eth-utils

ethereum utilities, dev tools, scripts, etc

• gethup.sh: primitive wrapper to geth
• gethcluster.sh: launch local clusters non-interactively (https://github.com/ethereum/go-ethereum/wiki/Setting-up-private-network-or-local-cluster)
• netstatconf.sh: auto-generate the json config of your local cluster for netstat (https://github.com/ethereum/go-ethereum/wiki/Setting-up-monitoring-on-local-cluster)

Usage

Launch an instance

GETH=./geth bash /path/to/eth-utils/gethup.sh <rootdir> <dd> <run> <params>...

This will

• if it does not exist yet, then create an account with password dd [NEVER USE THIS LIVE]
• bring up a node with instance id dd (double digit)
• using rootdir/dd as data directory (where blockchain etc. are stored)
• listening on port 303dd, (like 30300, 30301, ...)
• with the account unlocked
• launching json-rpc server on port 81dd (like 8100, 8101, 8102, ...)
• extra params are passed to geth

$ GETH=./geth bash ~/eth-utils/gethup.sh ~/tmp/eth/ 04 09 --mine console 
Welcome to the FRONTIER
> eth.getBalance(eth.coinbase)
'198400000000001'
>

Launch a cluster

Running a cluster of 8 instances under dir tmp/eth/ isolated on local eth network (id 3301), launch 05. Give external IP and pass extra param --mine.

GETH=./geth bash gethcluster.sh <root> <n> <network_id> <runid> <IP> [[params]...]

This will set up a local cluster of nodes

• <n> is the number of clusters
• <root> is the root directory for the cluster, the nodes are set up with datadir <root>/00, <root>/01, ...
• new accounts are created for each node
• they listening on port 303dd (like 30300, 30301, ...)
• json-rpc server is launched on port 81dd (like 8100, 8101, ...)
• by collecting the nodes' node-urls, they get connected to each other
• if enode has no IP, <IP> is substituted
• if <network_id> is not 0, they will not connect to a default client, resulting in a private isolated network
• the nodes log into <root>/00.<runid>.log, <root>/01.<runid>.log, ...
• <runid> is just an arbitrary tag or index you can use to log multiple subsequent launches of the same cluster, I recommend sequential double digit ids
• the cluster can be killed with killall -QUIT geth (FIXME: should record PIDs)
• the nodes can be restarted from the same state individually using the gethup.sh script
• if you want to interact with the nodes, use a json-rpc client
• you can supply additional params on the command line which will be passed to gethup.sh and eventually to geth for each node, for instance -vmodule=http=6 -mine -minerthreads=8 is a good one.

GETH=./geth bash gethcluster.sh ./leagues/3301/cicada 2 3301 05 77.160.58.3 -mine 
launching node 0/2 ---> tail -f ./leagues/3301/cicada/00.05.log
Welcome to the FRONTIER
launching node 1/2 ---> tail -f ./leagues/3301/cicada/01.05.log
Welcome to the FRONTIER

fill create:

./leagues/3301/cicada/
./leagues/3301/cicada/3301/
./leagues/3301/cicada/3301/00/
./leagues/3301/cicada/3301/00.05.log
./leagues/3301/cicada/3301/00.05.glog
./leagues/3301/cicada/3301/01/
./leagues/3301/cicada/3301/01.05.log
./leagues/3301/cicada/3301/01.05.glog
./leagues/3301/cicada/3301/

You can kill and restart individual nodes or the entire cluster safely, by using different runid you can separate logs for the individual runs in a neat way.

killall -QUIT geth

Using the -QUIT signal is very useful because it dumps the stacktrace into the glog file which you can attach to any bugreport or issue.

Monitor your local cluster:

Installing the eth-netstats monitor

git clone https://github.com/cubedro/eth-netstats
cd eth-netstats
npm install

####Configuring netstat for your cluster

bash /path/to/eth-utils/netstatconf.sh <number_of_clusters> <name_prefix> <ws_server> <ws_secret> 

• will output resulting app.json to stdout
• number_of_clusters is the number of nodes in the cluster.
• name_prefix is a prefix for the node names as will appear in the listing.
• ws_server is the eth-netstats server. Make sure you write the full URL, for example: http://localhost:3000.
• ws_secret is the eth-netstats secret.

For example:

git clone https://github.com/ethersphere/eth-utils
cd eth-utils
bash ./netstatconfig.sh 8 cicada http://localhost:3301 kscc > ~/leagues/3301/cicada.json

####Installing eth-net-intelligence-api

git clone https://github.com/cubedro/eth-net-intelligence-api
cd eth-net-intelligence-api
npm install
sudo npm install -g pm2

Starting the eth-net-intelligence-api

to start the eth-net-intelligence-api client for your cluster

cd eth-net-intelligence-api
pm2 start ~/leagues/3301/cicada.json
[PM2] Process launched
[PM2] Process launched
┌──────────┬────┬──────┬───────┬────────┬─────────┬────────┬─────────────┬──────────┐
│ App name │ id │ mode │ pid   │ status │ restart │ uptime │ memory      │ watching │
├──────────┼────┼──────┼───────┼────────┼─────────┼────────┼─────────────┼──────────┤
│ cicada-0 │ 1  │ fork │ 93855 │ online │ 0       │ 0s     │ 10.289 MB   │ disabled │
│ cicada-1 │ 2  │ fork │ 93858 │ online │ 0       │ 0s     │ 10.563 MB   │ disabled │
└──────────┴────┴──────┴───────┴────────┴─────────┴────────┴─────────────┴──────────┘
 Use `pm2 show <id|name>` to get more details about an app

####Starting the monitor

Use your own eth-netstat server to monitor a league on a port corresponding to a league

cd eth-netstat
PORT=3301 WS_SECRET=kscc npm start &

and enjoy:

open http://localhost:3301