Claymore's Dual Ethereum AMD+NVIDIA GPU Miner v15.0 (Windows/Linux) - page 472.

digibum

newbie

Activity: 11

Merit: 0

Quote from: toxiroxi on January 18, 2018, 11:19:23 AM

hi guys,

maybe someone can help. i have a problem which i cannot really identify.
I am running 2 rigs which are completely identical. The first one runes 1709 win10x64 with beta blockchain drivers and does that pretty stable. sometimes the watchdog is restarting the rig but that's ok in anyway for windows.

The rig2 has the exact same architecture and hardware, but as i have been experiencing issues with 1709 setup and drivers i decided to go with 1703. installation is completely the same.
But after 2-3 days approximately the rig stops responding. the watchdog is not kicking in restarting the rig - i cannot figure out why. There is also nothing in the logfile which might help.

Any ideas? I have attached the last lines from the claymore logfile. btw. i am running 10.3.
Thanks so much for your help - i would be great getting those freezing state rid of. rebooting is ok, but it doesnt at some point....
...
there is no error or anything in the log - this are the last lines before hanging.....

any ideas or help is appreciated. thanks in advance!

This smels like wrong bios configuration...
First update motherboard BIOS! And then check these configurations...

DMI Max Link Speed => Gen2
PCIEX16 slot's Link Speed => Gen2
M.2 slot's Configuration => PCIE mode (if you have M.2 slot risers)
Above 4G Decoding => Enable

rdluffy

legendary

Activity: 2366

Merit: 1408

Quote from: MassDestruction on January 18, 2018, 12:30:09 PM

Hey Guys

I'm having this weird issue with my rig where it will crash and down the system everyday at 7AM. I'm currently using nvoc 19(ubuntu) with claymore miner to mine ETH. Currently this feels like a software issue, and I'm exploring my options to see what the possible culprit is.

So what could Claymore be doing to cause this issue if it is? If so are there any possible solutions? I've provided info about my rig, settings, and the mining pool if that helps.

Mining Pool:
us1.ethermine.org:4444

Claymore Settings:
-epsw x
-mode 1
-dbg -1

Rig info:
Mother Board: Biostar TB250 Pro
Video Cards: 1 x GTX 1070, 7 x GTX 1060, 2 x GTX 1050Ti
Ram Size: 16 GB
Storage: SanDisk 120GB SSD
Processor: Intel Celeron G3900
Power Supply: Corsair AX860, Corsair VX550

I'm having a lot of issues too
I have a stable rig running 4 RX 580 but the another one is crashing a lot
Sometimes crashes at same time for 3 days....after that I have to reinstall drivers, reapply AMD Pixel Patch...I reinstalled Windows, I made the BiosMod again in every GPU and it's unstable
The system crashes and I have to manual reboot

I don't know what to do this time

I already:
Reinstall windows 10 pro
Use differents claymore
Change Risers, change positions on mobo

Metroid

sr. member

Activity: 2142

Merit: 353

Xtreme Monster

Quote from: j2james on January 18, 2018, 01:08:39 PM

Quote from: Claymore on January 18, 2018, 07:52:03 AM

v10.5:- added version for CUDA 9.1

Hi. Tell me please what should I do to use CUDA 9.1?

Looks like I still use 8.0:

CUDA Driver Version/Runtime Version: 9.1/8.0

you need to update the nvidia drivers to latest.

j2james

newbie

Activity: 80

Merit: 0

Quote from: Claymore on January 18, 2018, 07:52:03 AM

v10.5:- added version for CUDA 9.1

Hi. Tell me please what should I do to use CUDA 9.1?

Looks like I still use 8.0:

CUDA Driver Version/Runtime Version: 9.1/8.0

wtfonly16

full member

Activity: 285

Merit: 105

try mine for 1 hour, then close miner.

what happens?

MassDestruction

newbie

Activity: 10

Merit: 0

Hey Guys

I'm having this weird issue with my rig where it will crash and down the system everyday at 7AM. I'm currently using nvoc 19(ubuntu) with claymore miner to mine ETH. Currently this feels like a software issue, and I'm exploring my options to see what the possible culprit is.

So what could Claymore be doing to cause this issue if it is? If so are there any possible solutions? I've provided info about my rig, settings, and the mining pool if that helps.

Mining Pool:
us1.ethermine.org:4444

Claymore Settings:
-epsw x
-mode 1
-dbg -1

Rig info:
Mother Board: Biostar TB250 Pro
Video Cards: 1 x GTX 1070, 7 x GTX 1060, 2 x GTX 1050Ti
Ram Size: 16 GB
Storage: SanDisk 120GB SSD
Processor: Intel Celeron G3900
Power Supply: Corsair AX860, Corsair VX550

Swiped

member

Activity: 100

Merit: 10

toxiroxi

newbie

Activity: 17

Merit: 0

Quote from: Victorio on January 18, 2018, 11:50:27 AM

Quote from: toxiroxi on January 18, 2018, 11:19:23 AM

I am running 2 rigs which are completely identical. The first one runes 1709 win10x64 with beta blockchain drivers and does that pretty stable. sometimes the watchdog is restarting the rig but that's ok in anyway for windows.

The rig2 has the exact same architecture and hardware, but as i have been experiencing issues with 1709 setup and drivers i decided to go with 1703. installation is completely the same.
But after 2-3 days approximately the rig stops responding. the watchdog is not kicking in restarting the rig - i cannot figure out why. There is also nothing in the logfile which might help.

Have you disabled Windows Updates and Driver Update? You can use Win 10 tweakers to do it, e.g. Win 10 Spy Disabler.
I've seen similar issues when a new driver is installed from Windows Update repository.

Thanks for your reply - yes those are disable, but i will double check again to be sure.

Victorio

full member

Activity: 281

Merit: 100

Quote from: toxiroxi on January 18, 2018, 11:19:23 AM

I am running 2 rigs which are completely identical. The first one runes 1709 win10x64 with beta blockchain drivers and does that pretty stable. sometimes the watchdog is restarting the rig but that's ok in anyway for windows.

The rig2 has the exact same architecture and hardware, but as i have been experiencing issues with 1709 setup and drivers i decided to go with 1703. installation is completely the same.
But after 2-3 days approximately the rig stops responding. the watchdog is not kicking in restarting the rig - i cannot figure out why. There is also nothing in the logfile which might help.

Have you disabled Windows Updates and Driver Update? You can use Win 10 tweakers to do it, e.g. Win 10 Spy Disabler.
I've seen similar issues when a new driver is installed from Windows Update repository.

embombed

newbie

Activity: 1

Merit: 0

Anyone have any luck getting 10.4 running on Vega 64 running ubuntu 16.04.03 ?

getting the following:
\
AMD Cards available: 1
GPU #0: gfx900 (Device 687f), 8176 MB available, 64 compute units (pci bus 3:0:0)
GPU #0 recognized as Vega
POOL/SOLO version
Cannot build OpenCL program for GPU 0
GPU #0: algorithm ASM
No NVIDIA CUDA GPUs detected.
Total cards: 1
AMD ADL library not found.

No pool specified for Decred! Ethereum-only mining mode is enabled
ETHEREUM-ONLY MINING MODE ENABLED (-mode 1)
ETH: miner-proxy stratum mode
Watchdog enabled

ETH: Authorized
Setting DAG epoch #164...

GPU #0: gfx900 (Device 687f), 8176 MB available, 64 compute units (pci bus 3:0:0)
ETH - Total Speed: 0.000 Mh/s, Total Shares: 0, Rejected: 0, Time: 00:00
ETH: GPU0 0.000 Mh/s
Incorrect ETH shares: none
1 minute average ETH total speed: 0.000 Mh/s
Pool switches: ETH - 0
Current ETH share target: 0x00000000ffb34c02 (diff: 4300MH), epoch 164(2.28GB)
GPU0 t=62C fan=13%

Setting DAG epoch #164 for GPU0
Create GPU buffer for GPU0
GPU0, OpenCL error -48 (0) - cannot create DAG on GPU
GPU 0 failed

toxiroxi

newbie

Activity: 17

Merit: 0

hi guys,

maybe someone can help. i have a problem which i cannot really identify.
I am running 2 rigs which are completely identical. The first one runes 1709 win10x64 with beta blockchain drivers and does that pretty stable. sometimes the watchdog is restarting the rig but that's ok in anyway for windows.

The rig2 has the exact same architecture and hardware, but as i have been experiencing issues with 1709 setup and drivers i decided to go with 1703. installation is completely the same.
But after 2-3 days approximately the rig stops responding. the watchdog is not kicking in restarting the rig - i cannot figure out why. There is also nothing in the logfile which might help.

Any ideas? I have attached the last lines from the claymore logfile. btw. i am running 10.3.
Thanks so much for your help - i would be great getting those freezing state rid of. rebooting is ok, but it doesnt at some point....

Code:

ETH: 01/18/18-14:59:08 - New job from eu1.ethermine.org:4444
14:59:08:700 a88 target: 0x0000000112e0be82 (diff: 4000MH), epoch 164(2.28GB)
14:59:08:716 a88 ETH - Total Speed: 57.854 Mh/s, Total Shares: 476, Rejected: 0, Time: 09:24
14:59:08:716 a88 ETH: GPU0 29.061 Mh/s, GPU1 28.794 Mh/s
14:59:08:716 a88 SC - Total Speed: 1041.379 Mh/s, Total Shares: 240, Rejected: 0
14:59:08:716 a88 SC: GPU0 523.090 Mh/s, GPU1 518.289 Mh/s
14:59:18:873 a88 got 248 bytes
14:59:18:873 a88 buf: {"id":0,"jsonrpc":"2.0","result":["0x2aa5cfd97efc69f86d2396d0502a258b70e51cce1576b4177bf1adcd69fe4825","0xd5d85ca51867514082660716ef2bedbb55a65250622456390b4a76c4fe98db95","0x0112e0be826d694b2e62d01511f12a6061fbaec8bc02357593e70e52ba","0x4b37ad"]}

14:59:18:873 a88 parse packet: 247
14:59:18:873 a88 ETH: job changed
14:59:18:873 a88 new buf size: 0
14:59:18:889 a88 ETH: 01/18/18-14:59:18 - New job from eu1.ethermine.org:4444
14:59:18:889 a88 target: 0x0000000112e0be82 (diff: 4000MH), epoch 164(2.28GB)
14:59:18:889 a88 ETH - Total Speed: 59.505 Mh/s, Total Shares: 476, Rejected: 0, Time: 09:25
14:59:18:889 a88 ETH: GPU0 29.096 Mh/s, GPU1 30.409 Mh/s
14:59:18:904 a88 SC - Total Speed: 1071.093 Mh/s, Total Shares: 240, Rejected: 0
14:59:18:904 a88 SC: GPU0 523.726 Mh/s, GPU1 547.367 Mh/s
14:59:19:701 a88 send: {"id":6,"jsonrpc":"2.0","method":"eth_submitHashrate","params":["0x38b356a", "0x0000000000000000000000000000000000000000000000000000000006bac143"]}

14:59:19:748 a88 got 39 bytes
14:59:19:748 a88 buf: {"id":6,"jsonrpc":"2.0","result":true}

14:59:19:748 a88 parse packet: 38
14:59:19:748 a88 new buf size: 0
14:59:22:764 11fc em hbt: 15, dm hbt: 15, fm hbt: 62,
14:59:22:764 11fc watchdog - thread 0 (gpu0), hb time 125
14:59:22:764 11fc watchdog - thread 1 (gpu0), hb time 46
14:59:22:764 11fc watchdog - thread 2 (gpu1), hb time 62
14:59:22:764 11fc watchdog - thread 3 (gpu1), hb time 140
14:59:29:249 d7c GPU0 t=58C fan=55%, GPU1 t=68C fan=86%
14:59:30:827 a88 ETH: checking pool connection...
14:59:30:827 a88 send: {"id":3,"jsonrpc":"2.0","method":"eth_getWork","params":[]}

14:59:30:874 a88 got 248 bytes
14:59:30:874 a88 buf: {"id":3,"jsonrpc":"2.0","result":["0x2aa5cfd97efc69f86d2396d0502a258b70e51cce1576b4177bf1adcd69fe4825","0xd5d85ca51867514082660716ef2bedbb55a65250622456390b4a76c4fe98db95","0x0112e0be826d694b2e62d01511f12a6061fbaec8bc02357593e70e52ba","0x4b37ad"]}

14:59:30:874 a88 parse packet: 247
14:59:30:874 a88 ETH: job is the same
14:59:30:874 a88 new buf size: 0
14:59:39:719 a88 send: {"id":6,"jsonrpc":"2.0","method":"eth_submitHashrate","params":["0x37ac1b4", "0x0000000000000000000000000000000000000000000000000000000006bac143"]}

14:59:39:766 a88 got 39 bytes
14:59:39:766 a88 buf: {"id":6,"jsonrpc":"2.0","result":true}

14:59:39:766 a88 parse packet: 38
14:59:39:766 a88 new buf size: 0

there is no error or anything in the log - this are the last lines before hanging.....

any ideas or help is appreciated. thanks in advance!

luis31

newbie

Activity: 10

Merit: 0

okay my sollution was: i decided to change the grafikcards, i use actually only nvidia cards.

but now i have the next problem i use 13 gtx 1070 cards they run without problems but
how can i overclock all cards? msi afterburner shows me only 8 cards i cant overclock the forther 5 cards?

it is not easy to assemble an good rig... but problems are here to solve it... :-)

i will rise the amount of the cards to 18 is this possibel with the asus mining exptert mobo?

thanks...

tf2addict

full member

Activity: 675

Merit: 100

Quote from: wtfonly16 on January 17, 2018, 09:21:22 PM

this exploit has already been discovered long ago

https://github.com/nicehash/NiceHashMinerLegacy/issues/96

And yet

And it still exists?

Sparki01

newbie

Activity: 2

Merit: 0

Quote from: Claymore on January 18, 2018, 07:52:03 AM

v10.5:

- applied another workaround for ADL, you must not see "Temperature control thread hangs" message now.
- added version for CUDA 9.1.
- now miner shows a warning if you enabled full remote management (-mport > 0) and did not specify -mpsw parameter.
- now miner shows a warning if you specified default ETH wallet (from default start.bat).

wow, this version 10.5 runs HOT (compared to 10.3). at least 3-4 degrees hotter on each card (CUDA 8.0) and 5-6 degrees hotter on each card (CUDA 9.1), anyone else seeing this?

4x GTX1070

On the up side

Code:

-ttli

works

never seen it kick in before...

androstan1234

jr. member

Activity: 251

Merit: 6

Thank you for the quick and helpful response.

Claymore

donator

Activity: 1610

Merit: 1325

Miners developer

Quote from: androstan1234 on January 18, 2018, 08:49:23 AM

Claymore, your miner is logging the following messsge:

Quote

AMD topology bug detected, fix it.

I don't really know what this means, so I don't know how to address it, or if it needs to be addressed.
4x rx 570 and 1x gtx 1070. Stable on latest adrenalin driver and latest GeForce running v10.4.
Can you help me understand this message?
If it helps, I have another rig that does not get this message. 5x rx 580 and 1x gtx 1070. Stable on aug23 blockchain driver and latest GeForce, running v10.4.
Both win10 pro.

Don't worry about this message, it means that miner applies a workaround for bug in Adrenalin drivers which breaks temperature/fan management.

androstan1234

jr. member

Activity: 251

Merit: 6

Claymore, your miner is logging the following messsge:

Quote

AMD topology bug detected, fix it.

I don't really know what this means, so I don't know how to address it, or if it needs to be addressed.

4x rx 570 and 1x gtx 1070. Stable on latest adrenalin driver and latest GeForce running v10.4.

Can you help me understand this message?

If it helps, I have another rig that does not get this message. 5x rx 580 and 1x gtx 1070. Stable on aug23 blockchain driver and latest GeForce, running v10.4.

Both win10 pro.

kkourmousis

member

Activity: 119

Merit: 10

Quote from: Matias on January 17, 2018, 02:07:01 PM

Quote from: BennyT on January 17, 2018, 01:50:07 PM

Quote from: Matias on January 17, 2018, 01:40:55 PM

Is it really so, that mining two coins doesn’t decrease Ethereum hashrate? What is the best coin to mine with Ethereum?

Yes it does lower the ethereum hash rate. Most people think the decrease is not significant. Check what to mine for profitability on the second coin, knowing this always changes. I like Decred

But the profits from the second coin outweighs the 2% devfee (instead of 1%) and the decrease in the ethereum hashrate?

You have more factors to consider if you must wether dual mining is profitable for you:

lower ETH hashrate
+1% dev fee
+power consumption (which is pretty much actually)
higher temperatures
faster hardware wear

all of the above factors are different for each gpu, so do your research

In my 580s, it is indeed profitable to mine ETH + DCR (even though in my country I pay 0.3$/kwh Cry

)

dwdhu

newbie

Activity: 37

Merit: 0

hi,
please submit your updated hashrate results to
https://gpustats.com
thank you

BulldogPO

newbie

Activity: 20

Merit: 0

Quote from: ?? on ??

Quote from: BulldogPO on January 18, 2018, 01:24:47 AM

Quote from: ?? on ??

Quote from: BulldogPO on January 16, 2018, 05:48:22 AM

I tried to mod bios but now Windows does BSOD when installing blockchain drivers.

Try installing the latest 17.12.2 Adrenalin drivers. After it installed you can set the Compute mode in the settings.

Radeon settings -> Gaming -> Global settings -> Per GPU - Compute mode

Also how did you mod? Did you download the BIOS from your card and edit that? Could be that the changes are too high for your card and when it initialises it crashes.

I restored original ASUS bios to my RX 570 4GB cards, removed drivers totally with DDU, rebooted, installed Adrenalin 17.12.2 drivers, rebooted, switched to compute mode, updated Claymore to v10.4 and now I got 16MH/s Ethereum speed per card Sad

What next?

Can you check HWMonitor and see if your cards are going to full clocks? I had an issue that some cards get stuck on half/low clocks. I fixed it by adding -minspeed so the miner restarts when a specific speed isnt met. This way it will restart untill its corrected. Also -lidag 1 can help not getting them stuck.

I check that today and will report back.
Looks like Afterburner is not working while GPU´s are in compute mode?

Topic: Claymore's Dual Ethereum AMD+NVIDIA GPU Miner v15.0 (Windows/Linux) - page 472. (Read 6590718 times)