Topic: Hiveon. №1 OS for mining - page 18. (Read 80759 times)

I'm not using their pool, i'm mining to miningpoolhub and didn't see any difference in the hashrate there but i'm not sure if MPH registers 1hr cut or does it has to be longer to be noticed there.
Would be nice that the @OP could comment?
My friend had same cut at same time.

There was again 15min cut 10 mins ago

Thanks for your response. Yeah it lasted about an hour. Between 7pm and 8pm EST of January the 15th.
I just noticed that all of my machines were marked offline but I was still able to shell into them during this time and everything appeared to be mining still. I didn't take any chances in case there was a hack or something (like happened to Nicehash several years ago) so I  stopped the mining process leaving the machines running.
It sounds like you were effected too the only difference is that you continued to mine to their pool and probably won't get paid for it if it's showing the 1 hour cut to your hash rate. I stopped almost immediately after noticing it.

Not sure if the HiveOs servers were down, i have these cuts on my hash rate, there was 1hr cut about 24h ago.
I think it still continues to mine..

0.6-193@210115 🐧

What's new?

SYSTEM 🖥
⚙️ Improved overclocking AMD Vega10 family GPUs: Vega 56/64/FE (reduced power consumption; improved stability and allows you to reach higher memory clocks; fixed incorrect overclocking in some cases for Vega 64 and Vega FE)
⚙️ Fixed nvidia-oc (do not apply OC on miner start if no delay is set; apply OC on miner start if delay < 10s is set, but without this delay)
⚙️ Improved sreboot (improved disk remount which useful for slow flash drives)
⚙️ Updated CUDA 11.1 RT libs from CUDA Toolkit v11.1.1 (build 105)

MINERS👷‍♂️
⛏ PhoenixMiner v5.5b (updated kernels for AMD Polaris, Vega and Navi GPUs that are slightly faster and use less power than before when mining ETH automatically set -ttli instead of -tmax when the later is not supported by the driver. This will throttle down the GPUs when they reach the specified temperature to avoid overheating; added native kernels for AMD RX6800 and RX6900 GPUs. These are faster than the generic kernels and produce a lot less stale shares )
⛏ Gminer v2.42 (fixed performance regression for Ethash/Etchash algorithm on Nvidia miding edition cards; added auto-tune for Ethash/Etchash algorithm on Nvidia cards, miner automatically select fastest kernel, also you can select specific kernel manually by using --oc flag, currently 3 kernels available, 0 means auto-tune by miner; added display of shares per minute; added option --report_interval to control hashrate report interval in seconds; removed cuckarooz29 algorithm due Grin will not support it after hardfork)
⛏ XMRig v6.7.2 (fixed broken since v6.7.0 solo mining)
⛏ XLArig v5.2.1 fork XMRig (sync with XMRig v6.7.0 upstream)
⛏ NanoMiner v3.1.5 (updated AMD Navi GPU Ethash kernels, new performance tuning applied)
⛏ SRBMiner-Multi v0.6.3 (added algorithm 'phi5' (Combode) for GPU mining, fee 0.85%; added algorithm 'yespowertide' (Tidecoin) for CPU mining, fee 0.85%; fixed 'cryptonight_xhv', now it works on coins other than Haven; performance increase on 'phi5' algorithm on CPU; bug fixes)
⛏ Bminer v16.4.5 (fixed compatibility issues with some Etherum mining pools)
⚙️ lolMiner: fixed console coloring

0.6-192@210111 🐧

What's new?

SYSTEM 🖥
🐞 Fixed missed symlink to cudart 11.2 lib
⚙️ Added initial version of installation helper for 3G/4G modem (currently supports only Huawei E3372; could be found in /hive/opt/3g4g-modem)

MINERS👷‍♂️
⛏ T-Rex v0.19.7 (improved stability for 30xx series GPUs on ethash/octopus/kawpow; verify OC stability on ethash after DAG rebuild 'Instability detected' message is printed in case there are issues; added gpu-report-interval parameter to control hashrate summary report frequency based on the number of share submissions; added feature to internal watchdog to display a list of GPUs caused miner restart with GPU is idle error)
⛏ lolMiner v1.19 (added automatic tuning mode for --zombie-tune which on by default and running miner with option --4g-alloc-size set only will run the zombie mode automatic tuning, see details on miner manual; ethash stratum connection will now reconnect after three pool rejected shares in a row that did pass own CPU verify before)
⛏ NBMiner v36.1 (lowered power consumption for 20、30 series Nvidia GPU on octopus; improved hashrate 2% on 16 series Nvidia GPU on octopus; slightly reduced stale ratio on ethash; general improved overall stability, fixed a random crash; added detail datetime & cpu usage in summary log; ethash: if DAG verification failed, display corresponding GPU name in red in summary)
⛏ XMRig v6.7.1 (fixed log initialization; fixed AstroBWT on OpenCL)

So did anyone else's farms go offline today around 7pm Eastern Standard Time?
It was still using power, but HiveOS said that everything was offline. Weird.

I just started using HiveOS and I have a couple rigs of RX Vega 56.  I used PhoenixMiner at first and OC would take effect but not good hashrates.  I switched to TRM and OC wouldn't take effect at all and only about 38Mh/s.  admmemtweaks would work via shell but still only upto 43Mh/s.  Now hoping it was related to the version I downloaded.  Also didn't know RX Vegas didn't need the specific version for them and seems they can use the latest stable image.  So I will have to pull and flash the drive and try again. 

Wonder if other vega owners are facing similar issues or am I the only one.

fans died on two cards, so they were doing 32, the rest 46+.

0.6-191@210109 🐧

What's new?

SYSTEM 🖥
⚙️ Improved nvidia-driver-update tool (added support drivers 460.x series; added support CUDA 11.2; update supported kernels versions)
⚙️ Added CUDA RT v11.2 libraries
⚙️ Update PCI IDs database
⚙️ Improved hive-replace tool (sorting images when calling hive-replace -l: the top three are the latest Stable, Beta and Vega images, then other images by date; added hint to stop X server when running from X server console; added support for extended image descriptions)
⚙️ Improved hpkg command (added 'purge' option to remove all miners)
⚙️ Fixed hang motd in some cases (if wallet config was absent)
⚙️ Small improvement for amd-oc (show timestamp in amd-oc log)
⚙️ Small improvements and fixes for nvidia-oc (now OC delay isn't applied on 1080 when using Pill in special mode with a negative delay; show timestamp in nvidia-oc log; improved error handling)

MINERS👷‍♂️
⛏ GMiner v2.41 (changed default --dag_mode values for polular AMD GPUs: will improve performance compared with previous version when no --dag_mode specified; display valid/stale/invalid shares for Ethash/Etchash and KAWPOW algorithms when solution check on CPU enabled; added display invalid shares on Hive's dashboard: total and per GPUs)
⛏ CryptoDredge v0.26.0 (added KawPow algorithm; added Chukwa-v2 algorithm; fixed MTP issue related to 'invalid device symbol'; added support NVIDIA Ampere RTX 30xx; added --temperature-limit and --temperature-start options; added off flag to Nimiq optimizer --optimizer off; removed some no longer supported algorithms; Notice: this build not supported Ubuntu 16.04 based images and requires latest Nvidia drivers)

New Stable Branch Image release v0.6-190@210108 💿

🔑 Key features:
* Hive Linux client: v0.6-190@210106
* Universal boot mode: BIOS | UEFI
* Linux kernel: v5.4.80-hiveos
* AMD drivers:  OpenCL v20.40
* Nvidia drivers: v455.45.01
* Updated drivers for ethernet and wi-fi adapters

This Image is recommended for Nvidia 10xx/16xx/20xx/30xx and AMD Polaris RX4xx/5xx, Vega and Navi RX 5xxx

💾 Available for download as a torrent or XZ archive in the "GPU" section "Stable Image" over on https://hiveos.farm/install or it can be reflashed via the hive-replace tool or using balenaEtcher or Rufus tools on PC.

It's quite bad results for 7 pcs Vega's in my opinion. aprox. 41MH/s per card right?

0.6-190@210106 🐧

What's new?

SYSTEM 🖥
⚙️ Improved amd-info tool (slightly code refactoring; more GPU info; added unique id, more voltages, states, clocks and so on; added filtering by GPU index and bus id number)
⚙️ Minor fix for nvidia-oc (limit minimum OC delay to 30 sec)
⚙️ Fixed support some 3rd-party hardware (fixed octofan controller fan autodetection)

MINERS👷‍♂️
⚙️ Introduced auto selection CUDA11 builds as latest for some popular Nvidia miners (will work if Nvidia drivers 455+ installed for ethminer, nanominer, t-rex)
⛏ Bminer v16.4.4 (fixed compatibility issues with ETH pools that use the ethproxy protocols e.g. Sparkpool; fixed bugs that lead to high rejection rates on AMD GPUs ETH mining)
⛏ CPUminer-Opt-rplant v5.0.11 (added yespowerTIDE algo for Tidecoin)
⛏ Gminer v2.40 (significant performance improvements for KawPoW, especially on rigs with large number of GPUs; improved compatibility with Ravencoin pools; added option to control DAG buffering mode single or dual via option --dag_mode; added option to support "Zombie Mode" for Ethash, Etchash and KawPoW algortihms via option --dag_limit; improved detection of freezing AMD GPUs; fixed display of core and memory clocks; display GPU PCI index on miner startup; removed unpopular algorithms: sero, vprogpow, progpowz, cuckarood29, cuckarood29v)
⛏ lolMiner v1.18a (improved Zombie mode power draw & speed Polaris GPUs: depending from config uses from 0.5 to 1W less energy and from 0.2 to 0.4 MH/s faster; added --zombie-tune parameter for Polaris GPUs this can increase the performance of zombie mode up to 15%; fixed: segmentation fault when the DNS resolve of a pool fails; fixed: miner does not restart after connection loss; applied potential fix for "address or port already in use" bug)

Hello,

Got a rig of vega 64s (x7), everything was fine (280-290mh) till I upgraded the image from 0.6-16@190218-2 to the newest one 1 month ago. Nightmare started as OC settings were not applied, rig became instable, loosing 2-3 cards, and rebooting all the time.
Tried to go back to precious image, but the link for 4) It seems to be broken:
 
> Loading Hive OS images list
> Select version to install:
0) Exit
1) [email protected] · Latest Stable · K 5.0.21 · N 450.66 · A OpenCL legacy+pal 20.30 · Dual-Boot BIOS+UEFI / Universal image for AMD and Nvidia
2) [email protected] · Latest Beta · K 5.6.0 · N 440.100 · A OpenCL legacy+pal 19.30 · Dual-Boot BIOS+UEFI (Navi Beta3)
3) [email protected] · Latest Vega · K 5.4.0 · N 410.73 · A OpenCL PAL 19.10
4) [email protected] · K 4.15.18 · N 410.73 · A OpenCL 16.60
5) [email protected] · K 5.0.21 · N 440.82 · A OpenCL legacy+pal 19.30 · Dual-Boot BIOS+UEFI (Navi Beta2)
6) [email protected] · K 5.0.21 · N 440.31 · A OpenCL legacy+pal 19.20 · Dual-Boot BIOS+UEFI
7) [email protected] · K 4.13.16 · N 418.56 · A 18.10-572953
[email protected] · K 5.0.21 · N 440.100 · A OpenCL legacy+pal 19.20 · Dual-Boot BIOS+UEFI / no AMD Navi support
9) [email protected] · K 5.0.21 · N 430.64 · A OpenCL legacy+pal 19.20 · Dual-Boot BIOS+UEFI
> 4
***********************************
Hive OS Image Installation
***********************************
http://download.hiveos.farm/history/[email protected]
Total RAM=7918 Mb
Current OS booted from sda and whole drive size is 29568 Mb
Warning: After Hive OS installation, all your data on sda will be lost!
Type "yes" if you want to install Hive OS on sda: yes
> Stopping Hive OS services
> Getting image size
Unable to get image size
> Downloading image file
--2020-12-14 10:08:06-- http://download.hiveos.farm/history/[email protected]

Resolving download.hiveos.farm (download.hiveos.farm)... 104.22.10.47, 104.22.11.47, 172.67.28.84
Connecting to download.hiveos.farm (download.hiveos.farm)|104.22.10.47|:80... connected.
HTTP request sent, awaiting response... 404 Not Found
2020-12-14 10:08:06 ERROR 404: Not Found.
Download Hive OS image failed. Check url
press any key to close session
 
Tried both latest stable and beta image for vega, Tuning and amd-oc setting were not applied, maximum hash is 32/card but rig is unstable like hell. Using phoenixminer, but tried different versions and other miners as a test, all were dropping the cards even worst then phoenix.
https://ibb.co/8bRW2x0

Now my best results is 227mh on 0.6-101@191209, but the watchdog is restarting the miner minimum twice in every hour as 1 gpu drops. If I upgrade to any newer version my calvaria starts again with the instability.

Can someone give me a link to an image, where the OC settings are working properly and the rig is stable please?
Or to vega 0.6-16@190218-2, that was the last one stable.
Or anything how to fix this.



Thank you.

0.6-189@210104 🐧

What's new?

SYSTEM 🖥
🐞 Fixed amd-oc (OC not applied on some configurations; fixed applying Navi SoC Frequency/Voltage when it set as 0 and added checking for proper Navi SoC voltage range; ⚠️  need a reboot after updating: use 'Reboot after complete' option from Update menu to do it!)
🐞 Fixed selfupgrade tool (grub config not updated in some cases)
🐞 Fixed hive-replace (minor fix)
🐞 Fixed Wi-Fi scanning (was broken in v0.6-188 update)

MINERS👷‍♂️
⛏ CPUminer-Opt-rplant v5.0.10 (added phi5 algo for mining Combode coin)
⛏ SRBMiner-Multi v0.6.2 (added algorithm 'circcash' for CPU mining Circcash coin, fee 0.85%; added algorithm 'phi5' for CPU mining Combode coin, fee 0.85%)
⛏ SGminer-fancyIX v0.7.2-1 (added 35% faster optimized bin kernel for 'neoscrypt'; added 'neoscrypt-xaya' algo support; fixed global offset bug for 'yescrypt')
⛏ WildRig v0.28.3 (added 'phi5' algo for mining Combode coin)
🐞 XPMclient: fixed stats

Force reload web page

Improved amd-oc (added SoC clock and voltage adjustment for Vega and Navi; slight code refactoring; OC log can be print with amd-oc log)

How to adjust SoC clock ?
I updated HiveOS to the latest version but i cannot finf SoC clock option in overclock page ?

0.6-188@201230 🐧 🎄🎁🎉

What's new?

SYSTEM 🖥
⚙️ General improvements (optimized boot sequence by using new hivex control; cleanup output to syslog by removing colors codes; added to exclusion list from mining more AMD APU/IGP; other small fixes and improvements)
⚙️ Improved nvidia-oc (optimize Nvidia overclocking with delay: can be useful for Nvidia 30xx series memory OC and pill for Pascal with GDDR5X memory; OC log can be print with nvidia-oc log)
⚙️ Improved amd-oc (added SoC clock and voltage adjustment for Vega and Navi; slight code refactoring; OC log can be print with amd-oc log)
⚙️ Improved hive-replace tool (added option --download=PATH to download and save image to specified location without replacing; added option --repo to use custom repo, see repomirror for details)
⚙️ Improved repomirror package (new option -g or --get for downloading Hive OS images to repo mirror: can be used with --stable or --beta and afterall these images can be used by hive-replace with --repo option)
⚙️ Improved net-test tool (added DoH status reporting)
⚙️ Improved error reporting (added more info on reboot due high LA; added reporting Xid errors for Nvidia GPUs)
⚙️ Updated nvtool to v1.3.5 (minor fixes)
⚙️ Improved wifi (added support for open networks and connection to any network. Note: Be aware that it's unsecure, we don't recommend it!)

MINERS 👷‍♂️
⛏ Gminer v2.39 (added kernel auto-tuning for KAWPOW algorithm, this feature improves hashrate up to 4% on some GPUs; fixed AMD GPUs detection: now miner detects Vega and Navi)



Happy New Year! 🍹💪

Regards,
the whole Hive OS team

0.6-187@201229 🐧

What's new?

MINERS👷‍♂️
⛏ Bminer v16.4.2 (fixed crash ETH mining om Nvidia cards; improved performance on mining ETH on AMD 4G cards)
⛏ NBMiner v36.0 (fixed crashing when mining kawpow algo on some GPUs in versions 35.x; removed algos: eaglesong, trb, hns, sipc, cuckaroo, cuckaroo_swap and reducing binary size)
⛏ SRBMiner-Multi v0.6.1 (added algorithm 'balloon_zentoshi' for Zentoshi coin for CPU mining, fee 0.85%; 'cryptonight_xhv' kernel small changes, pool side hashrate should be a little higher; auto setup for 'cryptonight_xhv' algorithm is a little more agressive now: if you experience some issues, use --gpu-auto-intensity 1 for lighter settings; removed parameter '--gpu-worksize')
⛏ CPUminer-Opt-rplant v5.0.9 (improved auto-stratum connect; added new balloon algo: for Zentoshi coin mining)

No you can still flash the bios back to the original bios it won't be a problem.
There is option in HiveOS under cards where you can put all modded bios for this GPU. Just use the search button to find those modded bios for this model.