v1.11.2

This is v1.11.2 of the Smartnode.
It is a recommended update for node operators using Geth, Teku, Lodestar and Nethermind that haven't updated manually to the latest versions yet, as those new client versions include some important optimizations and bug fixes. It is also an important update for oDAO nodes with new features described below.

To upgrade, follow the Smartnode upgrade instructions.

---

Client Updates

• Nethermind updated to v1.24.0 - This version requires less bandwidth and computer resources (up to 8% less) when syncing and serving. ⚠️ This version is NOT downgradable. Once your node is upgraded to this version or synced from scratch on it, you won’t be able to downgrade to lower versions.
• Teku updated to v23.12.1 - ⚠️ Breaking change: By default, Teku won't allow syncing from genesis, users should use a checkpoint sync URL when starting a new node.
• Lodestar updated to v1.12.1 - Improvements and bug fixes (https://github.com/ChainSafe/lodestar/releases/tag/v1.12.0, https://github.com/ChainSafe/lodestar/releases/tag/v1.12.1)
• Geth updated to v1.13.6 - gas estimation and logging improvements (https://github.com/ethereum/go-ethereum/releases/tag/v1.13.6)

Smartnode Changes

• oDAO nodes will start to submit price updates to the Scroll network.
• oDAO nodes will continue the rewards voting process in case the IPFS upload fails for any reason. In this case, the rewards file CID will be calculated locally.
• Changed the default Lighthouse P2P QUIC port to 8001 to avoid port collisions. ⚠️ If you're using Lighthouse make sure to check this configuration setting and if your chosen port is being properly forwarded to your node IP on your router.
• The Smartnode will now check if all the ports selected are unique before saving the configuration.
• Teku now supports doppelganger protection.
• Users can now see information about the Rescue Node add-on using node status even if their Execution Client is offline or syncing. Thanks to Patches for this contribution!

v1.11.2-dev

This is a development prototype for v1.11.2.
This is under development and will change frequently, DO NOT use this unless you were instructed to do so or are willing to break things.

v1.11.1

This is v1.11.1 of the Smartnode.
It is a recommended update for all node operators due to client upgrades and new features.

To upgrade, follow the Smartnode upgrade instructions.

---

Client Updates

• Updated Besu to v23.10.2
• Updated Geth to v1.13.5
• Updated Nethermind to v1.23.0
• Updated Nimbus to v23.11.0
• Updated Prysm to v4.1.1
• Updated Teku to 23.11.0
• Lodestar was not updated to the latest version due to a bug report.

Smartnode Changes

• Added the Rescue Node add-on. You can now enable/disable the Rescue Node directly from the Smartnode TUI. Thanks Patches for this contribution!
• Added Rescue Node info to the node status command. This command now shows you for how long your Rescue Node credentials are going to be valid.
• New installations will use Path-Based State Scheme (PBSS) as the default for Geth. Thanks objectObject for this contribution!
• Added the option to change the BN P2P QUIC port settings. This is not mandatory but users who want to benefit from using the faster QUIC protocol will need to forward this port to the node IP address on your router configuration. The default port is 9002 (UDP) and the QUIC protocol is only used by Lighthouse (v4.5.0+).

Important Note

This release exposes an additional port to use the QUIC protocol for P2P connections. By default, this port is 9002, if you've already been using this port for regular P2P connections (on a fallback or second node), you'll need to change the QUIC port to use a new port number.

1. run rocketpool s c
2. go to the ETH2 section, and change the "P2P QUIC port" to a different port number, like 9003, then press enter and save your changes.

v1.11.1-dev

This is a development prototype for v1.11.1.
This is under development and will change frequently, DO NOT use this unless you were instructed to do so or are willing to break things.

v1.12.0 Development Prototype

This is a development prototype for v1.12.0.
This is actively under development and will change frequently, DO NOT use this unless you were instructed to do so or are willing to break things.

v1.11.0

This is v1.11.0 of the Smartnode.
It is a recommended update for all node operators due to client upgrades and new features.

To upgrade, follow the Smartnode upgrade instructions.

---

Important Dates

• The Oracle DAO must update to v1.11.0 or later before epoch 238214 (October 26, 2023 1:29 AM UTC) to maintain consensus for the next rewards interval.

New Test Network: Holesky

This version of the Smartnode supports the Holesky test network.
Holesky is quite similar to the Prater test network; there is an Execution Layer explorer and a Beacon Chain explorer, you can create practice nodes with practice minipools using free "fake" ETH and RPL, and we have support for getting both of them via our Discord server.

⚠️ Prater Deprecation

With this release, the Prater test network is deprecated and Holesky will replace it as Rocket Pool's long-term test network. This is expected to be the last version of the Smartnode that will have full support for Prater.

If you have a node on Prater, please exit your testnet minipools to gracefully remove them from the validator set. Follow our guide to learn how to do this.

Client Updates

• Updated Besu to v23.10.0.

• Updated Geth to v1.13.2.

• Updated Nethermind to v1.21.0 on Mainnet.

  • NOTE: Nethermind has been intentionally held back to v1.20.4 on the test networks due to a bug that causes it to report incorrect syncing progress.

• Updated Lighthouse to v4.5.0 on Mainnet, and a custom build on the test networks due to a bug.

• Updated Lodestar to v1.11.3.

• Updated Nimbus to v23.9.1.

• Updated Prysm to a custom build on the test networks due to a bug.

• Updated Teku to v23.10.0.

• Updated Grafana to v9.4.15.

• Updated Prometheus to v2.47.1.

Smartnode Changes

• Added support for the Holesky test network.

• Added support for Path-Based State Scheme (PBSS) to Geth.

  • PBSS is a new mode of Execution Layer storage that removes the need for manual pruning. Instead, it constantly prunes itself automatically in real-time.
  • To enable it, check the box in the "Execution Client (ETH1)" page of the service config TUI called "Enable PBSS".
  • ⚠️ NOTE: If you enable it, you will need to delete your Geth database and resync it. You can do this with the rocketpool service resync-eth1 command.
    If you don't have a fallback node specified, your node will be offline and cannot attest during this time until Geth finishes resyncing!

• The "Cache Size" parameter for Geth has been removed, as it is no longer recommended with Geth v1.13.

• Removed Blocknative from the list of MEV-Boost relays, as the maintainers have shut it down.

• rocketpool service logs now supports aliases for various containers:

  • ec, el, and execution will now show the logs for eth1 (e.g., rocketpool service logs ec)
  • cc, cl, bc, bn, beacon, and consensus will now show the logs for eth2 (e.g., rocketpool service logs cc)
  • vc will now show the logs for validator (e.g., rocketpool service logs vc)

• Implemented Rewards Spec v7, which is expected to be used for rewards interval 15.

  • In this update, minipools that have entered staking status (finished their scrub check and submitted the second deposit) will now be eligible for monthly RPL rewards, even if they haven't been activated on the Beacon Chain yet.

• You can now add custom scrape jobs to Prometheus by including scrape job YAML files in the .rocketpool/extra-scrape-jobs directory.

  • Thanks to @jshufro for this submission.

Bug Fixes

• Fixed a bug with the Rolling Records cache that didn't update information for multiple minipools that share the same pubkey (i.e., solo stakers that attempted migration but timed out, then tried a second time).

Misc Changes

• Nethermind's cache size has increased from 256 MB to 512 MB on systems with 8 GB of RAM, as the new version requires a larger cache to function.

• Removed the deprecated --network flag from the rocketpool service install command.

v1.11.0 Development Prototype

This is a development prototype for v1.11.0.
This is actively under development and will change frequently, DO NOT use this unless you were instructed to do so or are willing to break things.

v1.10.2

This is v1.10.2 of the Smartnode.

It is a recommended update for node operators using Besu, Geth, Nimbus, Prysm, or Teku that haven't updated manually to the latest versions yet, as those new client versions include some important optimizations and bug fixes.

For everyone else, it is a low priority.

To upgrade, follow the Smartnode upgrade instructions.

Client Updates

• Updated Besu to v23.7.1.

• Updated Geth to v1.12.2.

• Updated Nimbus to v23.8.0.

  • Nimbus will now only subscribe to 2 subnets total, similar to Lighthouse and Lodestar. This significantly reduces its CPU usage and bandwidth consumption, but may result in a small number of missed attestations for users with many validators.
  • If you'd like to subscribe to all of the subnets again to minimize the missed attestations, add the --subscribe-all-subnets flag to the "Additional Beacon Client Flags" field of the service config TUI. This is only recommended for users with powerful node hardware that can handle the load, so you may want to consider running without it first for comparison.

• Updated Prysm to v4.0.8.

  • NOTE for Docker / Hybrid users: Prysm's canonical image uses the incorrect version (it reports v4.0.7), so this Smartnode will come with images we built using their published binaries.
  • If you want to use the canonical image and ignore the mismatched version, please use the following Docker containers (noting they only work for x64 modern / non-portable systems):
  • BN: prysmaticlabs/prysm-beacon-chain:HEAD-0ca364-debug
  • VC: prysmaticlabs/prysm-validator:HEAD-0ca364-debug

• Updated Teku to v23.8.0.

• Updated Node-Exporter to v1.6.1.

• Updated Prometheus to v2.46.0.

Smartnode Changes

• The "bloXroute Ethical" relay has been removed from the MEV-Boost options since it no longer exists.

  • As this was the only option for relays that mitigated sandwich bundles, the "No Sandwiching" profiles have been removed as well.

• rocketpool minipool distribute-balance changes:

  • The minipool options are now sorted from most-rewards to least-rewards.
  • New option: use --threshold that will hide minipools with lower rewards than a specified threshold amount.
  • For example, rocketpool minipool distribute-balance --threshold 0.1 will hide all minipools that have less than 0.1 ETH in pending rewards.

• Commands like rocketpool minipool status now work properly on nodes with more than 1000 minipools.

Bug Fixes

• Lodestar's VC will now properly request blocks to sign from MEV-Boost when enabled.

v1.10.2 Development Prototype

This is a development prototype for v1.10.2.
This is actively under development and will change frequently, DO NOT use this unless you were instructed to do so or are willing to break things.

v1.10.1

This is v1.10.1 of the Smartnode.

It is a medium priority update for node operators Lighthouse or Lodestar, for anyone that generates their own rewards trees (including the Oracle DAO), and for anyone that wants to expose their Prometheus API port.

For everyone else, it is a low priority.

To upgrade, follow the Smartnode upgrade instructions.

Client Updates

• Updated Lighthouse to v4.3.0.
  • NOTE: This version of Lighthouse has an important change - it will now only connect to two subnets on the Beacon Chain total, rather than one subnet per validator. This will considerably reduce its bandwidth usage but may result in slightly reduced block rewards for users with a large number of validators. To force it to subscribe to all subnets again, add the --subscribe-all-subnets to the "Additional BN Flags" field in the service config TUI.
• Updated Lodestar to v1.10.0.
  • NOTE: Lodestar has the same behavior change as above.

Important Dates

• If already opted into Rolling Records, users (including the Oracle DAO) should update to v1.10.1 or later by epoch 225614 (August 31, 2023 05:29:59 AM UTC) to ensure that the records are properly created (see the Bug Fixes section below).

Smartnode Changes

• rocketpool node send now accepts an arbitrary token address as the token-to-send instead of just rpl, fsrpl, eth, or reth so you can use it to safely move tokens that are currently in your node wallet directly via the CLI.

Bug Fixes

• Finalized minipools are now correctly subtracted from the number of staking minipools in Grafana.

• Fixed a bug that prevented the Smartnode from starting when exposing Prometheus's HTTP port was enabled.

  • Thanks to @0xfornax for this fix!

• Rolling records will now use the proper starting slot when creating new records for subsequent intervals after reporting if one or more of the new interval's first slots were missing.

  • The Smartnode will now invalidate checkpoint files that were made using previous Smartnode versions without this fix. Upon upgrading, it will invalidate any checkpoint files you may have already generated from v1.10.0 and start building a new record from the beginning of the interval. This is expected.

Misc Changes

• Added an Oracle Dao duty to report prices to the Base L2 chain.