minershive/hiveos-linux

Hive OS client for GPU rigs

Hive OS client for GPU rigs

Notes:

1. Since v0.6-201 source codes no longer updated here
2. For bug reports and feature requests please write to [email protected]

hiveos-linux's Issues

AMD Memory Tweak Tool

https://bitcointalk.org/index.php?topic=5123724.0

Thought I would share. Looks useful, though not to the average user.

GTX 1070 memory and fan speed settings not changing

My rig of 1070 and 1070 TI cards are not changing to the OC settings
only power limit settings are changed

Update delay bug

I am experiencing update bug with last few HIVE versions. Few days after Telegram notification I am unable to update. The update button is highglighted but after depressing it shows "you have the latest version" and do nothing. In two to five days the rig updates without any changes from my side. It is a bit annoying just before today's Monero fork.
Please look into it.

Custom Miner Download On Bleeding Edge / Vega

It seems to be an SSL error. I figured out if I add:

if [ -f /usr/lib/x86_64-linux-gnu/libcurl-compat.so.3.0.0 ]; then
    echo "Exporting Libcurl"
    export LD_LIBRARY_PATH=/usr/lib/x86_64-linux-gnu
fi

It fixed on the custom miner h-run.sh side, and wget works. But users can't seem to get to the point to download a custom miner first.

Bminer 15.0.0 listed on github but not available on flight sheet configuration

Bminer 15.0.0 listed on github but not available on flight sheet configuration. are we just waiting for server upgrade? how can i use bminer 15 that is shown on github?

no teleconsole link after teleconsole restart

Monero Algo OC

I am mining monero but when I check the overclock setting on Hive 2, it is using the cryptonight profile but nit the v7 profile.

added ethminer v0.19.0 please

https://github.com/ethereum-mining/ethminer/releases/download/v0.19.0-alpha.0/ethminer-0.19.0-alpha.0-cuda-9-linux-x86_64.tar.gz

Special characters may break wifi setup

If you have a special character in your wifi password the wifi config generation script may break because of sed use of special characters.

For instance, if you have a wifi password like FOO&BAR the substitution script at https://github.com/minershive/hiveos-linux/blob/master/hive/sbin/wifi#L131

will generate a bad wifi.txt.

WIFI_SSID="..."

WIFI_PASSWORD="FOOWIFI_PASSWORD=""BAR"

(template: https://github.com/minershive/hiveos-linux/blob/master/hive/etc/hive-config.stub/network/wifi.txt)

The same may apply to the WIFI_SSID substitution.

Thank you

HiveOS not showing proper CPU temperatures

System is an HP DL 580 G7. HiveOS shows 8.3 degrees for all processor cores. This is a bug in HiveOS.

Installing and running lm-sensors shows the correct processor core temperatures ranging from 59 to 75 degrees.

HiveOS is picking this wrong sensor for all the processor cores:

acpitz-virtual-0
Adapter: Virtual device
temp1: +8.3°C (crit = +31.3°C)

That is the BUG. It should be using these sensors:

coretemp-isa-0002
Adapter: ISA adapter
Core 0: +59.0°C (high = +79.0°C, crit = +95.0°C)
Core 1: +55.0°C (high = +79.0°C, crit = +95.0°C)
Core 2: +59.0°C (high = +79.0°C, crit = +95.0°C)
Core 8: +63.0°C (high = +79.0°C, crit = +95.0°C)
Core 17: +61.0°C (high = +79.0°C, crit = +95.0°C)
Core 18: +62.0°C (high = +79.0°C, crit = +95.0°C)
Core 24: +64.0°C (high = +79.0°C, crit = +95.0°C)
Core 25: +59.0°C (high = +79.0°C, crit = +95.0°C)

coretemp-isa-0000
Adapter: ISA adapter

coretemp-isa-0003
Adapter: ISA adapter

coretemp-isa-0001
Adapter: ISA adapter

Please fix

/hive/bin/message: line 46: echo: write error: Broken pipe

There were some GPUs falling from bus. Hashrate falls under the watchdog limits. Watchdog barked and then:

Mar 6 11:30:09 ch1 hive-watchdog[1761]: BARK ethminer 0 kHs < 229000 kHs for 282 seconds
Mar 6 11:30:19 ch1 hive-watchdog[1761]: BARK ethminer 0 kHs < 229000 kHs for 292 seconds
Mar 6 11:30:29 ch1 hive-watchdog[1761]: BARK ethminer 0 kHs < 229000 kHs for 302 seconds
Mar 6 11:30:29 ch1 hive-watchdog[1761]: 2
Mar 6 11:30:29 ch1 hive-watchdog[1761]: /hive/bin/message: line 46: echo: write error: Broken pipe
Mar 6 11:30:29 ch1 hive-watchdog[1761]: > Sending "warning": #033[0;33methminer low hashrate (0 kH/s), restarting#033[0m
Mar 6 11:30:29 ch1 hive-watchdog[1761]: { "method": "message", "jsonrpc": "2.0", "id": 0, "params": { "rig_id": "307602", "passwd": "***", "type": "warning", "data": "ethminer low hashrate (0 kH/s), restarting" } }
Mar 6 11:30:31 ch1 hive-watchdog[1761]: parse error: Invalid numeric literal at line 1, column 10
Mar 6 11:30:31 ch1 hive-watchdog[1761]: parse error: Invalid numeric literal at line 1, column 10
Mar 6 11:30:31 ch1 hive-watchdog[1761]: ---
Mar 6 11:30:31 ch1 hive-watchdog[1761]: Restarting ethminer after 5 minutes
Mar 6 11:30:31 ch1 hive-watchdog[1761]: ---
Mar 6 11:30:31 ch1 hive-watchdog[1761]: #033[0;33mRestarting miner#033[0m
Mar 6 11:30:31 ch1 hive-watchdog[1761]: Stopping screen session 25790
Mar 6 11:30:32 ch1 hive-watchdog[1761]: Starting #033[0;36methminer#033[0m, #033[0;36mcustom#033[0m

Ethminer update 0.15rc2

Please update Hive's Ethminer. There is a release candidate version available from dev's Github

https://github.com/ethereum-mining/ethminer/releases

Keep pushing! 👍

ATIFLASH support for 10+ GPUs

Current version of ATIFLASH does not work with more than 10 GPUs.
Here is fixed version:
atiflashfixed.zip
Atiflash v4.25.1 patched by me
commands -p -s -i -ai now work with 16 GPUs using A B C D E F for 10+ GPU.

bminer - bind monitoring port to all interfaces

Hello.
Can you please change binding interface to 0.0.0.0.
It will allow external apps to access api in lan.

Agent stop working

Sometimes agent stop working after reboot (it hangs in gpu-stats script on getting data from pwm1 or temp_input1 of some AMD cards). It looks like this

Run gpu-stats with timeout function does not solve this problem. Watchdogs does not help also. HW watchdog works and does not reboot rig. Hashrate watchdog does not work as miner does not start.

I can suggest 3 ways to fix this and other similar problems.

1. Ping HW watchdog from agent. (Does not help if there is no watchdog)

2. Check from cron if agent is working. For example, if last update time is more than 1 minute - restart agent. If more than 3 minutes - reboot rig. (Agent must update something before sleeping)

3. Run gpu-stats in separate screen. It can also help with situation when nvidia-smi needs more time to get statistics (up to 10 seconds per gpu with high load average). At this moment if nvidia-smi does not get data in 10 seconds - there will no temps & fans
   

Add RTX 3000 series memory temprature!

Hello,
HWInfo has it please add it too!

Thank you

Custom Miner: SSL missing

I updated to latest release, and noticed that ssl dependencies was missing.

For me adding to my custom miner h-run.sh:

export LD_LIBRARY_PATH=/usr/lib/x86_64-linux-gnu:/hive/lib

Fixes, but before doing so, it was having issues connecting to websites with https:// such as miners that connect to pools for API/Profit information. May affect non-custom miners as well.

hive-help (enhancement)

Maybe there is one that I don't know about, but I propose a hive-help bash script being made for users to see all hive terminal commands, and what they do. It may be cool to add basic and extremely useful linux commands as well for total newbies to linux.

I personally am having to constantly pass users forum posts of hive command lists. It seems it would just be easier to tell them to type hive-help. I may be writing this not knowing there is one already...if there is...I can't find it.

There are a lot of commands at this point. Even I can't remember them all. Haha- Not complaining.

Question on watchdog rebooting

When a GPU starts flaking out, the watchdog service running on HiveOS seems to want to reboot the entire system. Isn't there something else that can be done; perhaps disable the problem GPU and re-enable it?

Here is an example:

WATCHDOG: GPU 0 hangs in OpenCL call, exit
WATCHDOG: GPU 0 hangs in OpenCL call, exit
Rebooting

Sending with payload "danger": Claymore Reboot: WATCHDOG: GPU 0 hangs in OpenCL call, exit
{"method":"message","jsonrpc":"2.0","id":0,"params":{"rig_id":"xxxxxx","passwd":"**********","type":"danger","data":"Claymore Reboot: WATCHDOG: GPU 0 hangs in OpenCL call, exit\r","payload":"[3305 bytes]"}}

Thanks

Custom-get correction

Hello!
Find some error in custom-get file:
Due to the fact that the archive file contains the full name (name-version) and not just the name need replace:
if [[ -d /hive/custom/$miner ]]; then to
if [[ -d /hive/custom/$basename ]]; then

rm -rf /hive/custom/$miner to
rm -rf /hive/custom/$basename
and
chown -R user /hive/custom/$miner to
chown -R user /hive/custom/$basename

The following argument was not expected: -HWMON

When ETH miner is used in progpow version, the miner exits with the following message.
The following argument was not expected: -HWMON

Git commits

Здравствуйте, господа! Мне нравится ваша ОС, но мне не нравится то как вы ведете историю коммитов. Пролистав страницы я увидел коммиты с одинаковым названием, которые дублируются 2, а иногда даже 3 раза, ну это такая придирка конечно, это ВАШ проект и ВАМ решать как использовать функционал гита, но я бы на вашем месте сквошил такие коммиты (git rebase -i HEAD~n), ну и возможно добавлял бы к названию коммита префикс что было реализовано ([Fix], [Feature], [Release]), я думаю вам стало бы проще ориентироваться. Успехов в разработке!

P.S. Ну а вообще гитхаб не очень удобный как по мне в разработке, мне больше нравится gitlab, а то я даже хз как лейбл повесить)

xmrig unified miner config not working

trying to make a flight sheet for xmrig unified miner and i get "the given data was invalid"

what's the hardware requirement?

What's the hardware requirement? Need GPU with 4GB VRam? Can CPU or low end CPU(mybe 1GB VRam) mining?

GPU driver error no temps

Hi,

in light of this forum thread https://forum.hiveos.farm/t/3080-gpu-driver-error-no-temps/29827

There seems to be an unresolvable problem right now where the GPU would randomly throw a GPU driver error on mostly RTX 30 series cards.
Users already tried several things such as lowering OC, replacing thermal pads, etc. without success.

From my own experience the error appears more frequently on latest nvidia driver but still very often on 455.45.01.
Hashrate Watchdog also seems to be unreliable as it doesn't always reboot HiveOS leading to several hours of idling away.

Any guidance on this issue?
Thank you!

No way to update wallet apikey setting if I created wallet first

Creating wallet and LATER apikey, wallet can't see new apikey to select and update with it.
But if I create a new wallet now (after apikey is created) I can select this apikey.

Version: 0.6-193@210118. 5.4.80-hiveos

bminer doesn't reports hashrate

Hello.
When using bminer, hashrate is not shown next to each GPU and it's not reported via API.

Cortex Miner

Please add support for Cortex Miner (https://github.com/CortexFoundation/Cortex_Release/blob/master/cortex-miner/cortex_miner.tar.gz). Don't have access to the repo.
Thanks =)

srrv2 not sending slot #

so with the 55 update you added srrv2 extension board support but your code is flaw and dont send the slot #

i was able to take the original keepalive from smos renamed it and put it in place of your srrv2 file it work flawless

Cortex - Herominers : URL typo issue

Check herominers URLs to Cortex (CTXC), have typo, missing c:

Are i386 packages required?

I have took a look inside the OS itself and find there are A LOTS of duplicated i386/amd64 packages and I'm pretty sure they aren't required in order to make it works. This could shrink the OS a lot and make it more faster, saving precious power on CPU.

Am I wrong?

Gminer 1.28

Please add support for gminer 1.28 which fixes equihash 150/5 mining on nicehash.

CPU mining - multiple threads visual glitch

Hi.
Since RandomX is coming I'm preparing my setup.
With multiple threads, much more than possible GPUs attached, there is a visual glitch in dashboard.

RX 550 (Lexa/Polaris12 NOT Baffin 550) not flashable with atiflash

Hi,

I can flash every card except my Lexa Pro cards (polaris12) (it wont even detect them "No adapter found"). The Windows version can detect and flash them but I'd like to do this on Linux with your 10+ card flasher. Any way you can implement this?
Thank you.

Kind Regards
MoneroCrusher

hello fails when interface has more than one ip address

Fix
/hive/bin/hello
@@ -56,7 +56,7 @@
-lan_address=ip -o -f inet addr show | grep eth0 | awk '/scope global/ {print $4}'
+lan_address=ip -o -f inet addr show | grep eth0 | awk 'BEGIN {ORS=" "} /scope global/ {print $4}'

report_rejected_per_gpu not working for t-rex 0.20.0

It seems the summary JSON format has changed in t-rex 0.20.0 (API version 3.6).
rejected_count has moved from stat_by_gpu to gpus inside the shares dictionary.
As a result, "report_rejected_per_gpu": true in the t-rex config no longer works.

h-stats.sh should be updated to check for API version >= 3.6 and look in gpus inside the shares dictionary instead of stat_by_gpu.

See https://github.com/minershive/hiveos-linux/blob/master/hive/miners/t-rex/h-stats.sh.

If current temp > target temp and speed = max speed, decrease power to lower temps.

In nvOC and forks, the auto fan/temp control script will lower the PL of the card to ensure the target temp is accomplished if the fans are at their max setting and the temperate remains above, I believe this is a very useful feature and was wondering if it could be implemented into the autofan script. I believe in the standalone version it already exists per Steambot

Claymore 14.6

Claymore 14.6 Beta is now available for Linux

Feature Suggestion: AutoOC

Would love a feature that suggests OC profiles for the detected cards and algorithm. That would be amazing, especially for workers that have mixed cards. Could be a huge value proposition for paid accounts. MVP could be based off of what the majority of other HiveOS users are doing (use at your own discretion, etc).

Vega and ROCm

ROCm/ROCm#463

I don't know if this may interest you, but if I am reading this correctly, it looks like as of 10 days ago overclocking for Vega's were added to ROCm kernels, if overdrive is enabled.

It allows full pp_table adjustments, which is pretty critical it seems, as I have purchased some, and have been playing with them in linux.

Huge pages size for RandomX

Hi.
Currently vm.nr_hugepages is set to 128.
For RandomX it needs to be much bigger, something around 2400.
I have two Xeon with 4 channel memory.

Disable P2 mode for CUDA (prevent instability)

Hi!

I was looking for a fix to "Force P2 state for CUDA" as we have in Windows (through nVidia Inspector). After some searches, googles, hair loss, I found it: ethereum-mining/ethminer#456 (comment)

P2 is the cause of many instabilities. I think the fix could be tested.

There is three procedures explained in the link provided:
1- Disable P2 by setting P1 as default.
nvidia-settings -a [gpu:0]/GPUPowerMizerMode=1

2- Set Compute Mode to NVIdia cards (I don't have sure that it helps)
nvidia-smi --gom=COMPUTE

3- Set PERFORMANCE clocks (I think it prevents clock fluctuation)
nvidia-smi -q -d PERFORMANCE

Could you take a look at this? I think if it's correct we could improve stability and performance.

Thank you!

missing settings for xmrig

Hi,
I'm currently struggling to enable randomx.1gb-pages. I tried to configure it under flightsheet > miner config > extra config args, but that only resulted in error in config.json. So I tried to directly change that config through Hive Shell, but after restarting the miner all settings are reverted(looks like settings from my flightsheet are applied). I also had some issues changing cpu config. All changes in CPU config aldo resulted in error in config.json. Even pasting in the example resulted in same error.

w83627ehf module

Hi,

Is it possible to add w83627ehf module for hardware monitoring chip?

Hashrate Watchdog Algos mode

When the Hashrate Watchdog mode is set to Algos, only the first miner using each algorithm is used to calculate the algorithm's current hashrate.

Example:
ethash minimum set to 300 mH/s
gminer (ethash) - 259.77 mH/s, teamredminer (ethash) - 56.86 mH/s, total ethash 316.63 mH/s
ethash low hashrate, rebooting

sometimes miner does not restart (claymore and lot of gpus)

Miner does not restart after wallet change or restart command when using Claymore and 13x GPU. 13x NVIDIA - Always. 13x AMD - sometimes.
Stop miner, then restart - works! Restart miner, one more restart - works. Apply new wallet (or change settings in wallet) - does not work. Need restart after. Seems happen only with claymore. But it is possible with others.

There is "claymore is already running" in miner log after restart. So it seems that during restart screen is closed but claymore is not. And then it does not start as it is still running.

Some checks need to be added to be sure miner is closed.

Update noncerpro to 3.3.0 version

https://github.com/NoncerPro/noncerpro-nimiq-cuda/releases/tag/v3.3.0

Does not show CPU Temp on old motherboard

Hello

Motherboard P5LD2-SE ASUSTeK Computer INC. (1301 08/16/2007)
CPU 2 × Intel(R) Pentium(R) Dual CPU E2140 @ 1.60GHz
Kernel 5.4.80-hiveos
OS VERSION 0.6-203@210403 (today upgraded from hiveos-0.6-190-stable@210108)
CPU TEMP 0°

Temperature from CPU always zero degrees after update!
Temperarure from GPU show without problem!

sensors output:
atk0110-acpi-0
Adapter: ACPI interface
Vcore Voltage: +1.24 V (min = +0.85 V, max = +1.60 V)
+3.3 Voltage: +3.33 V (min = +2.97 V, max = +3.63 V)
+5 Voltage: +5.02 V (min = +4.50 V, max = +5.50 V)
+12 Voltage: +11.98 V (min = +10.20 V, max = +13.80 V)
CPU FAN Speed: 2057 RPM (min = 600 RPM, max = 7200 RPM)
CHASSIS FAN Speed: 0 RPM (min = 800 RPM, max = 7200 RPM)
POWER FAN Speed: 0 RPM (min = 800 RPM, max = 7200 RPM)
CPU Temperature: +47.0°C (high = +60.0°C, crit = +95.0°C)
MB Temperature: +36.0°C (high = +45.0°C, crit = +95.0°C)

coretemp-isa-0000
Adapter: ISA adapter
Core 0: +52.0°C (high = +86.0°C, crit = +100.0°C)
Core 1: +52.0°C (high = +86.0°C, crit = +100.0°C)

What I can do with it or how fix it?

EWBF 144_5 BTG User parameters in the flight sheet are not accepted for installation

When write "pers BgoldPoW" not working

Autofan Nvidia 0°C issue

Looks like some GPU has fallen off the bus. Due to some error, autofan does think that GPU temps are 0°C and begins to slow fans down! Indeed temps gone up to 80°C. Thanks god I was near the farm.

2018-10-20 20:21:26
/hive/sbin/autofan: line 159: [[: 01:00.0: syntax error in expression (error token is ":00.0")                                                                                                              
/hive/sbin/autofan: line 165: [[: 01:00.0: syntax error in expression (error token is ":00.0")                                                                                                              
/hive/sbin/autofan: line 177: [[: 01:00.0: syntax error in expression (error token is ":00.0")                                                                                                              
/hive/sbin/autofan: line 159: [[: 02:00.0: syntax error in expression (error token is ":00.0")                                                                                                              
/hive/sbin/autofan: line 165: [[: 02:00.0: syntax error in expression (error token is ":00.0")                                                                                                              
/hive/sbin/autofan: line 177: [[: 02:00.0: syntax error in expression (error token is ":00.0")                                                                                                              
/hive/sbin/autofan: line 159: [[: 03:00.0: syntax error in expression (error token is ":00.0")                                                                                                              
/hive/sbin/autofan: line 165: [[: 03:00.0: syntax error in expression (error token is ":00.0")                                                                                                              
/hive/sbin/autofan: line 177: [[: 03:00.0: syntax error in expression (error token is ":00.0")                                                                                                              
/hive/sbin/autofan: line 159: [[: 05:00.0: syntax error in expression (error token is ":00.0")                                                                                                              
/hive/sbin/autofan: line 165: [[: 05:00.0: syntax error in expression (error token is ":00.0")                                                                                                              
/hive/sbin/autofan: line 177: [[: 05:00.0: syntax error in expression (error token is ":00.0")                                                                                                              
/hive/sbin/autofan: line 159: [[: 06:00.0: syntax error in expression (error token is ":00.0")                                                                                                              
/hive/sbin/autofan: line 165: [[: 06:00.0: syntax error in expression (error token is ":00.0")                                                                                                              
/hive/sbin/autofan: line 177: [[: 06:00.0: syntax error in expression (error token is ":00.0")                                                                                                              
/hive/sbin/autofan: line 159: [[: 07:00.0: syntax error in expression (error token is ":00.0")                                                                                                              
/hive/sbin/autofan: line 165: [[: 07:00.0: syntax error in expression (error token is ":00.0")                                                                                                              
/hive/sbin/autofan: line 177: [[: 07:00.0: syntax error in expression (error token is ":00.0")                                                                                                              
/hive/sbin/autofan: line 159: [[: 0a: value too great for base (error token is "0a")                                                                                                                        
/hive/sbin/autofan: line 165: [[: 0a: value too great for base (error token is "0a")                                                                                                                        
/hive/sbin/autofan: line 177: [[: 0a: value too great for base (error token is "0a")                                                                                                                        
/hive/sbin/autofan: line 159: [[: 0b: value too great for base (error token is "0b")                                                                                                                        
/hive/sbin/autofan: line 165: [[: 0b: value too great for base (error token is "0b")                                                                                                                        
/hive/sbin/autofan: line 177: [[: 0b: value too great for base (error token is "0b")                                                                                                                        
/hive/sbin/autofan: line 159: [[: 0c: value too great for base (error token is "0c")                                                                                                                        
/hive/sbin/autofan: line 165: [[: 0c: value too great for base (error token is "0c")                                                                                                                        
/hive/sbin/autofan: line 177: [[: 0c: value too great for base (error token is "0c")                                                                                                                        
/hive/sbin/autofan: line 159: [[: 0d: value too great for base (error token is "0d")                                                                                                                        
/hive/sbin/autofan: line 165: [[: 0d: value too great for base (error token is "0d")                                                                                                                        
/hive/sbin/autofan: line 177: [[: 0d: value too great for base (error token is "0d")                                                                                                                        
/hive/sbin/autofan: line 159: [[: 0f: value too great for base (error token is "0f")                                                                                                                        
/hive/sbin/autofan: line 165: [[: 0f: value too great for base (error token is "0f")                                                                                                                        
/hive/sbin/autofan: line 177: [[: 0f: value too great for base (error token is "0f")                                                                                                                        
GPU      [Nvidia]: 0°C 22% ↑
GPU      [Nvidia]: 0°C 22% ↑
GPU      [Nvidia]: 0°C 22% ↑
GPU      [Nvidia]: 0°C 22% ↑
GPU      [Nvidia]: 0°C 22% ↑
GPU      [Nvidia]: 0°C 22% ↑
GPU      [Nvidia]: 0°C 22% ↑
GPU      [Nvidia]: 0°C 22% ↑
GPU      [Nvidia]: 0°C 22% ↑
GPU      [Nvidia]: 0°C 22% ↑
GPU      [Nvidia]: 0°C 22% ↑