Dmitrii Shmatko
Dmitrii Shmatko
It's very unusual case btw I'm not sure it could happen outside test
I think I got a clear view on this finally and think it could be split into several PRs: 1. Teku API to get finalized snapshot 2. Teku flag to...
Following tests are failing from unmerged PR https://github.com/ethereum/hive/pull/606: - ~~`block-latest-safe-finalized`~~ - looks like EL doesn't know block label `safe`, not a Teku issue Test `latest`, `safe`, `finalized` block labels on...
New failed test, info will be added ~~`unknown-pow-parent-transition-payload`~~ - false-positive, just a bit of `DEBUG` logging flakiness
We have some flakiness on `eth2-testnet` test, see here https://hivetests2.ethdevops.io/ , filter by "teku". `ethereumjs` is buggy but we should pass both `go-ethereum` and `nethermind` but Teku fails in 50%....
Current state of hive tests: `eth2-testnet`: - Most failures are due to not enough participation `beacon 0: participation not healthy (got: 0.93, expected: 0.95)`. This error was previously seen mostly...
Yeah, EL previously gave us this block, it's a good point. > We probably should identify this particular case and log an error saying the EL reported the selected transition...
but what if, say ttd = 1000 ``` pre-terminal (999) - terminal 1 (1001) \- terminal 2 (1002) - arrives a bit later ``` And we are sending fcU with...
yeah, CL could have requirement for hash, for example, right
We have the latest version javalin with #5235 and looking at the original issue, it was not solved yet