interrupt message repeats over and over again with no apparent end in sight. You got any suggestions for fixing this problem?
Nov 24th, 2004 at 3:51 pm
paha Says
I had the same problem with my Debian and Seagate ST340014A. But I replaced the IDE-cables with brand new 80-wire cables, and everything went OK.
Dec 4th, 2004 at 8:22 pm
gabriel Says
fucked up cable here too! thanks paha! that was driving me insane. Changed PATA cable, 100% recovery.
PS: never trust ASUS cables.
Nov 4th, 2006 at 9:53 pm
peter Says
hi folks,
I had same issues with the “lost interrupt”. The solution was the right cable PATA, for DMA 133 transfer.
Thanks you.
Aug 3rd, 2007 at 7:07 am