v1.1.1

This is v1.1.1 of Hyperdrive. This fixes some issues reported by the community, and adds a few important features related to upgrading.

This is a high priority update for anyone that switched from Holesky to Mainnet on the same machine; in general, all users should plan to update when they are able to do so.

Known Issues

• Lodestar v1.22.0 does not currently support x64 systems without the avx instruction set. This will be fixed in a later client release.

• Reth still has a configuration issue that prevents it from working with the StakeWise v3 Operator service. If you're a StakeWise operator, be advised that you will have to use a different client for now.

  • We believe we have a fix in place for this and will release it in a future version.

---

Hyperdrive Core

Clients

• Updated Besu to v24.10.0.

• Updated Nethermind to v1.29.1.

• Updated Reth to v1.1.0.

• Updated Nimbus to v24.10.0.

• Updated Prysm to v5.1.2.

• Updated Teku to v24.10.3.

• Updated Prometheus to v2.55.0.

Service

• Installing Hyperdrive updates via apt update now prompts you to automatically restart the service so the daemons will be up to date.

• Automatic TXs will now ignore the gas threshold setting if a manual max fee is specified, as the threshold setting is only useful for dynamic network-based fees.

• Automatic TXs will now lower the max priority fee down to the max fee if the priority fee is greater than the max fee.

Config

• Switching networks in the service config network dropdown now properly loads the new network settings before starting the containers again.

  • Thanks to @jshufro for reporting this issue!

• The default max fee has been reduced to 20, and the default max priority fee has been reduced to 0.1. Note this only affects new users that are configuring Hyperdrive for the first time.

• The config validation routine will now work properly if you don't have any of the default MEV-Boost relays enabled but do have your own custom relays.

• Changing the contents of the MEV-Boost custom relay option will now properly flag the mev_boost container as needing to be restarted.

CLI

• Updating manually via hyperdrive service install will now automatically restart the service unless the --no-restart flag is provided.

• New global flags:

  • --itsf can be used to ignore TX simulation failues and sign / submit TXs anyway, even if they will revert.
  • --fgl can force a specific gas limit on all transactions created by whatever command is being run.
  • Do NOT use these unless you know what you're doing and have a good reason.

• The max fee selection prompt now properly shows the low and high transaction cost range, instead of just showing the high cost twice.

• Running Constellation or StakeWise commands without the respective module enabled will now print a proper warning message instead of a connection error.

• Manual TXs will now lower the max priority fee down to the max fee if the priority fee is greater than the max fee, and prompt the user with a notice before confirming the fee choices.

---

Modules

Constellation

• Fixed a bug that broke Constellation's automatic signed-exit upload task if an exit message was manually uploaded first via cs minipool upload-signed-exit.

• Fixed a bug that caused Rocket Pool contract queries to hang on single-CPU systems.

• New flags on cs minipool create:

  • --sbc skips checking if the node balance is sufficient to cover the temporary staking lockup.
  • --slc skips checking if Constellation has enough ETH and RPL in its vaults to fund a new minipool.
  • Only use these if you intend to sign (but not submit) a TX for manual bundling, or submit at a later date when the on-chain conditions become correct. Otherwise the transaction may revert when submitted and cost you gas fees!

• Removed delegate info from cs minipool status since it is not relevant for Constellation node operators.

v1.1.0

This is the much-anticipated v1.1.0 of Hyperdrive. This adds support for the Constellation module on mainnet, many updates to support for nodeset.io requests, updates the Ethereum clients, and more!

⚠️ NOTE: As usual, after upgrading, it's important to run hyperdrive service config and/or hyperdrive service start before running any other commands so the latest daemon services are pulled first.

Known Issues

• Lodestar v1.22.0 does not currently support x64 systems without the avx instruction set. This will be fixed in a later client release.

• Reth still has a configuration issue that prevents it from working with the StakeWise v3 Operator service. If you're a StakeWise operator, be advised that you will have to use a different client for now.

  • We believe we have a fix in place for this and will release it in a future version.

---

Hyperdrive Core

Clients

• Updated Besu to v24.9.1.

• Updated Geth to v1.14.11.

• Updated Nethermind to v1.29.0.

• Updated Reth to v1.0.7.

• Updated Lighthouse to v5.3.0.

• Updated Lodestar to v1.22.0.

• Updated Nimbus to v24.9.0.

• Updated Prysm to v5.1.0.

• Updated Teku to v24.10.1.

• Updated Prometheus to v2.54.1.

• Updated Node Exporter to v1.8.2.

• Updated MEV-Boost to v1.8.1.

  • The release notes are for v1.8, but the Docker image being used is for v1.8.1.

Service

• The daemon has been updated to use v2 of the nodeset.io API. It now manages a connection to the server that can be used by multiple modules, rather than the modules managing their own independent connections.

• The configuration system has been overhauled. The options in the Network dropdown are now loaded dynamically from disk rather than being built directly into Hyperdrive. As such, advanced users can now create their own configurations and use Hyperdrive on custom Ethereum chains for things such as development and testing.

• The Eden MEV-Boost relay has been removed from the list of standard relay options because it has been merged with Titan.

• Signed exits are now encrypted locally before sending them to nodeset.io, in addition to regular TLS encryption.

  • Thanks to @poupas for suggesting this enhancement.

• The Hyperdrive daemon and module daemons now require JWT authorization for API access. The Hyperdrive CLI takes care of this automatically, but you can no longer run custom HTTP requests against the daemons without providing a JWT.

  • Thanks to @poupas for reporting this!

Config

• When running the service config TUI, if you have externally managed clients and have MEV-Boost enabled in Hyperdrive, you will now see a modal before the review screen warning you to ensure you have MEV-Boost turned on in your external Beacon Node configuration. If you don't have it enabled in your Beacon Node configuration, your Hyperdrive validators may not be able to propose blocks.

• New parameter: Hyperdrive and TX Fees > Client Timeout

  • This lets you control the timeout for HTTP requests to the Execution Client, Beacon Node, and nodeset.io.

CLI

• New command: hyperdrive nodeset can be used to manage your node's status with the NodeSet service.

  • hyperdrive stakewise nodeset registration-status has been moved to hyperdrive nodeset registration-status.
  • hyperdrive stakewise nodeset register-node has been moved to hyperdrive nodeset register-node.

• New command: hyperdrive constellation (hyperdrive cs) can be used to manage the new Constellation module.

  • See the section below for more information on this module.

• The service stats command has been deprecated. We recommend you just run docker stats -a as a replacement.

• Running service daemon-logs without any other arguments no longer hangs; instead, it will now print the list of daemon logs you can pass in as arguments.

• Fixed an issue that caused the CLI to hang when querying validator status from the Beacon Node if you only have 1 CPU.

---

Modules

Constellation

• Added support for the Constellation module. It can be accessed from the hyperdrive constellation (hyperdrive cs) command. Please consult the Constellation documentation for information about how Constellation works and how to use the module.

Changes since RC1

• cs minipool create now saves the validator pubkey before submitting the transaction. Note that if you perform a nonce override due to an initial low gas amount, a new pubkey will be created and the original pubkey will remain even though it will be useless.

• cs minipool create now prints a warning about the expense of the transaction, and forces the user to confirm that they understand submitting with a high gas fee may have an excessively long breakeven time.

• Fixed the text on cs minipool exit.

• Added the cs network stats command to view info about the Constellation settings and deposit pool status.

StakeWise

• Updated StakeWise v3 Operator to v2.1.3.

• Submitting voluntary exits now requires the Beacon node to be synced first.

• Signed exits now use the latest finalized epoch, rather than the head epoch.

• Signed exits are now encrypted locally before sending them to nodeset.io, in addition to regular TLS encryption.

  • Thanks to @poupas for suggesting this enhancement.

• nodeset.io deployment names are no longer tied directly to the Hyperdrive network selection. They are now handled by individual modules.

• Increased the delay in the tasks loop during a failed ready-check from 10s to 15s.

---

Miscellaneous

• Generating validator deposit data now prints the BLS signature (and other intermediate artifacts) to the debug log for easy reference.

• hyperdrive service version now prints the versions of the enabled modules.

• Added better error handling to the nodeset.io request manager for a better explanation about what went wrong during an error.

• Updated to Go 1.22.

v1.1.0 Release Candidate 1

This is Release Candidate 1 for v1.1.0 of Hyperdrive. This adds some improvements the community helped us find with Beta 2, and adds enhancements to nodeset.io API support.
It is a required upgrade for all Beta Testers.

⚠️ NOTE: As usual, after upgrading, it's important to run hyperdrive service config and/or hyperdrive service start before running any other commands so the latest daemon services are pulled first.

Known Issues

• Lodestar v1.22.0 does not currently support x64 systems without the avx instruction set. This will be fixed in a later client release.

• Reth still has a configuration issue that prevents it from working with the StakeWise v3 Operator service. If you're a StakeWise operator, be advised that you will have to use a different client for now.

  • We believe we have a fix in place for this and will release it in a future version.

---

Hyperdrive Core

Clients

• Updated Nethermind to v1.29.0.

• Updated Teku to v24.10.1.

• Updated MEV-Boost to v1.8.1.

  • The release notes are for v1.8, but the Docker image being used is for v1.8.1.

Service

• Signed exits are now encrypted locally before sending them to nodeset.io, in addition to regular TLS encryption.
  • Thanks to @poupas for suggesting this enhancement.

CLI

• The CLI now ensures the existence of daemon JWT authentication keys on every command instead of just service install and service start for enhanced backwards compatibility.

v1.1.0 Beta 2

This is v1.1.0 Beta 2 of Hyperdrive. This adds some improvements and bug fixes the community helped us find with Beta 1.
It is a recommended upgrade for users of Beta 1.

⚠️ NOTE: As usual, after upgrading, it's important to run hyperdrive service config and/or hyperdrive service start before running any other commands so the latest daemon services are pulled first.

Known Issues

• Lodestar v1.22.0 does not currently support x64 systems without the avx instruction set. This will be fixed in a later client release.

• Reth still has a configuration issue that prevents it from working with the StakeWise v3 Operator service. If you're a StakeWise operator, be advised that you will have to use a different client for now.

  • We believe we have a fix in place for this and will release it in a future version.

---

Hyperdrive Core

Clients

• Updated Geth to v1.14.11.

  • Geth images are now published for both arm64 and amd64 once again, so all users can use this release.

• Updated Reth to v1.0.7.

• Updated Nimbus to v24.9.0.

Service

• The Hyperdrive daemon and module daemons now require JWT authorization for API access. The Hyperdrive CLI takes care of this automatically, but you can no longer run custom HTTP requests against the daemons without providing a JWT.
  • Thanks to @poupas for reporting this!

Config

• When running the service config TUI, if you have externally managed clients and have MEV-Boost enabled in Hyperdrive, you will now see a modal before the review screen warning you to ensure you have MEV-Boost turned on in your external Beacon Node configuration. If you don't have it enabled in your Beacon Node configuration, your Hyperdrive validators may not be able to propose blocks.

• New parameter: Hyperdrive and TX Fees > Client Timeout

  • This lets you control the timeout for HTTP requests to the Execution Client, Beacon Node, and nodeset.io.

CLI

• Performing a manual installation over v1.0.1 now functions correctly and no longer complains about missing the network resources files.

• The service stats command has been deprecated. We recommend you just run docker stats -a as a replacement.

• Running service daemon-logs without any other arguments no longer hangs; instead, it will now print the list of daemon logs you can pass in as arguments.

• Fixed an issue that caused the CLI to hang when querying the Beacon Node in certain situations.

---

Modules

Constellation

• Attempting to create a minipool with cs minipool create while having less than 1 ETH in your node wallet now correctly shows the missing amount of ETH.

• Registering for Constellation with cs node register now checks if your node is registered with nodeset.io first.

• Minipool stats from cs minipool status now print "Constellation portion" rather than "Your portion".

StakeWise

• Updated StakeWise v3 Operator to v2.1.3.

• The sw_operator container now starts properly.

v1.1.0 Beta 1

This is v1.1.0 Beta 1 of Hyperdrive. It adds support for nodeset.io's new v2 API and the much-awaited Constellation module.

⚠️ This is a BETA release and should only be used on the Holesky test network.
For users that want to run this public beta, please follow the beta installation instructions.
Note that the typical installation and update instructions will NOT update to this version, as it is not a stable release.

Known Issues

• If you're installing manually, please run the following command before going through the normal process:

  sudo mkdir -p /usr/share/hyperdrive && cd /tmp && wget https://github.com/nodeset-org/hyperdrive/releases/download/v1.1.0-b1/hyperdrive-install.tar.xz && tar -xf hyperdrive-install.tar.xz && sudo cp -r install/deploy/networks /usr/share/hyperdrive
  

  This will be fixed in Beta 2.

• The StakeWise Operator service may error out with a message in the logs stating 'NoneType' object has no attribute 'config'. We are looking into this and will have it resolved for Beta 2.

---

Hyperdrive Core

Clients

• Updated Besu to v24.9.1.

• Updated Geth to v1.14.8.

  • Note: while Geth v1.14.9 has been released, they haven't published a multiarch image to Docker Hub yet so we cannot include it in this release. Users are welcome to update manually once an image has been published.

• Updated Nethermind to v1.28.0.

• Updated Reth to v1.0.6.

• Updated Lighthouse to v5.3.0.

• Updated Lodestar to v1.22.0.

• Updated Nimbus to v24.8.0.

• Updated Prysm to v5.1.0.

• Updated Teku to v24.8.0.

• Updated Prometheus to v2.54.1.

• Updated Node Exporter to v1.8.2.

Service

• The daemon has been updated to use v2 of the nodeset.io API. It now manages a connection to the server that can be used by multiple modules, rather than the modules managing their own independent connections.

• The configuration system has been overhauled. The options in the Network dropdown are now loaded dynamically from disk rather than being built directly into Hyperdrive. As such, advanced users can now create their own configurations and use Hyperdrive on custom Ethereum chains for things such as development and testing.

  • For Linux systems, the network resources directory is /usr/share/hyperdrive/networks.

• The Eden MEV-Boost relay has been removed from the list of standard relay options because it has been merged with Titan.

CLI

• New command: hyperdrive nodeset can be used to manage your node's status with the NodeSet service.

  • hyperdrive stakewise nodeset registration-status has been moved to hyperdrive nodeset registration-status.
  • hyperdrive stakewise nodeset register-node has been moved to hyperdrive nodeset register-node.

• New command: hyperdrive constellation (hyperdrive cs) can be used to manage the new Constellation module.

  • See the section below for more information on this module.

---

Modules

Constellation

• Added support for the Constellation module. It can be accessed from the hyperdrive constellation (hyperdrive cs) command. Please consult the Constellation documentation for information about how Constellation works and how to use the module.
  • Currently only the Holesky test network has Constellation support.

StakeWise

• Submitting voluntary exits now requires the Beacon node to be synced first.

---

Miscellaneous

• Generating validator deposit data now prints the BLS signature (and other intermediate artifacts) to the debug log for easy reference.
• Updated to Go 1.22.
• hyperdrive service version now prints the versions of the enabled modules.

v1.0.1

This is release v1.0.1 of Hyperdrive. This is a point release with client updates and fixes for some bugs reported by our users from v1.0.0.

Upgrade instructions if you already have a previous version installed

Installation instructions if you don't have Hyperdrive installed yet

---

Clients

• Updated Nethermind to v1.27.0.

• Updated Reth to v1.0.0.

  • ⚠️ While this release means Reth is no longer considered in beta, we have not been able to verify if it fixes the compatibility issues with the StakeWise Operator. If you have the StakeWise module enabled, you will likely still need to use a different Execution client for now.

• Updated Lighthouse to v5.2.1.

  • ⚠️ NOTE: Reth v1.0.0 has a known bug that will cause the Lighthouse BN to fill its log with warnings. This will cause a slight performance decrease. See the patch notes for details.

• Updated Nimbus to v24.6.0.

• Updated Prysm to v5.0.4.

• Updated Teku to v24.6.1.

• Updated Grafana to v9.5.20.

• Updated Prometheus to v2.53.0.

Service

• Added support for IPv6 to the stack. It can be enabled with the new Enable IPv6 option in the Hyperdrive and TX Fees config section.
  • Note that if you already have Hyperdrive installed and running, you'll need to do the following after checking the box:
  • hyperdrive service down to remove the containers and Docker network while preserving chain data and personal information like your config, wallet, and validator keys
  • Follow the instructions at https://docs.docker.com/config/daemon/ipv6/#dynamic-ipv6-subnet-allocation to enable IPv6 in Docker itself
  • hyperdrive service start to regenerate the containers and Docker network which should all now have IPv6 support
• Community member @benv666 fixed a bug in the Beaconcha.in API key parser that would reject legal = characters.
• Community member @dabdab-wrt fixed a bug in the Beaconcha.in machine name, which doesn't allow for spaces.
• service start will now deploy the ~/.hyperdrive/metrics/modules folder prior to instantiating the Docker compose templates. This should resolve a rare permissions problem that occurred when migrating from v0.5.0 to v1.0.0.

CLI

• New command: hyperdrive service down runs a docker compose down operation, removing Hyperdrive's Docker containers and networks while leaving its volumes (i.e. chain data, Grafana and Prometheus databases, and StakeWise Operator database) intact as well along with the Hyperdrive configuration and your user data (such as your wallet and validator keys).
  • This is useful for anyone needing to adjust the Docker service, such as by enabling IPv6, without wanting to delete any data that's important or expensive to regenerate.
• The External IP address retriever used in some of the hyperdrive service commands has had its timeout reduced from 30 seconds to 3 seconds, which will make those commands much more responsive.
• Removed the dialog in the service config wizard that prompted for the externally-managed MEV-Boost URL if you selected externally-managed clients. That dialog only now properly only applies for locally managed clients.
• Fixed the text printed in wallet init and wallet recover to correctly refer to the hyperdrive sw ns register-node command.

StakeWise

• Moved the StakeWise v3 Operator's database from the user data folder into its own Docker volume.
  • ⚠️ NOTE: the first time you upgrade to Hyperdrive v1.0.1, the StakeWise Operator will regenerate its database which will cause high CPU load for a few hours. This is expected.
  • We will add a function to clean up the old database from your user data folder in a later release. If you'd like to do it manually, you can remove the folder at ~/.hyperdrive/data/modules/stakewise/0x646f5285d195e08e309cf9a5adfdf68d6fcc51c4 on Holesky or ~/.hyperdrive/data/modules/stakewise/0xe2aeecc76839692aea35a8d119181b14ebf411c9 on Mainnet.
• The CLI will no longer print an error about the clients not being synced if an error occured during a sync check, and will instead report the actual error.

v1.0.1-dev

⚠️ DO NOT USE
Not guaranteed to be stable - used for dev purposes

v1.0.0

This is release v1.0.0 of Hyperdrive. This version adds support for running on Mainnet with access to Gravita's StakeWise vault.

Upgrade instructions if you already have a previous version installed

Installation instructions if you don't have Hyperdrive installed yet

---

Clients

• Updated Besu to v24.6.0.

• Updated Geth to v1.14.5.

• Updated Reth to v0.2.0 Beta 9.

  • ⚠️ Reth still has a bug that prevents it from working with the StakeWise v3 Operator service. If you're a StakeWise operator, be advised that you will have to use a different client until the bug is fixed.

• Updated Lighthouse to v5.2.0.

  • Lighthouse no longer has different modern and portable images - there is only one image now, which dynamically determines your CPU features at runtime.
  • ⚠️ If you use Nethermind with Lighthouse, Nethermind v1.26.0 has a known bug that will cause the Lighthouse BN to fill its log with warnings. The Lighthouse team recommends reverting to Nethermind v1.25.0 until v1.27.0 is released, or ignorning the warnings. See the patch notes for details.

• Updated Lodestar to v1.19.0.

• Updated Teku to v24.6.0.

  • ⚠️ Teku v24.6.0 has a bug that prevents the Beacon Node from starting up if your node is not behind a NAT system (such as a home router). We have mitigated this in Hyperdrive by directly adding the --p2p-advertised-ip to Teku, so it should function properly for all of our node operators. However, if you still experience problems with it, you should downgrade to v24.4.0 until they provide a fix.

Service

• Added support for Mainnet.

• Fixed a race condition that caused hyperdrive service start to incorrectly state your node wasn't registered with your NodeSet account if the service was stopped before running the command.

  • Hyperdrive will now report that it couldn't check your node's registration status if the check failed, rather than simply stating your node wasn't registered.

• Fixed a bug that prevented Lighthouse's VC from working when MEV-Boost was enabled.

• The MEV-Boost relay selection mode has been adjusted in preparation for mainnet.

  • "Unregulated" relays have been removed from Hyperdrive's built-in list due to liability concerns.
  • The selection modes are now All and Manual; All automatically enables all of the built-in relays (similar to the old Profile mode), and Manual allows you to individually select the built-in relays you want to use (similar to the old Relay mode).
  • The new "Custom Relays" box allows you to add additional URLs for any other relays you'd like to use. Custom relays can be anything as long as they are on the same Ethereum network as your node. The ETHStaker relay list is a good resource to explore and compare all of the relay options.

CLI

• New command: stakewise nodeset generate-deposit-data will generate and print the deposit data for all of your node's validators without uploading it to NodeSet. Useful if you want to view your deposit data, such as for custom verification.
• wallet recover will now check if the wallet is registered with your NodeSet account for convenience.

StakeWise

• Added support for the Gravita vault on Mainnet.
• Updated to Operator v1.3.2.
  • This version no longer spams the logs with warning about empty deposit data or missing validator keys, which were not applicable to NodeSet operators. It now only logs relevant events.
• Added support for 0xSplits v2.0, which is now used for StakeWise rewards on Holesky and will be used on Mainnet.
• The daemon's task loop now waits for your node to be registered before trying to perform any actions that require registration.
• Uploading deposit data to NodeSet now requires both your Execution Client and Beacon Node to be synced.
• Switched to the production nodeset.io URL on Holesky nodes.

v0.5.0 Beta 1

This is v0.5.0 Beta 1 of Hyperdrive.

⚠️ Important Note:

MEV-Boost (new feature, see the notes below) is now enabled by default. If you had a previous beta release installed, please run hyperdrive service config to select at least one profile or relay (or disable MEV-Boost if you don't want to use it). If you attempt to start the service without doing this, you will get an error about MEV-Boost being enabled without any relays.

Upgrade instructions if you already have a previous version installed

Installation instructions if you don't have Hyperdrive installed yet

---

Known Issues

• Reth is currently incompatible with the StakeWise Operator container. If you've enabled the StakeWise module, you will need to use a different Execution client until the incompatibility is fixed.

Clients

• Updated Besu to v24.5.2.

• Updated Nimbus to v24.5.1.

• Updated Grafana to v9.5.19.

• Updated Prometheus to v2.52.0.

• Updated Node Exporter to v1.8.1.

Service

• Added MEV-Boost support back to Hyperdrive with several built-in relays. MEV-Boost will work properly on the Holesky testnet, so feel free to enable it and explore the options.
• Added support for the new NodeSet service authentication system.
  • If you're a new StakeWise user starting with v0.5.0 Beta 1, you'll need to whitelist your node's address on the nodeset.io user account page first to declare ownership. Next, confirm you own the node with hyperdrive stakewise nodeset register-node (or hyperdrive sw ns r for short).
  • If your node was already registered from a previous beta, you don't need to do anything.
  • For more information on the new system and how to use it, take a look at the node authorization docs page.

CLI

• New command: hyperdrive stakewise nodeset registration-status gets the node's registration status with the NodeSet service.
• New command: hyperdrive stakewise nodeset register-node will register your node with your NodeSet user account if it's not already registered.

StakeWise

• Updated StakeWise Operator to v1.3.0.

v0.4.3-dev

DO NOT USE - UNSTABLE DEV