Core Devs Call - 2023/10/18
· 3 min read
Gnosis Core Devs Call Notes
Welcome to the Gnosis Core Devs weekly gathering. Every Wednesday, key members from the Gnosis team, contributors, and various team representatives convene to discuss, collaborate, and update one another on the Gnosis ecosystem's progression.
Participants represent teams:
Erigon, Gateway, Nethermind, Geth, Gnosis DevOps, Gnosis Core Devs, Gnosis Comms team.
With a diverse set of voices present, our discussions are rich, multifaceted, and aim to foster innovation within the community.
Missed the meeting? Catch the full recording on YouTube.
October 18, 2023
EL
-
Nethermind:
- Finished some Gnosis-specific features
- Configurable EIP-4844 configurations
- Cancun merged to main branch
- Kickstarted an initiative to rework the
eth_getLogs
system to optimize it
- Finished some Gnosis-specific features
-
Erigon:
- Sync issues (OOM on 32 GB nodes also running CL)
- Exclusively happens in Docker, GOMAXPROC should be a workaround (4 instead of 24 for example)
- High disk usage (400 GB with
--prune=hrtc
vs 170 GB on Nethermind for the standardgnosis
preset) - Fairly unreliable (I have to restart it every few days at this point, maybe too many attached validators? I miss quite a few epochs on mainnet too but don’t need to restart at least.)
- Configurable EIP-4844 configurations are merged
- Working on Erigon 3
- Sync issues (OOM on 32 GB nodes also running CL)
-
Geth:
- Tried the static peers and it seems to work better
- Still not getting a connection
- The old branch systematically uses eth/66 and that works
- The new branch gets disconnected when trying to connect to eth/66
Dencun
- Config
- Max churn limit of 2
- Target and max blobs per block
- Security first: 1/2
- Thoughts of EL devs?
- https://github.com/gnosischain/specs/pull/24
- BLOB_GASPRICE_UPDATE_FRACTION=1112826
Chain Infra
- Gateway
- Discovered a bug in the load balancer that resulted in 0.5 to 1% of errors at the current load
- Sudden increase in
eth_getLogs
, which are typically quite slow on the nodes, which were subsequently falling behind - The solution was to account for this slowness
- Sudden increase in
- Discovered a bug in the load balancer that resulted in 0.5 to 1% of errors at the current load
Gnosis
- The public RPC was down during the weekend, but that was because of Gnosis.
Hive Tests
- Solved the configuration issues for Dencun
- Working on an error when sending blobs
Devnet
- Has Erigon joined?
thorax/erigon:docker_gno_dencun_devnet_0
--chain=gno_dencun_0
- Issues with the internal CL
- Not tested yet
Research
- Automatic sync tests