Overheating northbridge?

nouse66 · 04-22-2006, 02:19 AM

I have an ABIT KV7 motherboard and I'm running Kubuntu on a SATA drive. I noticed that my disk performance was slow this past week until eventually last night all of a sudden my entire filesystem was read-only.

I opened the case and found my tiny northbridge fan had died. I replaced it with a huge heatsink today and although my system is running longer, eventually the filesystem slowed to a crawl and then went to read-only mode. The northbride heatsink is barely warm to the touch though.

Could the northbride chip really be this heat sensitve or is it more likely that the chipset is dying? or maybe my sata drive is going bad and the northbride fan problem was just a coincidence?

I have Win XP installed on another drive but it's p-ata so if it doesn't crash I still dont know what that means since it's using a different part of the motherboard.

thanks for any help!

Electro · 04-22-2006, 02:41 AM

Sounds like the SATA hard drive is going. Did you check the logs.

I hope you replace the northbridge heatsink with a passive heatsink because using a fan to cool off the northbridge heatsink does not work for long.

The Northbridge chip handles the memory and AGP slot but not the SATA. Though it does interconnect to the Southbridge chip which handles the SATA, PATA, floppy, PCI, ISA.

nouse66 · 04-22-2006, 11:17 AM

thanks. yes, i replaced it with a big passive heatsink to avoid future failures like that.
i left my system booted into windows last night defragging the ata133 drive and it had no trouble. i'll boot into linux again and check for clues in the logs.

nouse66 · 04-22-2006, 12:15 PM

from /var/log/messages

Quote:

Apr 22 09:47:48 localhost kernel: [4294715.904000] ata1: status=0x35 { DeviceFault SeekComplete CorrectedError Error }
Apr 22 09:47:48 localhost kernel: [4294715.904000] SCSI error : <0 0 0 0> return code = 0x8000002
Apr 22 09:47:48 localhost kernel: [4294715.904000] sda: Current: sense key: Hardware Error
Apr 22 09:47:48 localhost kernel: [4294715.904000] Additional sense: No additional sense information
Apr 22 09:47:48 localhost kernel: [4294715.904000] end_request: I/O error, dev sda, sector 48234703
Apr 22 09:47:48 localhost kernel: [4294715.904000] lost page write due to I/O error on sda1
Apr 22 09:48:05 localhost kernel: [4294733.086000] ata1: status=0x35 { DeviceFault SeekComplete CorrectedError Error }
Apr 22 09:48:05 localhost kernel: [4294733.087000] SCSI error : <0 0 0 0> return code = 0x8000002
Apr 22 09:48:05 localhost kernel: [4294733.087000] sda: Current: sense key: Hardware Error
Apr 22 09:48:05 localhost kernel: [4294733.087000] Additional sense: No additional sense information
Apr 22 09:48:05 localhost kernel: [4294733.087000] end_request: I/O error, dev sda, sector 52730138
Apr 22 09:48:06 localhost kernel: [4294734.120000] ata1: status=0x35 { DeviceFault SeekComplete CorrectedError Error }
Apr 22 09:48:06 localhost kernel: [4294734.120000] SCSI error : <0 0 0 0> return code = 0x8000002
Apr 22 09:48:06 localhost kernel: [4294734.120000] sda: Current: sense key: Hardware Error
Apr 22 09:48:06 localhost kernel: [4294734.120000] Additional sense: No additional sense information
Apr 22 09:48:06 localhost kernel: [4294734.120000] end_request: I/O error, dev sda, sector 52730146
Apr 22 09:48:08 localhost kernel: [4294735.470000] ata1: status=0x35 { DeviceFault SeekComplete CorrectedError Error }
Apr 22 09:48:08 localhost kernel: [4294735.470000] SCSI error : <0 0 0 0> return code = 0x8000002
Apr 22 09:48:08 localhost kernel: [4294735.470000] sda: Current: sense key: Hardware Error
Apr 22 09:48:08 localhost kernel: [4294735.470000] Additional sense: No additional sense information
Apr 22 09:48:08 localhost kernel: [4294735.470000] end_request: I/O error, dev sda, sector 52730154
Apr 22 09:48:09 localhost kernel: [4294736.470000] ata1: status=0x35 { DeviceFault SeekComplete CorrectedError Error }
Apr 22 09:48:09 localhost kernel: [4294736.470000] SCSI error : <0 0 0 0> return code = 0x8000002
Apr 22 09:48:09 localhost kernel: [4294736.470000] sda: Current: sense key: Hardware Error
Apr 22 09:48:09 localhost kernel: [4294736.470000] Additional sense: No additional sense information
Apr 22 09:48:09 localhost kernel: [4294736.470000] end_request: I/O error, dev sda, sector 52730162
Apr 22 09:48:10 localhost kernel: [4294737.561000] ata1: status=0x35 { DeviceFault SeekComplete CorrectedError Error }
Apr 22 09:48:10 localhost kernel: [4294737.561000] SCSI error : <0 0 0 0> return code = 0x8000002
Apr 22 09:48:10 localhost kernel: [4294737.561000] sda: Current: sense key: Hardware Error
Apr 22 09:48:10 localhost kernel: [4294737.561000] Additional sense: No additional sense information
Apr 22 09:48:10 localhost kernel: [4294737.561000] end_request: I/O error, dev sda, sector 52730170
Apr 22 09:48:10 localhost kernel: [4294737.835000] __journal_remove_journal_head: freeing b_committed_data

sure doesn't sound good...

Bobymc · 04-23-2006, 12:19 PM

JUST A CLUE

I got the same problem few times with Sata,I tought my system is going down. Well I just give a touch on SATA DATA cable ;seems like a bad interconnect. I plug and unplug the cable then the system going smoothly. SATA cable only stick to the male interconnect not like IDE grips the pin. Make sure its well conected. Norton GHOST sometimes works with UNIX, you can ghost your HDD data and saved for back up.

CU Folks

nouse66 · 04-23-2006, 08:13 PM

I ended up finding a good deal on a new sata drive so I replaced the old one and everything is running great now. I'm guessing that drive just was on it's last legs.

hw-tph · 04-26-2006, 08:27 AM

A hint: Use Smartmontools to monitor your disks for signs of problems. I have replaced two disks at home before they went bad with the help of this.

There is a Linux Journal article on smartmontools to get you started.

Håkan

nouse66 · 04-26-2006, 09:29 AM

thanks, i'll check that out. i installed hddtemp also to check for heat issues.