Quote from: candleking on January 10, 2018, 10:11:51 PM

I'm using 5x 1060s 6Gb, Claymore 10.2 and get the "temperature control thread hangs need to restart miner" stuff. Driver version 360.95. Anyone else has this issue and know why? Temps on cards are between 53-58 ^oC.

Check first post of on this page.. you can try -tt 0 as he said to test it. For me, I went back to 10.0.

Quote from: tstempone on January 10, 2018, 03:56:36 PM

Quote from: Claymore on January 10, 2018, 03:44:23 PM

Quote from: tstempone on January 10, 2018, 03:19:59 PM

Quote from: Claymore on January 10, 2018, 02:56:20 PM

Quote from: jackbox on January 10, 2018, 02:32:58 PM

Quote from: mu_enrico on January 10, 2018, 02:29:52 PM

Quote from: nico610 on January 10, 2018, 01:27:43 PM

Can you help me for this error : Temperature control thread hangs, need to restart miner!

My card has temps 62, the miner restart every 2 hours or less, every 20mn this morning

I have this error on claymore 10.2 and 10.3, i don't use previous version.
I haven't this problem with claymore for cryptonight or equihash

I also experienced same issue,
Any idea how to disable the temperature control thread? Thanks

You cannot. It is a driver issue.

I agree, try different drivers if you are sure that hardware is ok. I see this issue on latest Adrenalin drivers on some cards, for example, I cannot make it work stable on a system with stock 1x RX480 + 1x Duo Pro on 17.12.2 drivers, looks like latest drivers are unstable in some configurations.
Or you can specify "-tt 0" to disable temperature/fan control, but probably it will hang in OpenCL call in a few minutes anyway.

I don't think it driver issue, it might compatibility issue with new version of claymore and blockchain driver.

I use blockchain driver and having this peoblem with 10.2 and 10.3 after I went back to 10.0 never have this temperature thread hang again. (all same setting only change 10.3 to 10.0, everything stable for 24 hours, 10.3 give me hangs after 30-60 mins)

Check History.txt:

v10.2
...
- in rare cases ADL API calls can hang, now watchdog checks it as well.
...

So now if ADL hangs, miner will be restarted. In v10.1 and earlier miner will work without temperature control (all related options will stop work too), watchdog will hang too (because it was in same thread). So you can select what behaviour is the best for you: restart miner when ADL hangs (use v10.3) or stop temperature control and related options (use v10.1), but the reason is not in miner anyway.

Thank you for clarify, may I ask if I want 10.3 to operate as 10.0 which command I have to add?