Node FAQ
Download and install, then get started.
Make sure you have latest Microsoft Visual C++ Redist installed:
I’m getting high CPU usage, even when my node is synced to latest blockchain height.
Make sure you are performing VDF verification with GPU/iGPU (more info in VDF FAQ).
On Linux make sure user has access to GPU/iGPU through the video
and or render
groups:
sudo adduser $USER videosudo adduser $USER render
Check if timelord has been enabled by accident.
Timelord should be disabled by default. You can check if it is enabled by looking under SETTINGS in WebGUI. Or check the config/local/timelord
file.
I see two different heights for blocks, both in logs and block explorer. Which is right?
Both. Each block in MMX has two heights:
- Block
Height
(always the lower value, Network Height) VDF Height
(always the higher value)
In nearly all cases, block height is what is relevant for normal usage.
A block is prepared by timelord every 10 sec, and given an VDF height.
About 2% of the time a non-TX block is the result, because of no winning proof. In that case VDF height still increases, but block height stands still. Chain will still adjust itself to 8640 blocks each day, with ~10 sec block interval.
Can look at block height as the ‘public facing’ height, while VDF height is an ‘internal clock’ height.
One example of VDF height usage is FARMER / PROOF tab. You submit proofs that passed plot filter to an VDF height. Block height is unknown at this point. It will depend on potential non-TX blocks before it becomes a block in chain. If your proof was best, it will be listed with block height under FARMER / BLOCKS tab. Can be tricky to correlate at times.
Time Calculator can help show the relation, and lookup or estimate time and date for heights.
I’m getting a warning when trying one of the mmx ...
commands:
[Proxy] WARN: connect() failed with: xxx (localhost:1133x)
Make sure node is running. Except mmx wallet create
, all other commands require node running.
In Windows, start the MMX CMD shortcut from Start Menu.
If Linux binary package, enough to open a terminal.
If compiled from source on Linux (before using CLI commands):
cd mmx-nodesource ./activate.sh
Overview of CLI commands.
Look Blockhain and Challenge Generation sections of whitepaper.
I’ve just created some plots. How do I start farming them?
Through SETTINGS in WebGUI, you have a Plot Directory input field below the Harvester section. Input the directory with the plots you have created, press ADD DIRECTORY button.
Alternatively you can edit the config/local/Harvester.json
file:
{ "num_threads": 32, "recursive_search": true, "reload_interval": 3600, "plot_dirs": [ "E:/MMXplots/", "F:/MMXplots/", "/mnt/Harddisk123/MMXplots/", "/mnt/Harddisk456/MMXplots/" ]}
Only use forward slash /
, even if Windows. Last entry is not followed by a comma ,
.
Score indicates how close your plot has proofs for the challenge. The lower the score, the better the proof.
Not really. But you will help the network by allowing incoming connection. If possible, enable port forwarding on 11337
(TCP).
Per-default, automatic UPnP port forwarding is enabled. You can disable it through SETTINGS in WebGUI. Or set "open_port": false,
in config/local/Router.json
file.
I’ve got several computers. Full node on one, and a few remote harvesters. How do I set them up?
Look Remote Services guide.
Should rarely happen with mainnet. If forked, and not automatically getting back to synced. You can try to revert DB to a height about 1000 blocks before node got stuck.
Through SETTINGS in WebGUI, try the Revert DB to height
option.
If using CLI, try the mmx node revert <height>
command.
To force a 100% sync from scratch. Stop the node, delete following directory:
- Windows:
C:\Users\<user>\.mmx\mainnet\db\
- Linux:
~/.mmx/mainnet/db/
- Linux:
~/mmx-node/mainnet/db/
Start the node again. A clean sync will start from height 0.