-
Notifications
You must be signed in to change notification settings - Fork 842
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Besu shows INVALID block at 18947893
on mainnet
#6357
Comments
Same issue |
Besu/Lodestar here same issue. Tried:
Tries to resync and hits the same issue. JVM error trace is:
|
Same here (besu+teku) Upgraded from 23.10.2 to 23.10.3, with no luck. |
Same thing here. Have restarted Besu multiple times, as well as Teku, but it is still stuck on:
Besu version v23.10.2 |
Same issue here, but luckily recovered after upgrading from |
Can you describe further how you resolved please? |
What exactly you did after upgrading? I'm having same issue. |
As @yorickdowne mentioned, some Besu recover (see attached logs in his initial post), that's what happened to me. Logs:
|
Attempts with |
besu/v23.10.0/linux-x86_64/openjdk-java-17
|
i experienced the same. my besu stopped at 12:47 CET (UTC+1)
Besu 23.10.2 EDIT: both of my besu machines actually died. :-( |
If feeling adventurous, this takes 3+ hours and may or may not help: Enable debug with It'll force a backwards sync and world state sync. We do not know this resolves it, and it does take 3+ hours depending on hardware. |
i am not sure if this is needed to start the backward sync.
After restarting the service. This is besu 23.10.2. EDIT: or is this different to what you suggested? |
Different. Some people have a backward sync just starting on its own, we don't know yet what anyone can do to trigger it. The debug_resyncWorldState forces a more complete resync, and we don't know yet whether it'll help: For people where backward sync doesn't "just" kick in If you are seeing a backward sync already let it complete, it should fix it |
Working fine for me with no intervention. I can see from the logs that Besu v23.10.3 did fail to import 18,947,893 and subsequent blocks, but then did a backward sync and recovered all by itself. I'm running Nimbus v23.11.0 as my CL client if that matters. |
From a Besu dev: "A possible way to force a backward sync, that at least works with Teku, is to stop the CL client, delete its beacon db, be sure it is configured to checkpoint sync, and restart" For Eth Docker users this is |
This cant be real, i actually missed a proposal. Im pissed now. After 5 months of nothing. EDIT: What did i learn from this? Dont have the same software on your backup node... I swapped the EL to Erigon on my Backup node. |
Hey Besu team, Traces from Nethermind node for a block 18947893. Let me know if we can help you in any other way. |
Had the same issue on 23.10.2 and 23.10.1, updating to 23.10.3 fixed it. |
I recommend to follow this https://discord.com/channels/905194001349627914/938504958909747250/1193223408058630295 for people having this issue |
Discord is an incredible mess (or the app is totally messed up for non subscribers). Could you please copy paste the relevant text? Thanks |
I did it, it doesn't solve the problem on my computer. The backward synch happened but Besu is still locked in the bad block loop. |
Besu world state issue recap. Around 2024-01-06T11:29:36 UTC, Besu started reporting errors like this one: The cause of this issue are still under investigation, but after collecting feedback from user and testing some options, there are some workarounds to recover you node, until a proper fix is released. Recovering options:
https://hyperledger.jfrog.io/artifactory/besu-binaries/besu/23.10.3/besu-23.10.3.tar.gz / sha256 73c834cf32c7bbe255d7d8cc7ca5d1eb0df8430b9114935c8dcf3a675b2acbc2
|
Same here. Now I initiated a world state resync. I think it takes a few hours, I will report the results. |
Updating to 23.10.3 and restarting did the trick for me with a backward sync which was failing on .2 |
My validator is back up after a world state sync. |
We have just created a hotfix. Even if your node works well it is better to update Besu to avoid future issues. If your node already works well, no need to do anything else, just an update and use hotfix |
Fantastic reactivity, especially over a weekend. Thanks and congratulations to the team! |
Resolved with 23.10.3-hotfix. Thanks! |
Description
Besu is showing an invalid block at
18947893
on mainnet. This has been reported across the ecosystem on23.10.1
,23.10.2
,23.10.3
and24.1-SNAPSHOT
Most Besu do not recover from this. I have one that did. Will add logs.
Logs (if a bug)
Logs of Besu that recovered
Logs of Besu that did not recover
In failure case, Besu does not backwards sync and stays failed, here's current logs:
The text was updated successfully, but these errors were encountered: