[quote=“jmnemonicj;5639”][quote=“jmnemonicj;5610”][quote=“hamednik;1198”]I have integrated tdxminer for HiveOS, so far its for the AMD cards only. Its a dirty patch but it gets the job done until Hive dev team add it to their next release!
…
[/quote]
I have mixed rig (1x gtx 1060 6gb, 6x rx580 8gb)
…
[/quote]
Images:
https://imgur.com/dvyqXQT
https://imgur.com/TBuwT0A[/quote]
only amd rx series
[quote=“idrisou;5647”][quote=“jmnemonicj;5639”][quote=“jmnemonicj;5610”][quote=“hamednik;1198”]I have integrated tdxminer for HiveOS, so far its for the AMD cards only. Its a dirty patch but it gets the job done until Hive dev team add it to their next release!
…
[/quote]
I have mixed rig (1x gtx 1060 6gb, 6x rx580 8gb)
…
[/quote]
Images:
https://imgur.com/dvyqXQT
https://imgur.com/TBuwT0A[/quote]
only amd rx series[/quote]
I have 6x AMD RX 580 8gb, what do you mean?
Interesting for a T-Rex miner
Рандомно вылетает по одной карте на каждой ферме. Помогает только ребут. Не подскажите как можно победить? https://forum.hiveos.farm/uploads/editor/az/f9a1dxstbbc0.png
Standalone working well, but Integrated not working at all, saying “no algorithm specified”. How to fill dstm wallet with -a lyra2z?
Tried everything, pls give me example of wallet fill on hiveOS dstm
{
“name”: “GIN”,
“props”: {
“dstm”: {
“pass”: “x”,
“port”: “20012”,
“server”: “stratum+tcp://angrypool.com”,
“template”: “%DWAL%.%WORKER_NAME%”
},
“dwal”: “GZB9aMueSVG9npgy6ZX3bGgkNo3tztFr8r”
}
}
@heartly you’ve missed a step otherwise you shouldnt have this issue. Algorithm selection is hardcoded in my patch for dstm/tdxminer.
@olezko46 i am sorry but i dont speak russian, but from the picture you have attached i can see you may be referring to the GPUs falling off the bus, this is most likely OC issue. Also you’re running your fans on 100%, dont do that.
@hamednik I’ve noticed two situations where reporting from log parsing could be better:
I’ve taken the liberty and somewhat changed line in tdxHelper that is parsing through the log file. If you find it satisfying you can include it in the next version of your very helpful manual:
getGPUstats=grep -i "Stats GPU" $LOGS | tail -n $GPUS | sed 's/.*lyra2z: \(.*\)/\1/' | cut -d "M" -f1 | sed '/0.000/d' | sed -e ':a' -e 'N' -e '$!ba' -e 's/ /},{"hash_ps":/g'
This way I first take only lines that are of interest to me, and tail will take only last n lines of miner stats log. This gives me ordered list all the times (or almost all the times). Additionally, I am removing 0.000 hasrates from reporting. That way when HiveOS receives this info, it displays stats for live GPUs and the difference is reported as missing GPUs which is helpful if you are running more than few rigs.
@givo thanks for your comment, my work was really a quick and dirty job of getting everything to work and i really havent spent my time trying to perfect the patch. I was hoping my adaptation would really motivate hiveos devs to include tdxminer in their next release with their own official patches, however that hasnt happened. I understand there is a release of HiveOSv2 in order which will address custom minder issues, would be interesting what they have inmind in terms of miners with no API and only log parsing options.
In anycase, I will test your code and if all was okay will include it in my files.
HiveOS dev team already works and testing “Custom miner API” and see on it big profit for all
More details later
@HaloGenius yeah i am quiet curious tbh to findout how you’re planning to handle the custom miners and the stats feed back to the hive api agent.
Also on an off topic note, can you please do something about the “quote” feature of this forum, its really annoying that the “quote” option is not clickable. Ive tried with different browsers and as soon as the mouse hovers out of the main comment box, the “quote” option disappears. I find myself trying to reply to specific comments however its not possible and i have to quote their username instead. I think its a bug in the VanillaForums that you use.
Does somebody try rx460/560/550 with Rocm drivers in hiveos with this miner?
@hamednik
I’m fully agree with you about forum core - they are too buggy. Main idea change and convert it to powerful support portal, something like IPB forum core.
Returning to our past conversation, also Dmitro’s dialogue with users on facebook, also requests on forum to add one or another miner to HiveOS we understood what we are faced with many questions like:
Yes, sorry for bad story with tdxminer integration but it’s will good to start for all HiveOS users and developers at near future
It is really nice that the system is free up to 3 rigs, but I really think you do need to make a distinction between your paying and non-paying customers. Perhaps some sort of support level agreement would also be beneficial, but it all depends on Hive’s team business development and strategies. As for knowing which miner is popular, i think a look at every pool’s (at least YIIMP style pools) will give you any current popular miner for specific algorithm, here is a look at https://bsod.pw/site/miners and you’ll see tdxminer is on the second one. Lolminer and that entire coin (minexcoin/mnx) which i believe is some sort of a scam anyways, is a special case. A general rule of thumb would be for example that if a miner only supports “one specific coin” then it probably shouldnt be in your list of priorities.
Finally and as far as a am concerned, i dont think the tdxminer was a bad story, for me it was a learning curve and i got to know your system a little better and it shows that we can make HiveOS stronger with the community’s support and input.
Lyra2z, tdxminer - it can be or will be a short story for some coins or not - this is not the main thing.
And in fact, the issue is not only in business, if you are about this. It’s just a fragment or part of something.
With your opinion why we did not integrate the TDXMiner and the opinions of other users about where this or that miner, we came to the conclusion that in fact we rested on the wall of our development. This is a dead-end road. Sooner or later it would have happened.
We start to make changes on this and I’m hope it’s really help to make system more strong and flexible.
Now we make changes on the fly for upcoming release HiveOS 2.0 to be ready for next steps of changes.
@givo i’ve tested your “sed ‘/0.000/d’” addition and decided not to include it in the tdxHelper, if a GPU has fallen of the bus and is no longer hashing, you may want to know which one it is, if you simply remove the the GPU with zero hashes from the trail you are effectively sending one less GPU to the hive api and hive will always incorrectly show the very last GPU to be offline, instead of showing which gpu has stopped hashing. thank you anyways.
@hamednik Fair enough 
It would be excellent if Hive could adjust its dashboard presentation of GPUs with 0 hashrate so to make a note of such GPUs at the top of the page where there are numbers of active/inactive miners and GPUs. Because, those are failing GPUs and we need some kind of presentation of that info on one screen. With several pages of miners on dashboard, you need to scroll up and down looking for a failing GPU.
In terms of (in)correctly showing failing GPU - please note (as I have mentioned before) your parsing of a log file is not sorting the result, so it sometimes starts with GPU0 (correctly) and sometimes with any other random GPUxx. Here, part of log file from hive agent where you can see that miner stats show 0 hashrate on GPU5 and 10 secs later on GPU4:
[Tue Jul 3 19:18:27 +04 2018] > {“method”:“stats”,“jsonrpc”:“2.0”,“id”:0,“params”:{“rig_id”:“xxx”,“passwd”:“yyy”,“miner”:“dstm”,“miner_stats”:{“hs”:[2.745,2.745,2.745,2.745,2.745,0,2.745,2.745,2.745,2.745,2.745,2.745,2.745],…
[Tue Jul 3 19:18:38 +04 2018] > {“method”:“stats”,“jsonrpc”:“2.0”,“id”:0,“params”:{“rig_id”:“xxx”,“passwd”:“yyy”,“miner”:“dstm”,“miner_stats”:{“hs”:[2.745,2.745,2.745,2.745,0,2.745,2.745,2.745,2.745,2.745,2.745,2.745,2.745],…
It is not such a big deal, since one can identify failing GPU by comparing temps and power usage among GPUs, but this is one small potential improvement that can be addressed…
@givo I actually looked for this scenario to see if tdxminer is outputting GPU stats in an unsorted list, but what i saw in every instance was the same format which happens to be sorted and shows the correct reflection of what GPU is producing what hash. Here is a copy of tdxminer which i based my parsing on. Check your miner log and you should see something similar.
[2018-07-03 19:28:52] Stats GPU 0 - lyra2z: 3.365Mh/s (3.364Mh/s)
[2018-07-03 19:28:52] Stats GPU 1 - lyra2z: 3.365Mh/s (3.364Mh/s)
[2018-07-03 19:28:52] Stats GPU 2 - lyra2z: 3.365Mh/s (3.364Mh/s)
[2018-07-03 19:28:52] Stats GPU 3 - lyra2z: 3.365Mh/s (3.364Mh/s)
[2018-07-03 19:28:52] Stats GPU 4 - lyra2z: 3.365Mh/s (3.364Mh/s)
[2018-07-03 19:28:52] Stats GPU 5 - lyra2z: 3.365Mh/s (3.364Mh/s)
[2018-07-03 19:28:52] Stats Total - lyra2z: 20.190Mh/s (20.185Mh/s)
In the case of t-rex miner however i noticed that its exactly as you said and GPU listing is all over the place:
[2018-07-03 19:30:35] GPU#2: MSI GTX 1070 Ti, 2796.99 kH/s
[2018-07-03 19:30:35] GPU#2: MSI GTX 1070 Ti, 0.026MH/W, 0.0015MH/Mhz
[2018-07-03 19:30:35] GPU#2: MSI GTX 1070 Ti, 64C(F:48%) 1878/4233MHz(109W)
[2018-07-03 19:30:58] GPU#3: MSI GTX 1070 Ti, 2767.19 kH/s
[2018-07-03 19:30:58] GPU#3: MSI GTX 1070 Ti, 0.025MH/W, 0.0015MH/Mhz
[2018-07-03 19:30:58] GPU#3: MSI GTX 1070 Ti, 63C(F:49%) 1873/4233MHz(109W)
[2018-07-03 19:32:09] GPU#1: MSI GTX 1070 Ti, 2787.37 kH/s
[2018-07-03 19:32:09] GPU#1: MSI GTX 1070 Ti, 0.026MH/W, 0.0015MH/Mhz
[2018-07-03 19:32:09] GPU#1: MSI GTX 1070 Ti, 64C(F:48%) 1886/4233MHz(109W)
For t-rex miner parser (which i havent realsed) i just pipe the log to "sort -V -k 3 " and that deals with the problem. If you think for whatever reason you’re not getting sorted results on tdxminer, play around with the sort command and you’ll get a sorted list.
Noob here. I have a few rigs with mostly Sapphire RX 580s. I am using tdxminer, integrated with HiveOS, to mine two different coins on two separate pools. I am not earning anything close to what the calculators estimated. Any ideas on what is causing this?