CPU temp above threshold

LIsLinuxIsSogood · Veteran Joined: 13 Feb 2016 Posts: 1179

Turning to the forum for some help with whatever is going on with ACPI and CPU temp (especially!). What is this and should I go about to be fixing it? Do these messages seem at all related?

eccerr0r · Posted: Sat Aug 19, 2017 9:46 pm Post subject:

They look separate from each other, your cpu probably really was overheating if you were running it under heavy load. I get those a lot during the summer, I may also need to really clean my heatsink/fan and get some fresh thermal interface material...
_________________
Intel Core i7 2700K/Radeon R7 250/24GB DDR3/256GB SSD
What am I supposed watching?

russK · l33t Joined: 27 Jun 2006 Posts: 665

I don't know what the ACPI issue is.
The CRC error with the disk is potentailly bad, but the disk may have handled it gracefully. It happened around 8 minutes before the cpu.
It's a good idea to keep heatsinks free of dirt and fans running OK. If machine is in a warm place, help it out.
Heat can effect a hard drive too. You may want to check the hard drive smartctl info. smartctl can also tell you temperature.

https://wiki.gentoo.org/wiki/Smartmontools

Section_8 · l33t Joined: 22 May 2004 Posts: 627

I see those temperature messages sometimes, as my system runs a couple of boinc projects. I wouldn't be surprised to see them if you're compiling some big package.

LIsLinuxIsSogood · Veteran Joined: 13 Feb 2016 Posts: 1179

The disk error is still happening today...could I please get some help with these very foreign looking messages I do not require detailed explanation, but more a simple overview...

Thanks to those with suggestions, like the steps I will be taking to implement SMART, in the meantime what can I do to query the hardware to check it or find out more info. I am still unaware how the device is being referred to in the output of the log as ata1? How does that correspond to /dev/sdx for naming of disks? Does that refer to the entire disk or just a partition of the disk, and where can I find this information?

I can't seem to understand the messages except that "bus error" and "hard resetting link" those don't seem good...I know that much but that's all I know. Help!

russK · l33t Joined: 27 Jun 2006 Posts: 665

I'm no expert but I'll take a stab at some of it ...

The errors are a repeat & similar to the first post.
Both times, the initial messages happened within one second, as in the 2nd episode shown by [68437.nnnnnn].

Within the first 11 microseconds of the first message, the driver decided to reset the link. I suspect the drive was trying in earnest to read some unrecoverable data at 68437.088944, 68437.088946, 68437.088948 and 68437.088950 and the driver was either not prepared to hear the bad news or simply impatient. The driver may have been upset about how long it took for the drive to respond. Some of this is conjecture on my part.

Finally the driver decided to reset the link to get back to a known state. At 68437.434977 the reset was complete and the drive was ready.

I suspect there is some data on the disk that was not recorded properly. Every time the drive tries to read the data, it retries over and over and then finally gives up. During this period, the driver gets a little impatient and resets the link.

smartctl can tell you about the health of the drive.
There are tools and web pages about discovering bad blocks and getting them re-allocated.
ddrescue is a good tool for recovering data from a bad drive.

Your drive may or may not be in need of replacement. I would use smartctl to see if the drive believes it is healthy.

Depending on how many drives you have, something like this might be useful:

russK · l33t Joined: 27 Jun 2006 Posts: 665

Also note, sometimes errors like this can be due to the cables or bad connections.

Small_Penguin · Tux's lil' helper Joined: 27 May 2005 Posts: 140

You can also investigate with sys-apps/gsmartcontrol which has a more user-friendly interface.

bunder · Bodhisattva Joined: 10 Apr 2004 Posts: 5934