Bitcoin Forum
May 05, 2024, 07:19:17 PM *
News: Latest Bitcoin Core release: 27.0 [Torrent]
 
   Home   Help Search Login Register More  
Pages: [1]
  Print  
Author Topic: windows 10 6 card mining rig freezes  (Read 315 times)
coenraad101 (OP)
Newbie
*
Offline Offline

Activity: 2
Merit: 0


View Profile
March 22, 2018, 02:36:44 PM
 #1

Hi guys

So I have a 6 card mining rig that has been running solidly for about 6-8 months now with no issues. Suddenly after the latest windows 10 build update the PC keeps freezing . Event viewer only says display driver failed , but recovered successfully. Pc still however freezes on me. I already removed the drivers with DDU Uninstaller and installed the latest blockchain driver. I am hosting at my office and if it freezes I need to manually restart the pc. So I lose money till the next morning or I have to driver to the office the same night. Please any help will be appreciated.

Build is: Biostar bt250 btc M/B Pentium CPU 4gb Ram 120gb WD SSD 6x RX 570 4gb aorus GPU
1714936757
Hero Member
*
Offline Offline

Posts: 1714936757

View Profile Personal Message (Offline)

Ignore
1714936757
Reply with quote  #2

1714936757
Report to moderator
You can see the statistics of your reports to moderators on the "Report to moderator" pages.
Advertised sites are not endorsed by the Bitcoin Forum. They may be unsafe, untrustworthy, or illegal in your jurisdiction.
1714936757
Hero Member
*
Offline Offline

Posts: 1714936757

View Profile Personal Message (Offline)

Ignore
1714936757
Reply with quote  #2

1714936757
Report to moderator
DannHomes
Jr. Member
*
Offline Offline

Activity: 30
Merit: 2


View Profile
March 22, 2018, 02:47:28 PM
 #2

What's your storage size (SDD/HDD) and do you use virtual memory? As I know the size of new version of window 10 is quite large. the page files may be too large so it causes your rig get error.
chandrarahmadewa
Member
**
Offline Offline

Activity: 308
Merit: 49


View Profile
March 22, 2018, 03:39:23 PM
 #3

Hi guys

So I have a 6 card mining rig that has been running solidly for about 6-8 months now with no issues. Suddenly after the latest windows 10 build update the PC keeps freezing . Event viewer only says display driver failed , but recovered successfully. Pc still however freezes on me. I already removed the drivers with DDU Uninstaller and installed the latest blockchain driver. I am hosting at my office and if it freezes I need to manually restart the pc. So I lose money till the next morning or I have to driver to the office the same night. Please any help will be appreciated.

Build is: Biostar bt250 btc M/B Pentium CPU 4gb Ram 120gb WD SSD 6x RX 570 4gb aorus GPU
Hello my friend,
Does your GPU go through bios mods? if yes
visit my post https://bitcointalk.org/index.php?topic=3175339.msg32870578#msg32870578
Would be good if you want to try this method, And please give us an update after.
leonix007
Sr. Member
****
Offline Offline

Activity: 1008
Merit: 297


Grow with community


View Profile
March 22, 2018, 03:53:01 PM
 #4

Hi guys

So I have a 6 card mining rig that has been running solidly for about 6-8 months now with no issues. Suddenly after the latest windows 10 build update the PC keeps freezing . Event viewer only says display driver failed , but recovered successfully. Pc still however freezes on me. I already removed the drivers with DDU Uninstaller and installed the latest blockchain driver. I am hosting at my office and if it freezes I need to manually restart the pc. So I lose money till the next morning or I have to driver to the office the same night. Please any help will be appreciated.

Build is: Biostar bt250 btc M/B Pentium CPU 4gb Ram 120gb WD SSD 6x RX 570 4gb aorus GPU

seems like not a windows update prob to me, since you already done DDU and stuffs to restore the drivers (assuming you installed the correct driver), I guess you need to visit your rig onsite and try to check/replace your riser. check also your OC settings.

Palmerson
Sr. Member
****
Offline Offline

Activity: 434
Merit: 255


Live cams shows pimped with cryptocurrency


View Profile
March 22, 2018, 03:59:11 PM
 #5

Hi guys

So I have a 6 card mining rig that has been running solidly for about 6-8 months now with no issues. Suddenly after the latest windows 10 build update the PC keeps freezing . Event viewer only says display driver failed , but recovered successfully. Pc still however freezes on me. I already removed the drivers with DDU Uninstaller and installed the latest blockchain driver. I am hosting at my office and if it freezes I need to manually restart the pc. So I lose money till the next morning or I have to driver to the office the same night. Please any help will be appreciated.

Build is: Biostar bt250 btc M/B Pentium CPU 4gb Ram 120gb WD SSD 6x RX 570 4gb aorus GPU

seems like not a windows update prob to me, since you already done DDU and stuffs to restore the drivers (assuming you installed the correct driver), I guess you need to visit your rig onsite and try to check/replace your riser. check also your OC settings.


After upgrading Windows 10 is not always possible to restore the old version. The problem is not in the equipment. My advice to you. Make the disc with minimum size. After reinstalling Windows and installing all the software cut off the remaining free space on disk C. this will not give in the future the opportunity to update Windows 10. It will write messages that it is impossible to install the update but it will not bother you.

 
           ▄▄██▄▄
       ▄▄███▀  ▀███▄▄
   ▄▄███▀▀  ▄▄▄▄  ▀▀███▄▄
▄██▀▀   ▄▄██▀▀██▄▄   ▀▀██▄
██▀  ▄▄██▀▀      ▀▀██▄▄  ▀██
██▄ ██▀              ▀██  ██
▀▀████▄▄▄             ██  ██
▄   ██▀▀▀████▄▄▄      ▀▀████
██  ██       ▀▀▀███▄▄▄     ▀
██  ██             ▀▀▀███▄▄▄
██  ██▄▄            ▄▄██ ▀██
██▄   ▀▀██▄      ▄██▀▀   ▄██
▀███▄    ▀██▄▄██▀    ▄███▀
    ▀▀██▄▄   ▀▀   ▄▄██▀▀
        ▀██▄▄  ▄▄██▀
           ▀▀██▀▀
          
♡   Live cams shows pimped with cryptocurrency   ♡
██Discord  |  Twitter  |  Medium  |  Youtube  |  Bitcointalk  |  Github██
           
           ▄▄██▄▄
       ▄▄███▀  ▀███▄▄
   ▄▄███▀▀  ▄▄▄▄  ▀▀███▄▄
▄██▀▀   ▄▄██▀▀██▄▄   ▀▀██▄
██▀  ▄▄██▀▀      ▀▀██▄▄  ▀██
██▄ ██▀              ▀██  ██
▀▀████▄▄▄             ██  ██
▄   ██▀▀▀████▄▄▄      ▀▀████
██  ██       ▀▀▀███▄▄▄     ▀
██  ██             ▀▀▀███▄▄▄
██  ██▄▄            ▄▄██ ▀██
██▄   ▀▀██▄      ▄██▀▀   ▄██
▀███▄    ▀██▄▄██▀    ▄███▀
    ▀▀██▄▄   ▀▀   ▄▄██▀▀
        ▀██▄▄  ▄▄██▀
           ▀▀██▀▀
coenraad101 (OP)
Newbie
*
Offline Offline

Activity: 2
Merit: 0


View Profile
March 23, 2018, 07:16:58 AM
 #6

Hi guys

Thanks so much for all the replies I will definitly keep you updated.. SSD is 120gb with 16gb of virtual memory. Cards are modded yes and I did the timing straps myself, as its just much safer downloading one from anortek ( believe me had that issue). I have like 16 rigs running and all have the same settings and it is only this one rig giving me the issues.
chandrarahmadewa
Member
**
Offline Offline

Activity: 308
Merit: 49


View Profile
March 23, 2018, 10:12:49 AM
 #7

..I have like 16 rigs running and all have the same settings and it is only this one rig giving me the issues.
I have some solutions for you
1. checked the PSU and power cables? I am afraid there is a burning,
2. tried replacing GPU riser, and USB cable.
Quickset45
Newbie
*
Offline Offline

Activity: 6
Merit: 0


View Profile
May 27, 2018, 02:08:55 PM
 #8

I have a very similar problem.  I had (2) seven card rigs running fairly stable for 3 or 4 months.  Monday morning 5/21 I got notification from ethermine that one of my rigs was off line, looking at the stats the other was only working at half capacity. (2 separate machines both took a significant hit within a 1/2 hour of each other.  Looking at both machines checking for any leaking capacitors, checking connections I rebooted, both machines once through the bios, windows would only go to a black screen.

My machines are both Asus Z270a Prime mobo, both have 8 gigs of ram, both have plenty of power, and I'm using MSI RX 580 armor cards. (all bios modded, I used anorak tech bio's and then backed off clocks a little for card life with AMD catalyst wattman.  I had been using the latest AMD drivers circa 12/31/2017 in compute mode.  I was hashing between 29 and 30 for each card, they were nice and cool, and I was content.  I am using unregistered win 10 Pro (an old version I own.. so I had to go through a bunch of updates upon install)  On both of the rigs I right after I built them I was getting a windows message "task host is preventing background tasks"  I youTubed a fix where you changed a HEX key value in start from 2 to 4, and this stopped it.

I toggled the PSUs Monday morning and came back this weekend, to maintenance the machines and trouble shoot.

First off, I updated bios double checked all my settings 4G encoding, Gen 2 etc. Unplugging all the cards, the pc booted right up no problems, I DL claymores latest miner, scrubbed Cayalyst drivers and DL the latest and started plugging in GPUs.  1 card no problem, 2 cards no problem, 3 cards machine slowing.. like 20 seconds of wading through mouse button clicks to switch cards to compute mode.  4 cards over a minute per mouse click... right click on screen (wait over a minute) left click AMD catalyst control center (wait 2 minutes) click gobal settings (wait 3 minutes)

At four cards installed, I started claymores.. no issues.  Everything running ok.

Installing the 5th card, Windows became unusable.  like 4 or 5 minutes between responses. sometimes freezing altogether.

I can switch to any three cards and all is ok, so I don't suspect riser card issues, but I have riser cards to swap out, maybe I'll try them. 
Before this event, I had maybe a 12 second lag per screen switching to compute mode with 7 cards installed. I"m assuming this is normal?

I'm looking for advice,
thanks.
nmersulypnem
Full Member
***
Offline Offline

Activity: 238
Merit: 100


View Profile
May 27, 2018, 07:23:15 PM
 #9

I am having a very similar problem.  After the Win 10 update, I cannot get system to boot up with more than 4 GPUs.   All GPUs work fine if there's only 4.  I do see lagging and GPUs behaving weird with 4 if the BC driver installed (it looks like an issue with the HD audio driver - but disabling in the bios strangely doesn't stop it).

mobo is ASUS z270-p, and I updated bios to 1205 - still no dice.  ...and there's no way to rollback the bios unfortunately.

I guess I'll try re-installing an earlier version of windows from USB...
Quickset45
Newbie
*
Offline Offline

Activity: 6
Merit: 0


View Profile
June 01, 2018, 11:28:30 AM
 #10

I found this suggestion from a guy named wiskeywiskey on another board, he adds a registry value for each gpu.. found where to add the values, but I couldn't figure out how to add the key.  Apparently, this solves the painfully slow time between switching from graphics mode to compute mode if you're not using blockchain drivers.  I'm wondering if this could help the crashing problem?


Windows Registry Editor Version 5.00
[HKEY_LOCAL_MACHINE\SYSTEM\CurrentControlSet\Control\Class{4d36e968-e325-11ce-bfc1-08002be10318}\0000] "KMD_EnableInternalLargePage"=dword:00000002
And for 0001, 0002, etc.
permalinkembedsavereportreply



Reply 1
[–]jellylovee 1 point 3 months ago
U da best
permalinksaveparentreportreply





Reply 2
[–]nFo911 1 point 10 days ago
thanks whiskeyx2 (recent w10 april update... and my 2nd gpu would not go back to 31 mh/s... added (new entry) the kmd setting + restart... and i m good to go)
androstan1234
Jr. Member
*
Offline Offline

Activity: 251
Merit: 6


View Profile
June 01, 2018, 12:09:51 PM
 #11

Almost every miner that has updated to 1803 has has problems such as sluggishness, freezing, etc.  The first thing you should have done is roll back Windows.
xlm19877
Member
**
Offline Offline

Activity: 140
Merit: 20


View Profile
June 01, 2018, 12:42:39 PM
 #12

before one week this problem happend  with me

i try to stop 3 gpus and test the rig about one hour then no problem

after that i run gpus 4 then test then 5 then 6 until i know which card do the problem

i replace bios for the card and it work well

some time the problem from riser

so you must test
PRC_May
Member
**
Offline Offline

Activity: 224
Merit: 15


View Profile
June 01, 2018, 12:53:15 PM
 #13

bro

check the PSU Cable

at the afterburner do some under clock and keep watch claymore to see any card hung

and some time the problem from riser or his USB cable 
Three33
Newbie
*
Offline Offline

Activity: 3
Merit: 0


View Profile
June 01, 2018, 12:57:06 PM
 #14

make clean install Windows 10 1703
Wotan Wipeout
Member
**
Offline Offline

Activity: 438
Merit: 27


View Profile
June 01, 2018, 02:24:56 PM
 #15

Because you control your rig from the office i expect its not nearby.

So my first step would be to start each GPU alone.

Mostly its a overloaded PCI-E lane or a Riser.

In case its the PCI-E Lane and you have connected lets say 2 GPUs to one cabel, expect to have
a crash at 2 GPUs.

If its only the Riser it should be at 1 GPU.
cubbo
Newbie
*
Offline Offline

Activity: 26
Merit: 0


View Profile
June 01, 2018, 02:48:22 PM
 #16

I would say your best bet is to do a fresh install of your previous version of windows 10 and there is a registry entry where you tell the machine to get its updates from a local wsus server rather than microsoft update. If you set this address to a random one on your network windows won't be able to get any updates and should keep you on Windows 10 (1703 or whatever it was) until YOU decide that you want to do updates.

I am going to do this over the weekend as I'm having exactly the same problems as yourself and it is so frustrating to have the lower hashrates and the crashing every day.
blacktux88
Member
**
Offline Offline

Activity: 308
Merit: 10


View Profile WWW
June 04, 2018, 08:47:00 AM
 #17

do you have overlockted the cards ?

we had lots of problems with mem overlocking in windows 10 .

that could be an issue too.

thanks

Pages: [1]
  Print  
 
Jump to:  

Powered by MySQL Powered by PHP Powered by SMF 1.1.19 | SMF © 2006-2009, Simple Machines Valid XHTML 1.0! Valid CSS!