Visit lnplay.guide to learn more about this project.
WARNING: This software is new and should be used for testing and evaluation only!
This repo allows you to deploy LNPlay quickly in a modern docker engine using docker swarm mode. LNPlay is a docker container application that deploys a bitcoin core daemon and one or more core lightning nodes to a docker engine (local or remote). Clams is deployed as the web-frontend for interacting with each of the CLN nodes over the websocket interface. Connection information can be embedded in QR codes for quick client onboarding. Each node is funded with 1.0000000 (regtest) bitcoin, so they can go around and open channels, learn about channel liquidity, test BOLT12 offers, etc.
Want to try this software but don't have the skill to host it yourself? Consider renting your own Private Lightning Network at LNPlay.live. For more information about this project, visit lnplay.guide.
LNPlay runs in regtest, signet, or mainnet modes. When running in regtest, various channel setups can be created. All lightning nodes are able to communicate with each other using a docker overlay network.
To get started, clone this repo and its submodules:
git clone --recurse-submodules https://github.com/farscapian/lnplay
As mentioned, you can run LNPlay in any docker engine, either local or remote. Don't have docker engine installed in one of those locations? You can run the ./install.sh file to get it installed. After running it, you may need to restart your computer or log out and back in to refresh your group membership.
Environment files (contained in ./environments/) are where you specify the parameters of your deployment. Anything you specify in your env file overrides anything in ./defaults.env. Here's an example env file called domain.tld that will deploy 10 CLN nodes to a dockerd running on domain.tld in signet with TLS enabled. Check out our AWS hosting guide for more details on hosting your own LNPlay instance on AWS.
DOCKER_HOST=ssh://ubuntu@domain.tld
BACKEND_FQDN=lnplay.domain.tld
FRONTEND_FQDN=remote.domain.tld
ENABLE_TLS=true
BTC_CHAIN=signet
CLN_COUNT=10
First, update active_env.txt to set the active environment file, then run the following scripts:
Brings LNPlay up according to your active environment definition.
Brings your LNPlay down in a non-destructive way.
Deletes docker volumes related to your active env so you can reset your environment. This is very useful for development. Note mainnet is NEVER deleted! But if you absolutely must, you can run docker volume rm VOLUME_NAME.
This is just a non-destructive down.sh, then up.sh. Just saves a step. Like down.sh, you can pass the --purge option to invoke purge.sh.
This script allows you to perform load testing against a remote LNPlay deployment.
Allows you to interact with the current bitcoind instance.
Allows you to interact with the CLN instances. Just add the --id=12 to access a specific core lightning node. For example: ./lightning-cli.sh --id=12 getinfo
The default environment deploys everything in regtest mode to your local docker daemon. By default there are 5 CLN nodes all backed by a single bitcoind node having a block time of 5 seconds (override with REGTEST_BLOCK_TIME) (see EBT below). Each CLN node is connected to docker overlay network so they're gossiping on the same P2P network. Each CLN node is funded with 100,000,000 sats, aka 1 BTC.
Effective Block Time: In the case of LNPlay, the more CLN nodes you deploy, the longer the Effective Block Time (EBT) -- or the time which each respective CLN node "notices" a change in the block height. Each CLN node pollsbitcoindto check the blockheight, and becauseLNPLaydeploys a lot of CLN nodes, there is a need to "spread out" the polling activity evenly among the nodes so as to not overburdenbitcoind. The Effective Block Time is UX time -- what the user experiences for the block time. It can be found in the CLN yaml output asCLN_BITCOIND_POLL_SETTING. By default, LNPlay targets a EBT of 5 seconds for a 200 CLN node count environment.
If you want to run signet, set BTC_CHAIN=signet in your env file. The scripts will stop if the bitcoind signet wallet is inadequately funded. If the balance is insufficient, an on-chain address will be shown so you can send signet coins to it. It is recommended to have a bitcoin Core/QT client on your dev machine with a signet wallet with spendable funds to aid with testing/evaluation.
It is not recommend to run mainnet at this time due to how new this software is. But it runs similarly to signet.
The following table shows the most common configuration settings.
| Environment Variable | default value | Description |
|---|---|---|
BTC_CHAIN |
regtest |
Set the chain you want to deploy: regtest, signet, mainnet. |
CLN_COUNT |
5 |
The total number of CLN nodes to deploy. Max is MAX_SUPPORTED_NODES=200 |
BACKEND_FQDN |
lnplay |
Default backend lnplay FQDN (e.g., lnplay.domain.tld |
FRONTEND_FQDN |
remote |
Default frontend remote FQDN (e.g., remote.domain.tld) |
DEV_PLUGIN_PATH |
null |
Override the local (i.e., 127.0.0.0) CLN plugin path which gets mounted into the CLN containers. |
ENABLE_TOR |
false |
Deploy a TOR proxy for each CLN node so you can create lightning channels with onion-only endpoints. |
ENABLE_TLS |
false |
If true, letsencrypt certificates will be generated. This requires DNS and firewall settings to be properly configured. |
REGTEST_BLOCK_TIME |
5 |
Adjust the blocktime (in seconds) used in regtest environments. |
CHANNEL_SETUP |
none,prism |
By default, no channels are created. If prism, a layout useful for developing prisms will be established. |
CREATE_PRISM |
false |
By default, a prism will not be created. |
ENABLE_BITCOIND_DEBUGGING_OUTPUT |
false |
If true, bitcoind will emit debugging information. |
CLN_P2P_PORT_OVERRIDE |
null |
If specified, this port will be used in the --announce-addr= on your mainnet or signet node 0. |
NAMES_FILE_PATH |
./names/names.txt | Provide a custom list of aliases for the CLN nodes. Should be a fully qualified path. |
COLORS_FILE_PATH |
./names/colors.txt | Provide a custom list of node colors. |
LNPLAY_SERVER_PATH |
$(pwd)/lnplay/stacks |
Specify where deployment articfacts are stored. |
DIRECT_LINK_FRONTEND_URL_OVERRIDE_FQDN |
null |
If specified, overrides the https://${BACKEND_FQDN} to specified value: e,g., 'app.clams.tech' |
ENABLE_CLAMS_V2_CONNECTION_STRINGS |
true |
If true, will emit Clams v2 Connection String format with "LARP mode". |
DO_NOT_DEPLOY |
false |
Set to true to use as a safeguard against inadvertant state changes to an environment. |
CONNECT_NODES |
true |
By default, all regtest nodes are connected to each other to bootstrap the p2p network. |
DEPLOY_CLAMS_REMOTE |
true |
By default, Clams gets deployed to port 80/443. |
There are other options in there that might be worth overriding, but the above list should cover most use cases.
The prism channel setup is useful for testing BOLT12 Prisms. In set to prism Alice (node0) opens a channel to Bob (node1), then Bob opens multiple channels with every subsequent node after Bob. Then on Bob, a BOLT12 Prism is created. When Alice pays Bob's BOLT12 Prism Offer, Bob splits the regtest coins payment to the remaining n nodes.
When you bring your services up, the ./show_cln_uris.sh script will emit connection information, but also saves direct links to ./output/cln_connection_info_${BACKEND_FQDN}.csv. This file can be used as input for 1) the load testing submodule.
These QR codes can be generated by adding the --qrcode option to ./show_cln_uris.sh. It is recommended to print out cards before hand, then use a label maker to print out the individual QR codes with sticker backing. Maintain the order of the connection information, if possible.
When deploying your application to a local docker engine, the CLN plugin path will get mounted into each CLN instance (container). If you want to make updates to the ``, for example, make the change, then run reload_dev_plugins.sh which iterates over each CLN node and instructs it reload the bolt12-prism.py the newly updated plugin.
Global Network Graph Visualizers:
- This repo can be used to view a global channel graph.