Version 3.2.0 Beta 11 is ready

[UselessGuru]

https://github.com/MultiPoolMiner/MultiPoolMiner/releases/tag/Beta3.2.0

Please also report if all is working fine - only so I can know that we get rid of all the bugs :-)

[Majorbafe]

so.... why is my mpm not mining specific coins.... I have yet to see a mining command with the mc=*** tag in it. This has been happening with v3.1 and v3.2

[UselessGuru]

I have yet to see a mining command with the mc=***

ZergPool was the only pool that offered the functionality. But ZergPool is dead.

[Majorbafe]

Ok... so what are the *Coin.ps1 files in the pool folder used for?

[UselessGuru]

Ok... so what are the *Coin.ps1 files in the pool folder used for?

To mine Coins instead of algos...
See README: ...(dis-)allows mining selected coins only

[torsrex]

@torsrex

Should these have 1 and 2 threads respectively?

Should ;-) I am thankful for all your input 👍

Next - and hopefully last - try:

AMD_CPU_NVIDIA-FireiceCryptonight_v2.7.0.zip

Edit: This fix is also in 3.2.0 Beta 11.

I apologize for the late reply. Using beta 11 the config files look like this (it seems to enumerate all the threads twice):
ThreadsConfig-CPU-CryptonightV8-GenuineIntel4CoreCPU-2.txt
ThreadsConfig-CPU-CryptonightV8-GenuineIntel4CoreCPU-1.txt

[UselessGuru]

Where-Object {$.Path -eq $Miner.Arguments -and $.Arguments -EQ $Miner.Arguments}

The files are as intended. *-2 means TWO threads (per CPU core in this case).
For CPUs with HyperThreading only each other core should be used, e.g. as stated here
fireice-uk/xmr-stak#957 (comment) or
fireice-uk/xmr-stak#640 (comment).

[skovesh]

with beta 3.2.0 beta 11 still getting red lines as:
Write-Log : Miner (AMD-SRBMinerCryptonight_v1.7.2-1-GPU#00-GPU#01 {CryptonightHeavy@NiceHash}) has failed.
At F:\MINE\MultiPoolMiner\MultiPoolMiner.ps1:999 char:17

• ... Write-Log -Level Error "Miner ($($Miner.Name) {$(($Miner. ...

•             ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
  

• CategoryInfo : NotSpecified: (:) [Write-Error], WriteErrorException
• FullyQualifiedErrorId : Microsoft.PowerShell.Commands.WriteErrorException,Write-Log

[UselessGuru]

with beta 3.2.0 beta 11 still getting red lines as:

I bet you will get the same error if you run the command outside of MPM.
What needs to be changed that the error goes away?
I cannot reproduce your error. It benchmarks SRBMiner with all algos flawlessly.

[skovesh]

Doesn't it has to do nothing with MultiPoolMiner.ps1 ??
.....At F:\MINE\MultiPoolMiner\MultiPoolMiner.ps1:999 char:17....

[UselessGuru]

Doesn't it has to do nothing with MultiPoolMiner.ps1 ??
I bet you will get the same error if you run the command outside of MPM.

Look for the exact command that MPM used to launch the miner (check your MPM log file).
Rund exact this command outside of MPM. I bet you will get the same error.
It would be very helpful if you posted some more details so that I can assist you better. My crystal ball is currently in maintenance ;-)

[FruitFace]

@UselessGuru

I'm seeing an issue with AMD_NVIDIA-PhoenixminerEthash_v4.0a

For some reason, one of 4 GPUs is being left unused, despite the output stating that it will be used. I only noticed because I was reviewing afterburner profiles, and the actual GPU clocks were at idle for that card. Other miners seem to work properly.

I notice the command line string for Phoenixminer reads "-gpu 234" and the unused card per MPM is 1. Does the MPM numbering match that of the miner? I wonder if somehow my card 4 is ending up assigned to two separate miners at the same time. GPUs 1, 2,3 are 1070s, 4 is a 1080ti.

Thanks!

MultiPoolMiner_2018-12-16_23-02-23.txt
MultiPoolMiner_2018-12-16.txt

[skovesh]

... Look for the exact command that MPM used to launch the miner (check your MPM log file).
Rund exact this command outside of MPM. I bet you will get the same error....

Running alone I get:
Can't open PoolConf_NiceHash-CryptonightHeavy.json, neither is pool defined in parameters
Unknown error parsing pools config file
Thanks for the great help!

[UselessGuru]

Can't open PoolConf_NiceHash-CryptonightHeavy.json, neither is pool defined in parameters
Unknown error parsing pools config file

Did you run the command in the miner directory (NOT the MPM directory)?

Please zip all *.txt & *.json files and upload them here.. I also need the output from http://localhost:3999/devices

[UselessGuru]

I notice the command line string for Phoenixminer reads "-gpu 234" and the unused card per MPM is 1. Does the MPM numbering match that of the miner? I wonder if somehow my card 4 is ending up assigned to two separate miners at the same time. GPUs 1, 2,3 are 1070s, 4 is a 1080ti.
one of 4 GPUs is being left unused

Please post http://localhost:3999/devices

I don't understand why MPM start numbering your cards with GPU#01, It should start with GPU#00

I recommend you try '-UseDeviceNameForStatsFileNaming:true' (note: this will trigger re-benchmarking unless you manually rename the stat files). This parameter will make it easier for you to match cards & miners.

[FruitFace]

@UselessGuru

Here's the output of the devices page. I am running with the parameter now, and its benchmarking. My other rig with a single card is showing GPU0, but right, 1-4 on this one.

I will re-enable ethash later today, and see how it behaves. This issue may have begun in beta 10 or 9. I just gave those versions a skip, since I didn't have time to look into what was going on..

Thx
devices.txt

[skovesh]

...Please zip all *.txt & *.json files and upload them here.. I also need the output from http://localhost:3999/devices...
Here are all files:
txt_json.zip

[UselessGuru]

@skovesh

Did you run the command in the miner directory (NOT the MPM directory)?
Please zip all *.txt & *.json files and upload them here.. I also need the output from > http://localhost:3999/devices

Thank you for all the files... I only meant the txt/json file from the problematic SRB miner.

Can't open PoolConf_NiceHash-CryptonightHeavy.json, neither is pool defined in parameters
Unknown error parsing pools config file

PoolConf_NiceHash-CryptonightHeavy.json does exist in the ZOP file, so you probably made a mistake when trying to run SRB miner standalone.
Please try again and post a screenshot of the CMD window, make sure the command line is visible ;-)

[UselessGuru]

@FruitFace

Is ist just PhoenixMiner 4.0b behaving like this? All miners should be affected :-(

Your device config looks broken! Did you exclude any GPUs with '-ExcludeDeviceName'?
Why does your GPU-numbering start with GPU#01 - I expect GPU#00??? And all "index*" elements also start with 1 (instead of 0). This is wrong!

Can you tell me which values for '-gpu 234' would be working in your case? (I assume '-gpu 123' & '.gpu 4')
In the miners readme it says -gpus <123 ..n> Use only the specified GPUs (if more than 10, separate the indexes with comma) (note: starting with 1, MPM already increments the index values (starting from 0!) to take care of this).
In your case the indexes start already too high.

[FruitFace]

@UselessGuru

Is it possible that some miners ennumerate non-mining cards? I'm driving the display off an Intel Integrated display adapter.

Also seeing a few write errors...attached.
write error.txt

Just reenabled ethash to watch how the miners handle the cards

Claymore and excavator find all 4 cards and use them appropriately. They are addressed as cards 0-4, by the miners themselves.
Eminer doesn't appear to work, but does address the cards as 1-4.
EthMiner works, addresses a 0-4
Bminer sees 0,1,2
Phoenixminer sees only 3 cards.

I installed a fresh copy, in a new directory, no config.txt, just nicehash and ethash. Same issues. Attached logs & devices output below.

devices.txt
Logs.zip

[UselessGuru]

@FruitFace

Is it possible that some miners ennumerate non-mining cards? I'm driving the display off an Intel Integrated display adapter.

Yes, this might actually cause this issue. Let me think!

Also seeing a few write errors...attached.

These do not help me at all ;-(
Same procedure as always: Run the miner with the exact same parameters outside of MPM and see what happens. Most likely all issues are caused by the incorrect device id mappings.

I prepared a special version of Include.psm1. Please rename the attached file from .txt and replace the existing one and start MPM.
It will then create a file 'DumpOpenCL.txt' in the MPM directory. Please upload this file. Thank you!

Include.psm1.txt

[FruitFace]

@UselessGuru

Here you go.
DumpOpenCL.txt

[UselessGuru]

@FruitFace

The extra Intel Integrated display adapter is the cause.

Please replace include.psm1 again. It will create 2 DumpOpenCL* files.

Include.psm1.txt

[FruitFace]

@UselessGuru

It only appears to have created one.
DumpOpenCL.txt

[UselessGuru]

It only appears to have created one.

My bad - I uploaded the wrong file...

Include.psm1.txt

[FruitFace]

@UselessGuru
No worries, here they are.
DumpOpenCL.txt
DumpOpenCLPlatforms.txt

[UselessGuru]

@FruitFace

Thank you for the files.

Since I cannot test your scenario I would like you to some tests for me:
Attached 2 files as replacement for Include.psm1 (do not forget to rename them to .psm1).

For faster testing please remove (e.g. move them to a different directory) all miner files in the MinersLegacy folder except the ones listed below:

Then replace Include.psm1 with Include.psm1_Ignore_Intel.txt
and start MPM. Are the deviceIDs AND platfomr ID values working?
Then the same with Include.psm1_Ignore_Intel-CPU.txt

Which file version (name) does work for you? (if any :-((???)

[mkirov0]

I think -Interval is broken. It works as intended when -Interval 60 to 180, but I tested increases over 180 (up to 1200 now) and in reality it keeps 3 minute intervals when MPM is running. I tried changing the actual $Interval in MPM.ps1 and still no change. Just posting this FYI, the issue doesn't really bother me.

[UselessGuru]

I think -Interval is broken.
FYI, the issue doesn't really bother me

Thank you for posting. I'll investigate.

FYIO: 3.3.0 will bring changes to the loop handling.

[UselessGuru]

@mkirov0

I think -Interval is broken.

With what version / beta is that?

[mkirov0]

I tested Beta 11
I did not re-download and "reinstall" to make sure its not only on my end. Like I said it's not that big of a deal to me.

[FruitFace]

@UselessGuru

In short, "Include.psm1_Ignore_Intel-CPU.txt" seems to work correctly. Specifically NSGMiner and CryptonightV7 and Ethminer work ok. CryptonightV8 has errors relating to admin privileges and other things, but not related to enumeration. I don't mine Cryptonight, so can't really speak more to it. I don't have any AMD cards, so I can't speak toward those.

The other .psm1 file caused NSGMiner and Ethminer to not find the GPUs. CryptonightV7 was unaffected.

I attached the log files for each run.

Edit: I also tries Phoenixminer, and that is happy. I put the corrected psm into my non-test mining folder, for the rig, and am benchmarking. It's looking good so far :)

Cheers

Include.psm1_Ignore_Intel.zip
Include.psm1_Ignore_Intel-CPU.zip

[UselessGuru]

@FruitFace

Thank you for your feedback.

Cryptonight & Adminrights: Check the web for
#2257 (comment)

[FruitFace]

@UselessGuru

No worries, and the benchmark completed, no failed miners. Enumeration looks good!

@FruitFace

Thank you for your feedback.

[carlo0000]

i still have the problem, when multipool just stop working, i upgrade to betta11 yesterday (fresh install, i keep only start.bat) but got stuck today, couple hours behind

update an other computer, same problem happen, computer is idle and log is stop 1 hours behind
i already restarted multipool twice , it only finish 4 benchmark in cpu and 0 in gpu

i notice both computer are stuck after a failed download
then nothing happen anymore, my connection is bit slow, have 100gigas/month and end of month, it s full used and limited to 100kb/s, but why multipool get stuck? waiting for what? why it not moving on or retry

2018-12-28 17:29:12 WARNING: Cannot download C:\cgminer_skein\multipoolminer\Bin\NVIDIA-CcminerLyra2z_v0.3.0\ccminer.exe distributed at https://github.com/djm34/ccminer-msvc2015/releases/download/v0.3.0/ccminer.rar.
2018-12-28 17:29:12 WARNING: Cannot find C:\cgminer_skein\multipoolminer\Bin\NVIDIA-CcminerLyra2z_v0.3.0\ccminer.exe distributed at https://github.com/djm34/ccminer-msvc2015/releases/download/v0.3.0/ccminer.rar.

then nothing happens anymore

it s stuck again
2018-12-28 19:41:32 INFO: Saving hash rates.
2018-12-28 19:41:32 WARNING: Stat file (CPU-XmrigCryptonight_v2.8.3-1xAuthenticAMD4CoreCPU-2_CryptonightV8_HashRate) was not updated because the value (282.416666666667) is outside fault tolerance (216 to 264).
2018-12-28 19:41:32 INFO: Starting next run.
2018-12-28 19:41:32 INFO: Mining for you. Donation run will start in 00 hour(s) 37 minute(s).
2018-12-28 19:41:33 INFO: Updating exchange rates from Coinbase.
2018-12-28 19:42:33 INFO: Loading saved statistics.
2018-12-28 19:42:33 INFO: Waiting for pool information.

the other computer is also stuck here (current time 20.25)
2018-12-28 19:40:30 INFO: Finish waiting before next run.
2018-12-28 19:40:30 INFO: Saving hash rates.
2018-12-28 19:40:30 INFO: Starting next run.
2018-12-28 19:40:30 INFO: Mining for you. Donation run will start in 00 hour(s) 37 minute(s).
2018-12-28 19:40:30 INFO: Updating exchange rates from Coinbase.
2018-12-28 19:40:30 INFO: Loading saved statistics.
2018-12-28 19:40:31 INFO: Loading pool information - this may take a minute or two.

same computer 3
2018-12-28 19:45:08 INFO: Updating exchange rates from Coinbase.
2018-12-28 19:45:18 WARNING: Coinbase is down.
2018-12-28 19:45:18 INFO: Loading saved statistics.
2018-12-28 19:45:18 INFO: Waiting for pool information.

computer4 working fine for now, doing benchmarks

but not only computers with last version , notice some computers runing beta8a are stuck too
2018-12-28 16:34:26 INFO: Updating exchange rates from Coinbase.
2018-12-28 16:34:28 INFO: Loading saved statistics.
2018-12-28 16:34:29 INFO: Loading pool information.

looks like when internet is slow it just get stuck

[UselessGuru]

why it not moving on or retry

This is what it should do. There are timeouts built in everywhere
Can you please post the log files (last 100 lines)?

Honestly - with weak internet mining is rather pointless. You will have high latency which will make your shares arrive later at the pool which will reduce your profit (unless the pool counts stale shares too).

Edit: Could this be the cause?
#2274

[carlo0000]

i don't have high latency, all miner run very fine , most share are submitted below 120ms
but it has to do with slower internet, in begin of month, never sow a problem and miners show up on https://multipoolminer.io/monitor but end of month when my connection is limited most of the miners shows up offline on the website after some time, but does not mean that multipool stop working, sometime it just works and say stats update success but still show as offline

 B2S: 12/29/18-00:32:40 - New job from blake2s.eu.mine.zpool.ca:5766
ETH: 12/29/18-00:32:40 - SHARE FOUND - (GPU 2)
ETH: Share accepted (47 ms)!
 B2S: 12/29/18-00:32:52 - SHARE FOUND - (GPU 4)
 B2S: Share accepted (125 ms)!
ETH: 12/29/18-00:32:55 - New job from europe.ethash-hub.miningpoolhub.com:20555
ETH - Total Speed: 129.049 Mh/s, Total Shares: 717, Rejected: 0, Time: 07:10

00:28:27 | GPU 1 Thread 2 Lane 360 finds a Share, value 64420
00:28:27 | Accepted by the pool in 94 ms.
00:28:36 | GPU 0 Thread 0 Lane 419 finds a Share, value 64420
00:28:36 | Accepted by the pool in 94 ms.
00:29:00 | GPU 1 Thread 2 Lane 374 finds a Share, value 64420
00:29:00 | Accepted by the pool in 78 ms.
00:29:00 | Pool sends a new Job.
00:29:44 | GPU 1 Thread 3 Lane 775 finds a Share, value 64420
00:29:44 | Accepted by the pool in 141 ms.
00:29:55 | Pool sends a new Job.

but other stuff like surfing is bit slower, actually i don't understand why its not able to get pool data, what should it be, a couple of mb max ? can't be that much

i have around 30 miners running, so they need to share 100kb/s max
when i load a youtube video i m able to watch it mostly in 480p, so not possible that it use 100kb/s for the miners

i gonna change internet provider, to one with unlimited data

carlo0000_log1.txt

[carlo0000]

have an other problem on my 470ring
it s not doing the most profitable algo, i did reset profit and restart multipool, then it was doing eth
but later i check again and it s back on monero what is less profitable, never sow this on the old version

i manually modified the stats file and change the fluctuation, now it s back on eth but i guess it s gonna update soon and change it ? i don't know why it s 52% never sow this before)

AMD-JCECryptonote_v0.33b9-4xEllesmere4GB[0,90 %] CryptonightV8 3,34 KH/s 0,00029662 100 % 89,74723215
AMD_NVIDIA-PhoenixminerEthash_v4.0a-4xEllesmere4GB-EthashBlake2s-30[0,90 %][0,00 %] {Ethash, Blake2s} {108,00 MH/s, 3,24 GH/s} 0,00039562 {52 %, 99 %} {0,00359253, 0,00000343}

ouch yeah fluctuation was changing very fast again
so i fixed the problem i just toke the old stat file what has a very long duration and notice has a fluctuation close to 0
so i guess it s only a problem when stat file just created

[UselessGuru]

@carlo0000

2018-12-28 22:58:12 INFO: Loading saved statistics.
2018-12-28 22:58:13 INFO: Waiting for pool information.
...{stop}...

You need to find out which pool file might be causing this. Please remove (e.g. move to another folder) all pool files except for one (it must be a configured pool in -poolname' of course). See if the problem remains, then switch to the next pool file and so on... Which file is causing the stop?
Also upload the last 100 lines or so of transcript file named MultiPoolMiner_YYYY-MM-DD_HH-MM-SS.txt (please upload as a file - it is much better readable)

it s not doing the most profitable algo,

Please define most profitable algo.

The most profitable algo is not necessarily the one that shows the highest value in the 'profit' column. The most profitable algo is the one that has the highest 'reliable' profit (profit * accuracy). This is to avoid algo-hopping when the price of an algo jumps to unreal highs (thereby reducing accuracy, see below).

i don't know why it s 52% never sow this before)

What do you mean by 52%? Accuracy?
Accuracy tells you how stable (over time) the received price data for an algo is. The older that stat file is, the more reliable the values are (more samples give more stable information).

Note that nothing has changed since 3.0 (or even longer) when it comes to profit calculation.

[UselessGuru]

3.2.0 Beta 12 is available: https://github.com/MultiPoolMiner/MultiPoolMiner/releases/tag/Beta3.2.0

Please report all pending & possibly new issues in this thread:
#2277

[carlo0000]

i m only using multipool and zpool

also notice this
WARNING: Coinbase is down. could be that also

i think it does time out when request the data from pool or maybe coinbase and get stuck there

i will try removing zpool and avoid computer to restart and see what happens

What do you mean by 52%? Accuracy?
Accuracy tells you how stable (over time) the received price data for an algo is. The older that stat file is, the more reliable the values are (more samples give more stable information).

i understand it s why i restore on old stats file
the new one was mining monero when it s very lower profitable than eth
because i guess it says accuracy 52%
in future i will keep the old stats, when i delete all the rest

[carlo0000]

ok so 1 pool only multipool

it get stuck also

2018-12-30 22:28:27 WARNING: Coinbase is down.
2018-12-30 22:28:27 INFO: Loading saved statistics.
2018-12-30 22:28:28 INFO: Loading pool information.

[UselessGuru]

ok so 1 pool only multipool

What is multipool??? There is no such pool name
There is MiningPoolHub ???

How about posting the log as I requested?

If MiningPoolHub is your problematic pool then please try the files in this zip file:

MiningPoolHub_PoolFiles.zip

[carlo0000]

yeah sorry i mean miningpool
it s also possible it get stuck after it try conenct to coinbase and time out

i can't try anymore it's 1 January my connection is at full speed now,
as problem only shows up when speed is slower
i also toke an other internet provider, so not sure if i can test again before i get the new one

[UselessGuru]

@carlo0000

Thank you for reporting back!
Should the problem reoccur please post in a new thread.