Topic: rig.cards - remote rig monitoring and notifications by email and Telegram - page 2. (Read 12183 times)

This is a great tool! Very useful. I wish I knew about it before. Great work! Congratulations.

As for feature requests, I think it would be nice to have some kind of notification when something goes wrong with the miner. Some message saying the miner was restarted or something like that. Then you can monitor the rig stability, adjusting the GPU overclock, etc.

For example, I was using Gminer and there was one rig that was giving a lot of errors, but I didn't know until I looked inside the log file. So, sometimes I need to manually check the log file to monitor the stability of the rig. I wouldn't need to worry if the tool gave a warning message.

I've found that some people don't know how to get notifications to some Telegram accounts at the same time.

So, you don't need to create multiple accounts on the rig.cards, launch more than one client on the rig and other things.

1) Create group chat in Telegram
2) Add there people who also must get notifications (up to 200,000 people!)
3) Also add to group our bot — @rigcardsbot
4) Now you need send your secret key to the bot to set this group chat for notifications.
Just send message to the chat


Your secret key you can find in Account - Your Profile - Secret key

Thats all!

Now every user in group chat will receive notification.
At any time you can add or remove any user from the chat.

Here is also gif for better understanding:




Finally, guys, if you have any questions or any feature requests - just give me know. You can post here in topic or send email to [email protected]

Many miners use bus number ordering by default.
But if you want full support, what about adding an option in config file for default order / bus number order ?

Sure you can PM me

Thanks for suggestion! I'll check it and if have some questions can I PM u?

but here is not guarantee that any of miners will use same order... For example Claymore even has -altnum option for "alternative GPU indexing". It means that depends miner settings can be different order and it's not possible to check it... What do you think about such issue? Or any miner gives a list ordered by bus id? 
Also for 106-1000 for example I use nvidia-smi instead OHM. So in mixed rigs it will be too hard do determine GPUs...


Anyway, thx again, I must check it 

✅  Update

Client app updated to version 0.7.8.1

Minor fixes

(The update will occur automatically if you have not disabled the auto-update feature)

You just need to recompile openhardwaremonitor lib after adding order by bus number in code. I've done it for my own projects.

I've found that some people trying to set port number of pool instead of miner's API port!    Like 4444 for ethereum pools.

Guys, not pool port is needed! You need to set the port of miner's API! If you don't know what is it just give me know, I'll explain and help, or email to [email protected]

Event more, some miners, like Claymore, already has enabled API by default, so you don't need to do anything, just leave the "use default port" checkbox on the rig settings.

Such feature not present at the moment.

About possibility to add...
The difficulty is:
When client app grabs info of GPU parameters, the list of GPUs has some order.
But when we read data from miner about hashrate of each GPU, that list of GPUs can has a totally different order. So I can't to match these to GPU lists and can't know which hashrate value is for which GPU exactly.
Moreover the order of such lists can depend on miners, on OS, dirvers verion... so here is dont exists the only one right decision.

Will think about it...



So guys, is there any feedbacks about hashrate minitoring feature? Is everything works fine?
Or maybe any feedback about rig.cards as a whole? Any feature requests?

Its possible add the hash rate por each GPU?

⚡⚡⚡  Update


Client app updated to version 0.7.8

Added feature for miner auto detection 

On client start, within a few seconds will be a prompt to try to find working miner on rig for its hashrate monitoring — just press any key.




Of course miner must be launched and API is enabled on default port (for miners that has disabled api by default).


All supported miners at this time are available for auto-detecting:

• Claymore's Ethereum Miner
• PhoenixMiner Ethereum
• SRBMiner Cryptonight
• EWBF Equihash Miner
• Ethminer
• T-Rex
• TeamRedMiner
• NBMiner
• GMiner

Update!

  Added miners:

• GMiner

API is disabled by default, need to active before. Just add --api 80 option to the bat file:


Client app updated to version 0.7.7
(The update will occur automatically if you have not disabled the auto-update feature)


Ask if you want to see miner that don't added yet!

⚡  UPDATE

Client app updated to verson 0.7.6

⚠️  Users who use mining cards like NVIDIA P106-100 и P104-100 make attention!

As you know, to display such cards, you had first to switch them to WDDM mode, as described in the FAQ in the first post of this topic.
However, at some point (some new version of the drivers) this method not working already. Moreover, after switching to WDDM, cards can dissapear in OS and need to reinstall the drivers then.

This issue has been resolved with this update! Now the data for the mining cards is collected by an alternative method using NVIDIA's nvidia-smi utility.
Users using such mining cards, please report is this feature works normally?


How it works:

- if no found any GPU in the rig using the standard method — an attempt will be made to get data via nvsmi
(it's most typical case and everything will happen automatically and no user actions is required)


- if you have a mixed rig, where there are both mining cards and regular ones (say, GTX 1060-6 or RX580), then the alternative mode must be forced in the config.json.
This will affect all NVIDIA cards in the rig (information from the regular 1066, for example, will also be collected by an alternative method).

in config.json you need to add the following line (do not forget to put a comma on the previous one):

"nvsmi": 1

config will be looks like this:




- there may also you have installed NVIDIA drivers to a non-default folder

For such cases in config.json also need to specify the path to the nvidia-smi.exe file.

By default, this file is located here: \Program Files\NVIDIA Corporation\NVSMI\nvidia-smi.exe

You need to add the line:

"nvsmi_path": "\\fullpath\\to\\utility\\nvidia-smi.exe"

Note that the backslash \ must be escaped (in fact, it must be double)

So the result will be like this:




Known issues
When using alternative method there are no possibility to get information about fan rpm. Only % percent value is available.



In case of any questions, issues and other – post to this topic or mail to [email protected]   

Great! If you find any problems, let me know.
Thx for feedback.



So, do you try miner hashrate monitoring?

Just tried on one of my rigs running team red miner. Now I can see hashrate on main rig.cards screen, simple and awesome!

Update!


  Added miners:

• Ethminer
• TeamRedMiner
• T-Rex
• NBMiner

Ask if you want to see miner that don't added yet.


Client app updated to version 0.7.5
(The update will occur automatically if you have not disabled the auto-update feature)



For added miners here is settings to activate API if needed:

• Ethminer

• TeamRedMiner

or if you use non-default port:

• T-Rex

API enabled by default, working on port 4067, no any action required.
Anyway just FYI here is the launch string for miner with enabled api

• NBMiner

API enabled by default, working on port 22333, no any action required.
Anyway just FYI here is the launch string for miner with enabled api

UPDATE

Double check temperatures for notifications 


To avoid false notifications, as well as reducing total number of it (some users receive a hundreds notifications per day!), a double temperature check has been implemented.

This means that now the notification will be sent only after the server receives out of the range temperature value from the client app twice.
At the same time it will allow users not to be overloaded by notifications when the temperature is unstable and permanently changes up and down near the limit.


! This only affects GPU temperature notifications !


Feature is enabled by default for all users (both new and already registered).

However, for those who want to receive notifications after the first check (as before this update) there is a checkbox - you can turn off the double check.

I've created email for user support — [email protected]. There you can write on any questions relating to the work of the service.

Having issues or found a bug? New feature requests? → [email protected]

Of course, all comments and suggestions can be left also here, in this topic, or just write to me to PM. 

As usually I would also to see feedbacks from users of rig.cards

Ahh.. I see. OK

Got it! Added to tasks.


And the answer for vmozara
Yes. Also added to tasklist.


I hope I can continue works on implement new miners on the end of the week.

miner sw = miner software as to avoid confusion with i dunno, octo miner hw?

anyway, i use: ethminer, nbminer and teamredminer

BTW i only looked quickly at your tool, and it looks rather good, i think pretty much only alternative is not developed anymore rigmonitor.app - that was/is really great tool

Yes, but what you mean "miner sw"? What is sw? Sw..itching?  


As I said earlier, I choose some miners to implement just for example. The feature of hashrate monitoring was launched only some days ago.
In my environment, for example, users usually use Claymore or Phoenix for Ethash... So...

Anyway Ethminer added to tasklist!