2940783a9c
## Proposed Changes Adds some improvements I found when playing around with local testnets in #3335: - When trying to kill processes, do not exit on a failure. (If a node fails to start due to a bug, the PID associated with it no longer exists. When trying to tear down the testnets, an error will be raised when it tries that PID and then will not try any PIDs following it. This change means it will continue and tear down the rest of the network. - When starting the testnet, set `ulimit` to a high number. This allows the VCs to import 1000s of validators without running into limitations. |
||
---|---|---|
.. | ||
beacon_node.sh | ||
bootnode.sh | ||
clean.sh | ||
ganache_test_node.sh | ||
kill_processes.sh | ||
print_logs.sh | ||
README.md | ||
reset_genesis_time.sh | ||
setup.sh | ||
start_local_testnet.sh | ||
stop_local_testnet.sh | ||
validator_client.sh | ||
vars.env |
Simple Local Testnet
These scripts allow for running a small local testnet with multiple beacon nodes and validator clients. This setup can be useful for testing and development.
Requirements
The scripts require lcli
and lighthouse
to be installed on PATH
. From the
root of this repository, run:
make
make install-lcli
Starting the testnet
Modify vars.env
as desired.
Start a local eth1 ganache server plus boot node along with BN_COUNT
number of beacon nodes and VC_COUNT
validator clients.
The start_local_testnet.sh
script takes three options -v VC_COUNT
, -d DEBUG_LEVEL
and -h
for help.
The options may be in any order or absent in which case they take the default value specified.
- VC_COUNT: the number of validator clients to create, default:
BN_COUNT
- DEBUG_LEVEL: one of { error, warn, info, debug, trace }, default:
info
./start_local_testnet.sh
Stopping the testnet
This is not necessary before start_local_testnet.sh
as it invokes stop_local_testnet.sh
automatically.
./stop_local_testnet.sh
Manual creation of local testnet
These scripts are used by ./start_local_testnet.sh and may be used to manually
Start a local eth1 ganache server
./ganache_test_node.sh
Assuming you are happy with the configuration in vars.env
, deploy the deposit contract, make deposits,
create the testnet directory, genesis state and validator keys with:
./setup.sh
Generate bootnode enr and start a discv5 bootnode so that multiple beacon nodes can find each other
./bootnode.sh
Start a beacon node:
./beacon_node.sh <DATADIR> <NETWORK-PORT> <HTTP-PORT> <OPTIONAL-DEBUG-LEVEL>
e.g.
./beacon_node.sh $HOME/.lighthouse/local-testnet/node_1 9000 8000
In a new terminal, start the validator client which will attach to the first beacon node:
./validator_client.sh <DATADIR> <BEACON-NODE-HTTP> <OPTIONAL-DEBUG-LEVEL>
e.g. to attach to the above created beacon node
./validator_client.sh $HOME/.lighthouse/local-testnet/node_1 http://localhost:8000
You can create additional beacon node and validator client instances with appropriate parameters.
Additional Info
Adjusting number and distribution of validators
The VALIDATOR_COUNT
parameter is used to specify the number of insecure validator keystores to generate and make deposits for.
The BN_COUNT
parameter is used to adjust the division of these generated keys among separate validator client instances.
For e.g. for VALIDATOR_COUNT=80
and BN_COUNT=4
, the validator keys are distributed over 4 datadirs with 20 keystores per datadir. The datadirs are located in $DATADIR/node_{i}
which can be passed to separate validator client
instances using the --datadir
parameter.
Starting fresh
Delete the current testnet and all related files using. Generally not necessary as start_local_test.sh
does this each time it starts.
./clean.sh
Updating the genesis time of the beacon state
If it's been a while since you ran ./setup
then the genesis time of the
genesis state will be far in the future, causing lots of skip slots.
Update the genesis time to now using:
./reset_genesis_time.sh
Note: you probably want to just rerun
./start_local_testnet.sh
to start over but this is another option.