Worker (Ursula) Configuration Guide¶
NuCypher staking operations are divided into two roles “Staker” and “Worker” - This Guide is for Workers.
Worker - (aka “Ursula”) Active network participant who carries out re-encryption work orders.
The Worker is the bonded delegate of a Staker and an active network node. Workers must remain online to provide uninterrupted re-encryption services on-demand. Each staking account or Staker is bonded to exactly one Worker. The worker’s ethereum account must remain unlocked to send automated work confirmation transactions and have enough ether to pay for transaction gas; however, it is not necessary (and potentially risky) to hold NU tokens on a worker’s account for any reason.
In order to be a successful Ursula operator, you will need a machine (physical or virtual) which is kept online. As a frame of reference, maintaining an Ursula is similar to the deployment and maintenance of a high-availability web service, with the addition of Ethereum accounts management. The worker must be tolerant of internet connectivity problems, and power outages via a redundant power supply. However, short temporary service disruptions such as upgrades are understandable.
Aside from the base requirements for installation of the
Dedicated physical/virtual machine
Physical or SSH access
2GB RAM (minimum)
20GB HDD free storage - backups are required since data loss results in a malfunctioning worker
Publicly available IP address - static where possible, NAT management where applicable
TCP Port 9151 opened for network communication - firewall rules where applicable
Access to a fully synced Ethereum web3 provider e.g. local node, Infura, Alchemy etc. (see Using an Ethereum Node) is required to read and write to smart contracts
Workers can be run on cloud infrastructure – for example, Digital Ocean 4GB Basic Droplet satisfies the memory and processing power requirements listed above.
If also running a local Ethereum node on the same machine, additional requirements are needed.
Configure and Run a Worker¶
This guide assumes that you already have
nucypher installed, have initiated a stake, and bonded a worker.
Ensure that a Stake is available (see Staker Configuration Guide)
Run an ethereum node on the Worker’s machine eg. geth, parity, etc. (see Running an Ethereum node for Ursula)
nucypheron Worker node (see Installation Guide)
Create and fund worker’s ethereum address (see Fund Worker Account with ETH)
Bond the Worker to a Staker (see Bonding a Worker)
Configure and run a Worker node (see Configure and Run Ursula)
Ensure TCP port 9151 is externally accessible
Keep Worker node online!
1. Run an Ethereum node for Worker¶
Worker (Ursula) transactions can be broadcasted using either a local or remote ethereum node. See Using an Ethereum Node for more information.
2. Fund Worker Account with ETH¶
Ensure that the worker’s ethereum account has sufficient ETH to pay the gas for regular transactions, or it may forgo subsidies (inflationary rewards).
Reducing the gas costs burdened upon stakers/workers is an active and high-priority area of network development.
For testnet, the worker account can be funded with Rinkeby testnet ETH via https://faucet.rinkeby.io/.
3. Ensure Worker account is bonded to Staker¶
Ensure that the worker’s ethereum account is bonded to the Staker. See Bonding a Worker.
4. Run Worker¶
Run Ursula via CLI (Interactive)¶
(nucypher)$ nucypher ursula init --provider <YOUR PROVIDER URI> --network <NETWORK_NAME>
<YOUR PROVIDER URI> with a valid node web3 node provider string, for example:
ipc:///home/<username>/.ethereum/geth.ipc- IPC Socket-based JSON-RPC server
https://<host>- HTTP(S)-based JSON-RPC server
wss://<host>:8080- Websocket(Secure)-based JSON-RPC server
<NETWORK_NAME> is the name of the NuCypher network domain where the node will run.
If you are using NuCypher’s testnet, this name is
Create a password when prompted
Enter a password to encrypt your keyring: <YOUR PASSWORD HERE>
Run the Ursula!
(nucypher)$ nucypher ursula run --interactive
Verify Ursula Blockchain Connection (Interactive)
This will drop your terminal session into the “Ursula Interactive Console” indicated by the
Verify that the node setup was successful by running the
Ursula >>> status
To view a list of known Ursulas, execute the
Ursula >>> known_nodes
You can also view your node’s network status webpage by navigating your web browser to
Ensure that this URL can be accessed publicly: it means that your node can be seen by other NuCypher nodes.
Since Ursulas self-sign TLS certificates, you may receive a warning from your web browser.
To stop your node from the interactive console and return to the terminal session:
Ursula >>> stop
Run Ursula with Docker¶
Assuming geth is running locally, configure and run an Ursula using port and volume bindings:
export NUCYPHER_KEYRING_PASSWORD=<YOUR KEYRING_PASSWORD> export NUCYPHER_WORKER_ETH_PASSWORD=<YOUR WORKER ETH ACCOUNT PASSWORD> # Interactive Ursula-Worker Initialization docker run -it -v ~/.local/share/nucypher:/root/.local/share/nucypher -v ~/.ethereum/:/root/.ethereum -p 9151:9151 -e NUCYPHER_KEYRING_PASSWORD nucypher/nucypher:latest nucypher ursula init --provider file:///root/.ethereum/geth.ipc --network <NETWORK_NAME> # Daemonized Ursula docker run -d -v ~/.local/share/nucypher:/root/.local/share/nucypher -v ~/.ethereum/:/root/.ethereum -p 9151:9151 -e NUCYPHER_KEYRING_PASSWORD -e NUCYPHER_WORKER_ETH_PASSWORD nucypher/nucypher:latest nucypher ursula run
5. Monitor Worker¶
A reliable way to check the status of a worker node is to view the logs. As a shortcut, nucypher’s
logs can be viewed from the command line using
tail: tail -f $(nucypher –logging-path)/nucypher.log
Once Ursula is running, you can view its public status page at
It should eventually be listed on the Status Monitor Page (this can take a few minutes).
Ursula can optionally provide a Prometheus metrics endpoint to be used for as a data source for real-time monitoring. The Prometheus client library is not installed by default and must be explicitly installed:
(nucypher)$ pip install nucypher[ursula]
The metrics endpoint is disabled by default but can be enabled by providing the following
parameters to the
nucypher ursula run command:
--prometheus- a boolean flag to enable the prometheus endpoint
--metrics-port <PORT>- the HTTP port to run the prometheus endpoint on
The corresponding endpoint,
http://<node_ip>:<METRICS PORT>/metrics, can be used as a Prometheus data source for
monitoring including the creation of alert criteria.
By default metrics will be collected every 90 seconds but this can be modified using the
Collection of metrics will increase the number of RPC requests made to your web3 endpoint; increasing the frequency
of metrics collection will further increase this number.
During the Technical Contributor Phase of our testnet, P2P Validator contributed a self-hosted node monitoring suite that uses a Grafana dashboard to visualize and monitor the metrics produced by the prometheus endpoint.
Both the Ursula Status Page and Prometheus Endpoint are areas of active development.