Bitcoin Forum

As per subject, the longest of any of my miners are 1 week. I find it a rather almost full time job to have to run to my rig to hard reset them.

Any suggestion to remote resetting? I am exploring diskless system and it seems to be the most practical method.

just add reboot.bat to execute each 4 days or something reasonable. And add the miner.bat in the startup so that it starts automatically as your PC starts.

Probably around 2, maybe 3 months for most rigs with Win7 with a few tweaks and most scheduled tasks disabled so there's nothing going on in the background that could degrade it over time.

But then again 2-3 months without a short power outage is super rare. And some rigs are less stable whatever I do and only usually go for a couple of weeks tops.

For most rigs that won't go long it's almost always caused by poor USB powered risers. 3 parts per risers, it's a pain in the ass trying to find out which part (out of 18 for a 6 card rig) is at fault.

Mine is at 28 days now.
http://1cf13d.ethosdistro.com/#

It's only 28 days because I upgraded ethOS and that required a reboot. Prior to that it had been several months. The biggest issue for me is the occasional loss of power. It is now cold (relative to "normal" weather here) where I live so I suspect that the next unplanned reboot will be due to loss of power due to an ice storm.

Months at a time, 20+ Windows 10 rigs with 6 or 7 GPUs each. Only reason they go offline is to get blasted with compressed air, or power outage.

well with zec i had to upgrade claymore many times.  but if I leave the rigs alone 3 weeks is pretty easy sometimes 5 or 6.

most issues are psu lacks the balls it needs.

card gets too hot.

riser gets too warm.

How can it be a full time job if all you take is few minute to hard reset each rig (well unless you are talking about 10000 rig)? There are actually a lot of way to remotely reset but each come with problem too.

I myself tried wireless remote power socket but no luck, it work at first but might be due to the high heat from rig it spoil rather quick.

Max is probably a week. Then either some GPU will crash and system will either automatically reboot or it will  be a hard hang.

No idea how you guys can run it for 3 months straight, these aren't Antminers.

i didn't experience any crash perfectly stable, the only time i stop it is to change the miner version that's it

now apparently there is a bug on my system that prevent win 10 from updating, which come in handy for me

For the old R9 390, the riser is cold.

R9 390 draws more power from the connectors, newer RX cards draws more power from the slot so riser will get warmer there.

As for the thread only reason I shut down my rigs is to update claymore's miner. Other than that It's been months since something crashed.

by 5xrx480 rig working only a month, but never reset or crashed or anything, just working

mostly multiple months, till i decide to change something hardwarewise or relocate the entire rig

less if some hardware is faulty or underpowerd (psu)

im using win 8.1 to avoid win10 forced updates and still use 6 gpus easily

I have simiar experience.

Is there a simply way to use 8.1 for 6 gpu? It seems complicated

I just made a post about it ^^
https://bitcointalk.org/index.php?topic=1725792.0

i do not understand you all, my rig not reset or crash even once after i build it

Actually you can turn off updates in windows 10 just gonna know how to and were  :) . i do it all the time when i do a clean install with a unattend.xml setup file and  install with WinPE using a USB  32 GB SSD  drive, it's faster then using the setup.exe and i edit my install.wim after i convert it from ESD  :) then customize the install on a VM and test it so no issue for me on windows 10 at all and the new AMD drivers that have driver signing I patched to use with a modded bios seem to have stopped random crashes i was getting when i switched from ZEC to ETH mining without first either a shut down  or reboot on 5 GPU rig I still reboot/shutdown to be safe but those drivers for me any way seemed to fix it .

you are a lucky man!
Can you tell what mobo and cards are using ?

If you do not over clock, it is stable.

If you have a windows rig, it's better to reboot it every few days, because with windows, you always get performance degradation over time. My rig is SSD based so a reboot till mining is about 35sec, i have a couple monitoring scripts that connect to pools API's and miner logs that will see if mining is below some thresholds, or nothing @ all, or if the rig is hanged and doesn't answer anymore, and accordingly will take action (soft  reboot (shutdown scripts) or hard reboot (with a custom RPI or wifi plugs) or miner kill/restart (usually i prefer a reboot, it's so fast, and you are sure it's going to mine @ optimal), i also added some scripts for auto-mining switch depending on the current value of coins and im working on automatic downclocking scripts for rigs that keep being unstable for a while. In general for any switch or something going wrong, when ssd or usbkey OS it's better to reboot.

For unix based rigs, you can let them run all the time basically but i think at least a reboot every odd week is always good. And with the same monitoring of course.

On the more mature mining programs, such as Cryptonight or Ethereum ones, I can have Windows rigs run for months without issues. The only time I would reboot is when physically cleaning the rig. With newer coins such as ZEC, with the constant changes and bugs I had to reboot a couple of times, but the more recent version are fairly stable.

If you are experiencing rigs locking up a lot, you need to look at your system as a whole. Are they getting too hot, are your pushing your PSU too much, are you overclocking your cards too much, etc.?

ordinary h81 pro btc, 5x rx480 4GB (some have samsung, some have modded hynix memory), powered usb risers with own build cables, but have 2 cheap psu + 1 good modular psu - 760W at wall (it was cheaper (2x 17$ + 90$) than one big modular psu), windows 7 x64 with some optimize, undervolted core, overclocked memory, 1210 sol/s

I mod the shit out of my gpus, rigs are running 24/7 with 1 crash in weeks...go softer on modding (if you do)? Or maybe reinstall drivers? Also if you're on Linux, my experience mining on it with the AMD GPU PRO was horrible, constant crashes. Since I migrated to Windows it's been super stable.

I run the rig will low frequency and they are very stable.

Varies with the rig for me.

 Most of my Windows rigs are lucky to go a week without a crash.

 My Linux-based triple R290 TheStilt BIOS modded very overclocked rig (my by far highest power eating HOTTEST running rig) hasn't had a crash at all except for power glitches and when I've had to shut it down to move it.

It depends a lot on the hardware and whether you are overclocking the gpus .
I find mining with the older gpus like R9 280x, 290 and 390 tend to be more stable than say
RX470 and RX480.  But then it may be because I have overclocked the memory.
Linux tends to be more stable than Win (7 at least) so long as your gpus are not RX-series.
Also, I find it depends on the mobo.  I used to like Asus mobos a lot, I have two Z97-As and they are
very finicky.
With R9 gpus on Linux I could take a 3 week vacation and the miners were still running when I came back :-)

For the underclocked rigs, the longs is about 30 days.

It depends on the weather outside, which coin your mining and if you are using a open air vs server chasis.

I had to reset it after a month of working for a onboard LAN failure which seems very wierd :)