Author

Topic: [ mining os ] nvoc - page 235. (Read 418546 times)

hero member
Activity: 651
Merit: 501
My PGP Key: 92C7689C
September 08, 2017, 03:26:01 PM
salfeter mph gives me error

Code:
Traceback (most recent call last):
  File "/home/m1/mph_switch", line 14, in
    cfg=json.loads(open(sys.argv[1]).read())
  File "/usr/lib/python2.7/json/__init__.py", line 339, in loads
    return _default_decoder.decode(s)
  File "/usr/lib/python2.7/json/decoder.py", line 364, in decode
    obj, end = self.raw_decode(s, idx=_w(s, 0).end())
  File "/usr/lib/python2.7/json/decoder.py", line 382, in raw_decode
    raise ValueError("No JSON object could be decoded")
ValueError: No JSON object could be decoded

That's the line that loads the config file.  By the time it's run, 3main should have already generated mph_conf.json.  What do you have in that file?
hero member
Activity: 651
Merit: 501
My PGP Key: 92C7689C
September 08, 2017, 03:18:39 PM
MPH wise, we got the major ethash and equihash algos, we already have SALFTER's MPH switcher.

Only thing remaining is the zpool algos.

Someone asked me about implementing a switcher for zpool.  Took a quick look yesterday to see what information their API provides...turns out it's not much at all.  Unless they have some undocumented API calls available, there's nothing on which to base a switching decision.  I suppose there's WhatToMine, which could possibly lead to a more generalized switcher, but they appear to have somewhat limited algorithm coverage.

It's possible to calculate expected production for a coin based on your equipment's hashrate and the network's difficulty; that's basically what WhatToMine does.  If I can find a sufficiently diverse source of network difficulty figures, I might be able to knock something together.  (https://chainz.cryptoid.info/dmd/api.dws?q=summary might be a possibility; I need to compare it to what can be mined at NiceHash/MiningPoolHub/ZPool to see if it might be sufficient.  Back in the day, I was running a bunch of block explorers myself; I'd rather not go back to that if I can avoid it.)

Edit: On slightly deeper inspection, the aforementioned chainz.cryptoid.info doesn't even cover the likes of BTC or ETH, though it does have a ton of shitcoins I've never heard of.  Anybody have recommendations for a multiple-coin block explorer with an API that makes difficulty information available?
newbie
Activity: 7
Merit: 0
September 08, 2017, 02:50:42 PM
It works fantastic! Thank You! Do you happen to have a solution where this can mine Ubiq coin?
hero member
Activity: 651
Merit: 501
My PGP Key: 92C7689C
September 08, 2017, 02:01:30 PM
full member
Activity: 378
Merit: 104
nvOC forever
September 08, 2017, 10:56:00 AM
ZPOOL LBRYCOIN:


Code:
COIN="ZPOOL_LBC"

1bash:
Code:
# ZPOOL uses your BTC_ADDRESS
ZPOOL_LBC_POOL="stratum+tcp://lbc.mine.zpool.ca:3334"


3main:
Code:
if [ $COIN == "ZPOOL_LBC" ]
then
HCD='/home/m1/SPccminer/ccminer'
ZPOOLADDR="$BTC_ADDRESS.$WORKERNAME"
ZPOOLPASS="$WORKERNAME,c=BTC"

screen -dmS miner $HCD -a lbry -o $ZPOOL_LBC_POOL -u $ZPOOLADDR -p $ZPOOLPASS

if [ $LOCALorREMOTE == "LOCAL" ]
then
screen -r miner
fi

BITCOIN="theGROUND"

while [ $BITCOIN == "theGROUND" ]
do
sleep 60
done
fi

I have a question regarding this;

on ZPOOL site, it says

Code:
-o stratum+tcp://.mine.zpool.ca: -u [-p c=,]

We need to enter ALGO type, PORT related to algo, BTC Wallet Address and for the password it can be empty (I believe) and it can go like this right? '-p c=BTC'

It can be hard coded like this right?

Code:
screen -dmS miner $HCD -a lyra2v2 -o $ZPOOL_LYRA2V2_POOL -u $BTC_ADDRESS -p c=BTC

so that you only need to give the right pool embedded with right port? given we already have our BTC address?

but the above 'ZPOOL_LBC' has these:

Code:
ZPOOLADDR="$BTC_ADDRESS.$WORKERNAME"
ZPOOLPASS="$WORKERNAME,c=BTC"

&&

Code:
screen -dmS miner $HCD -a lbry -o $ZPOOL_LBC_POOL -u $ZPOOLADDR -p $ZPOOLPASS

Why can't we use it like this??

Code:
screen -dmS miner $HCD -a lbry -o $ZPOOL_LYRA2V2_POOL -u $BTC_ADDRESS -p c=BTC

Please correct me if I have missed something!?


It would be nice to implement unified zpool worker name into zpool password in the new 1bash and 3main

if use bitcoin for zpool
Code:
ZPOOLWORKER="$WORKERNAME,c=BTC"

Regarding this;

Its bit confusing, not sure if I'm missing something, why do we need worker name in the place of password? please correct me if I'm taking this completely wrong Smiley



newbie
Activity: 23
Merit: 0
September 08, 2017, 10:52:31 AM

Hello!

A few questions:

My farms are in the data center.

1. Will any remote monitoring of the rig be implemented, except SSH?

2. Is it possible to update nvoc 18 -> 19 without overwriting a new image?
full member
Activity: 686
Merit: 140
Linux FOREVER! Resistance is futile!!!
September 08, 2017, 08:36:40 AM
ZPOOL LBRYCOIN:


Code:
COIN="ZPOOL_LBC"

1bash:
Code:
# ZPOOL uses your BTC_ADDRESS
ZPOOL_LBC_POOL="stratum+tcp://lbc.mine.zpool.ca:3334"


3main:
Code:
if [ $COIN == "ZPOOL_LBC" ]
then
HCD='/home/m1/SPccminer/ccminer'
ZPOOLADDR="$BTC_ADDRESS.$WORKERNAME"
ZPOOLPASS="$WORKERNAME,c=BTC"

screen -dmS miner $HCD -a lbry -o $ZPOOL_LBC_POOL -u $ZPOOLADDR -p $ZPOOLPASS

if [ $LOCALorREMOTE == "LOCAL" ]
then
screen -r miner
fi

BITCOIN="theGROUND"

while [ $BITCOIN == "theGROUND" ]
do
sleep 60
done
fi
member
Activity: 117
Merit: 10
September 08, 2017, 07:57:09 AM
ok, I like to share that finally have a rig with nvOC 0.19 and find a problem with Claymore 9.8.
Rig has 3 1060 and 2 asus strix 970 OC.
Using Claymore 9.8 I notice that the miner may get the gpus id incorrectly.

system says (and is right)
0 gtx 970
1 1060
2 1060
3 1060
4 gtx 970

and Claymore 9.8 says (wrong)
0 1060
1 gtx 970
2 1060
3 1060
4 gtx 970  

Downloaded Claymore 10 and add the code to 1bash and 3main. It seems the bug is fixed on version 10.

The thing is that if a rig works only with a unique model of gpus they can be mix up by claymore 9.8 in the same way.
Bad thing for individual oc and settings.

I still did not manage to OC the 970's. They give me around 15Mh/s mining etc.
From windows using nicehash software they give around 19Mh/s mining daggerhashimoto.
Some says it can get around 22/23.

I would appreciate sharing oc settings + pl for this 970 cards.
And one more question:
Should the card of same model be in correlative order so like this:
0 gtx 970
1 gtx 970
2 1060
3 1060
4 1060

Instead of like this as I have then right now:
0 gtx 970
1 1060
2 1060
3 1060
4 gtx 970

Thanks  
full member
Activity: 686
Merit: 140
Linux FOREVER! Resistance is futile!!!
September 08, 2017, 06:12:46 AM
fullzero, I've noticed that even with "individual target temps" set to NO, guake shows the individual values being used. even though I want to use the target temp I identified up top, I have to adjust all the individual ones to get my intended results.

This has been already reported, fullzero said he is working on the fix. probably, the bug I've found also interlinked to this issue!!

For now you should use the first individual temp (GPU0) as your target temp, until fullzero fix it.
If you set individual to yes, they will work as they should
if set to no, the first one will be used for all.
full member
Activity: 378
Merit: 104
nvOC forever
September 08, 2017, 05:55:55 AM
fullzero, I've noticed that even with "individual target temps" set to NO, guake shows the individual values being used. even though I want to use the target temp I identified up top, I have to adjust all the individual ones to get my intended results.

This has been already reported, fullzero said he is working on the fix. probably, the bug I've found also interlinked to this issue!!
full member
Activity: 686
Merit: 140
Linux FOREVER! Resistance is futile!!!
September 08, 2017, 05:09:14 AM
salfeter mph gives me error


Code:
Traceback (most recent call last):
  File "/home/m1/mph_switch", line 14, in
    cfg=json.loads(open(sys.argv[1]).read())
  File "/usr/lib/python2.7/json/__init__.py", line 339, in loads
    return _default_decoder.decode(s)
  File "/usr/lib/python2.7/json/decoder.py", line 364, in decode
    obj, end = self.raw_decode(s, idx=_w(s, 0).end())
  File "/usr/lib/python2.7/json/decoder.py", line 382, in raw_decode
    raise ValueError("No JSON object could be decoded")
ValueError: No JSON object could be decoded


full member
Activity: 378
Merit: 104
nvOC forever
September 08, 2017, 04:56:01 AM
Any one faced the issue with POWER_LIMIT??

or Am I not using it properly!?

While using MPH_PROFIT_SWITCHER; for cryptonight algo, I had the following settings:

Code:
Cryptonight_POWERLIMIT_WATTS=60
__Cryptonight_CORE_OVERCLOCK=155
Cryptonight_MEMORY_OVERCLOCK=-300

but POWER_LIMIT as follows

Code:
POWERLIMIT="YES"
POWERLIMIT_WATTS=78

I've enabled 'Maxximus auto Temp', so when it starts mining XMR based on profitability; it changes the temp limit to '60' as expected but in the next minute 'Maxximus auto Temp' changing it back to '78'

I've tried this too, but still same:

Code:
POWERLIMIT="NO"
POWERLIMIT_WATTS=78

I do remember some one made a note of issue with POWER_LIMIT (its using individual power limit, not the global power_limit value).

Any one experienced this issue?? or any mistake in the way I'm using it currently?


Yeap i have this issue to Wink When i test to tweak my setup, i assume that from the switching value of templimit.
I was minining ETH on 125w per GPU and want to do to some test with lower TDP, and i do nvidia-smi -pl 110 for example.

What i have observed, it seems my system dont like this aggressive reduction of power limit and if my GPU ask for more, linux will upper the limit to the last state.
I was not able to get log from this ... Is like no log for this action ... Maybe i miss something

I think Maxximus007_AUTO_TEMPERATURE_CONTROL is using the below variable if something goes wrong!

Quote
TARGET_TEMP=70

__FAN_ADJUST=5      # Adjustment size in percent

POWER_ADJUST=2      # Adjustment size in watts

# Difference in actual temperature allowed before action:  Works only if current is BELOW target temp

ALLOWED_TEMP_DIFF=3

# Restore original power limit if fan speed is lower than this percentage

RESTORE_POWER_LIMIT=120

# lowest fan speed that will be used

MINIMAL_FAN_SPEED=20



I remember fullzero said he will look into it when it was reported last time (issue with individual power limits and global power limit)
full member
Activity: 420
Merit: 106
https://steemit.com/@bibi187
September 08, 2017, 04:44:00 AM
Any one faced the issue with POWER_LIMIT??

or Am I not using it properly!?

While using MPH_PROFIT_SWITCHER; for cryptonight algo, I had the following settings:

Code:
Cryptonight_POWERLIMIT_WATTS=60
__Cryptonight_CORE_OVERCLOCK=155
Cryptonight_MEMORY_OVERCLOCK=-300

but POWER_LIMIT as follows

Code:
POWERLIMIT="YES"
POWERLIMIT_WATTS=78

I've enabled 'Maxximus auto Temp', so when it starts mining XMR based on profitability; it changes the temp limit to '60' as expected but in the next minute 'Maxximus auto Temp' changing it back to '78'

I've tried this too, but still same:

Code:
POWERLIMIT="NO"
POWERLIMIT_WATTS=78

I do remember some one made a note of issue with POWER_LIMIT (its using individual power limit, not the global power_limit value).

Any one experienced this issue?? or any mistake in the way I'm using it currently?


Yeap i have this issue to Wink When i test to tweak my setup, i assume that from the switching value of templimit.
I was minining ETH on 125w per GPU and want to do to some test with lower TDP, and i do nvidia-smi -pl 110 for example.

What i have observed, it seems my system dont like this aggressive reduction of power limit and if my GPU ask for more, linux will upper the limit to the last state.
I was not able to get log from this ... Is like no log for this action ... Maybe i miss something
full member
Activity: 362
Merit: 102
September 08, 2017, 04:00:58 AM

try to lower the cc around 100, mc may go till 1700 (only if its samsung), why negative dcri?? what is the power utilization %??
[/quote]

I think it wasn't stable when I went under 150 and upped the memory, I can have another try though. Anyways, I may have just written DCRI incorrectly, but I have it at 400-450 which seems to be optimal for ETH/LBC dual mining profitability. I have the power limit at 300w.
full member
Activity: 378
Merit: 104
nvOC forever
September 08, 2017, 03:37:53 AM
Any one faced the issue with POWER_LIMIT??

or Am I not using it properly!?

While using MPH_PROFIT_SWITCHER; for cryptonight algo, I had the following settings:

Code:
Cryptonight_POWERLIMIT_WATTS=60
__Cryptonight_CORE_OVERCLOCK=155
Cryptonight_MEMORY_OVERCLOCK=-300

but POWER_LIMIT as follows

Code:
POWERLIMIT="YES"
POWERLIMIT_WATTS=78

I've enabled 'Maxximus auto Temp', so when it starts mining XMR based on profitability; it changes the temp limit to '60' as expected but in the next minute 'Maxximus auto Temp' changing it back to '78'

I've tried this too, but still same:

Code:
POWERLIMIT="NO"
POWERLIMIT_WATTS=78

I do remember some one made a note of issue with POWER_LIMIT (its using individual power limit, not the global power_limit value).

Any one experienced this issue?? or any mistake in the way I'm using it currently?
full member
Activity: 378
Merit: 104
nvOC forever
September 08, 2017, 03:16:38 AM
Just wanted to share my current settings after spending the last day tweaking. Still probably could maybe optimize more, but this is how far I've gotten so far.

1080ti mining eth/lbry

cc: 150
mc: 1500
dcri: -500

eth: 27 Mh/s
lbry: 294 Mh/s

Currently comes out to just a touch under $3 per day of profit on my rig spending .11/kWh. Sort of underwhelming, but still higher than mining ETH or LBRY alone. Any suggestions or tips are welcome. (These settings are not stable.)

try to lower the cc around 100, mc may go till 1700 (only if its samsung), why negative dcri?? what is the power utilization %??
full member
Activity: 362
Merit: 102
September 08, 2017, 02:16:32 AM
Just wanted to share my current settings after spending the last day tweaking. Still probably could maybe optimize more, but this is how far I've gotten so far.

1080ti mining eth/lbry

cc: 150
mc: 1500
dcri: -500

eth: 27 Mh/s
lbry: 294 Mh/s

Currently comes out to just a touch under $3 per day of profit on my rig spending .11/kWh. Sort of underwhelming, but still higher than mining ETH or LBRY alone. Any suggestions or tips are welcome. (These settings are not stable.)
newbie
Activity: 11
Merit: 0
September 08, 2017, 01:39:46 AM
in 3main
Code:
pkill -e miner
I know it is used to kill the miner process, but what does -e mean? Cannot find this argument in
Code:
man pkill

-e echoes the processes that were killed
newbie
Activity: 4
Merit: 0
September 08, 2017, 12:12:07 AM
Hi
Help or assist to understand with a problem.
when starting this error
/ home / m1 / 3main: line 53: nvidia-smi: command not found


 rig IP: 192.168.1.3


rig MAC: 70: 85: c2: 49: 78: 44


01: 00.0 VGA compatible controller: NVIDIA Corporation Device 1b06 (rev a1)
07: 00.0 VGA compatible controller: NVIDIA Corporation Device 1c03 (rev a1)
08: 00.0 VGA compatible controller: NVIDIA Corporation Device 1c03 (rev a1)


/ home / m1 / 3main: line 144: nvidia-smi: command not found



Thu Sep 7 13:54:01 EDT 2017
Action: Installing daemon (12.0.76279) for 'systemd' ...
installing /etc/systemd/system/teamviewerd.service (/opt/teamviewer/tv_bin/script/teamviewerd.service)
Try: systemctl enable teamviewerd.service
systemctl start teamviewerd.service

[WARN] Unable to load the library libutempter!
[WARN] The command will not work in guake!
[WARN] libutempter.so.0: can not open shared object file: No such file or directory

LAUNCHING: IAmNotAJeep_and_Maxximus007_WATCHDOG

process in screen wdog; attach with: screen -r wdog


LAUNCHING: Maxximus007_AUTO_TEMPERATURE_CONTROL

process in screen temp; attach with: screen -r temp

/ home / m1 / 3main: line 450: [: NO: integer expression expected
/ home / m1 / 3main: line 455: nvidia-smi: command not found

ERROR: Error querying enabled displays on GPU 0 (Missing
       Extension).


ERROR: Error querying connected displays on GPU 0 (Missing
       Extension).



ERROR: Error resolving target specification 'gpu: 0' (No
       targets match target specification), specified in
       assignment '[gpu: 0] / GPUGraphicsClockOffset [3] = 50'.


ERROR: Error querying enabled displays on GPU 0 (Missing
       Extension).


ERROR: Error querying connected displays on GPU 0 (Missing
       Extension).



ERROR: Error resolving target specification 'gpu: 0' (No
       targets match target specification), specified in
       assignment
       '[gpu: 0] / GPUMemoryTransferRateOffset [3] = 200'.

/ home / m1 / 3main: line 877: [: ==: unary operator expected
/ home / m1 / 3main: line 883: [: ==: unary operator expected
/ home / m1 / 3main: line 891: [: ==: unary operator expected
/ home / m1 / 3main: line 901: [: ==: unary operator expected
/ home / m1 / 3main: line 913: [: ==: unary operator expected
/ home / m1 / 3main: line 927: [: ==: unary operator expected
/ home / m1 / 3main: line 943: [: ==: unary operator expected
/ home / m1 / 3main: line 961: [: ==: unary operator expected
/ home / m1 / 3main: line 981: [: ==: unary operator expected
/ home / m1 / 3main: line 1003: [: ==: unary operator expected
/ home / m1 / 3main: line 1027: [: ==: unary operator expected
/ home / m1 / 3main: line 1053: [: ==: unary operator expected
/ home / m1 / 3main: line 1081: [: ==: unary operator expected

LAUNCHING: Mining Process

process in screen miner; attach with: screen -r miner
full member
Activity: 200
Merit: 101
September 07, 2017, 09:08:43 PM
Has anyone tried using Firefox in nvOC while mining? It is painfuly slow even on my rigs with i7 cpu's. Same thing happens in v18 and v19. If I kill the mining process Firefox works well. Strange... fullzero, any idea what might cause firefox to crawl like that?
Completely normal, GPU is under full pressure and cant render pages.

I surf all the time on all of my rigs with no slow down at all but Im running Windows...



exactly, no slowdown at all on windows. i wonder why there is such a slowdown on linux... usually linux handles everything better than windows.... there must be something wrong here... some setting or library not configured properly... linux should be able to alocate resources as needed...

on a side note.... if i ctrl-c the watchdog, maximus temp control and miner, the system hangs, i have to hard reset
... hrmm..Huh
Jump to: