When the Rivet team started work on EtherCattle, we did the initial development work against the Goerli testnet, so it saddens us to hear that the the Goerli testnet is going to be deprecated in the coming months. For those of you concerned about having Goerli support, rest assured that Rivet will support Goerli for as long as the network is stable, and we’ll offer a read-only endpoint to it for a couple of months beyond that.

But for those of you ready to say goodbye to yesterday’s news and move on to the next big thing, Holesky is here!

Since Holesky is a new network and we don’t have customers relying on it for critical operations yet, we’ve decided to use it to beta test a few new features to our stack.

On the Flume side, the changes aren’t very user facing, but we’re testing out a new method of building out a Flume database. We’ve tested this database quite extensively, but if you should run across anything that defies expectations, please reach out to our support team on Our Discord and let them know what you’ve found. We’ll be happy to comp you some RPC credits if you find any significant anomalies.

On the Cardinal EVM side, we’ve got a feature lots of users have been asking about: Archive mode. Since we launched Cardinal in 2022, users have been able to get low cost, low latency data about the state of the blockchain from the most recent 128 blocks. That architecture, however, didn’t give us an easy way to offer data going back further. With Holesky, we’re beta testing Cardinal’s new archival storage engine, meaning users can query for state data from anywhere in the history of the blockchain. Once again we’ve run fairly extensive tests to ensure the integrity of the data we’re serving, but this is a whole new storage engine for Cardinal, so it’s possible some issues will crop up. If you happen across any of those issues, please reach out to our support team.

Once testing on Holesky has had time to flush out any issues, we plan to start rolling out archive mode to more networks. It will take some time to build the archival Cardinal databases, but let us know what networks we support that you’d most like to see archive support for.