Thanks.

Those are 2 different models: the HS100 supports remote power on/off, scheduling, etc... but not power consumption monitoring. For that you need the HS110, which are a bit more expensive.


I decided to run in on a separate service for several reasons:
- I need to be able to keep monitoring the rest of the rigs during crashes and upgrades
- the raspberry will monitor rigs for "long" server freezes and reboot them automatically (this is in my TODO list)
- stability. those little things are very resilient. I have been running one for 3 years in a row without any issue

Or maybe all those reasons are bullshit and it's sys admin brain messing with me again

UPDATE:
Added support for Ethermine pool monitoring. Check version 1.1 https://github.com/rodneymo/rig-monitor
Current version supports currentStats and payouts APIs.
Upgrade instructions are here: http://randomcryptostuff.blogspot.nl/2017/09/update-ethermine-pool-monitoring-added.html

If you'd like other pools to be supported then please create issue

Great work!  I've been playing around with trying to do more of the data/graphing locally.  This really helps me get started.
 Here are some thoughts.

You might want to update your first post when you update this.  I was running off that for the latest instructions and they weren't up to date.  It's a convention thing to help "updates" not get lost in longer threads.

There are some issues with your walk through and user permissions.  Also I can see that you've edited out the create sql user in the create_db.sql, but your walk through does not indicate that you need to create the user.  It also does't setup permissions quite right for your script (grafana only needs SELECT but your script needs a lot more than that).  Not hard to fix, but just something to point out.  

I'm a bit confused by all the local storage of data in CSVs.  Aren't you using the database as well?

The rest of the bugs and issues I'll open on github, unless you'd prefer I put them here.

I use smartthings for HA and electricity monitoring.  I'll probably write something that either works with your script or just writes to the DB directly.  I believe I can get that data from their API.  ST is pretty popular here in the US so others might like to use that as well.

Ok. Here's the proper reply:


Thanks for the tip. I will do so this wkd


Fixed in dev branch


I could insert them directly, but keeping the data in CSV allows me to recover it in case I have issues with DBs or upgrades. I think your point is valid but it's more a matter of convenience. Not using CSVs would make the scripts simpler. What's your advice?


Cool. If you do please push it to the repo

You need to make sure you're following his instructions from his latest post.  It's the latest one on his blog.  The instructions have changed slightly for importing csv files.

I got stuck at "Create DB and rig-monitor metadata"
the comand "mysql -u root -p<mysql root password> < ./rig-monitor/sql/create_db.sql" error

it said: "bash: syntax error near unexpected token `<'"

Hi there, just a quick question, I planned to build a mining farm of around 150-200 gpus (total of around 20 rigs). In this case, do I need to link several Raspberry Pi together as a cluster computer? And if yes then how many RPis do I need, say if I want to keep database of around 2-3 days..?

Additionally, does it matter if some rigs are running Windows, some Linux, and some EthOS?

Thanks!
Ejzok

Sorry for the late reply

I'll take a look in a couple of weeks.

Sure.

I'll probably work on this first as I am facing similar issues.

Will keep you posted.
Thanks for the feedback