today our nodes got many errors in log and went OutOfSync state what can cause this behavior?
EDIT: my node is again in sync after updating to v1.7.15-0-gc1de9106 so being on the old version can causes errors. Nvm after updated the same problem most nodes went OutOfSync, most of it happend during the epoch shift
@Validators you experience the same?
The team is already checking, seems like the old version should be updated asap by all Validator runners.
Hi Klever Team, could you please answer? Most of the nodes are jailed
It doesn’t help.
I updated my node but it still keeps losing blocks
The new version seems to help, but it is still unstable. My nodes still lose blocks from time to time.
as long as you have updated all nodes to the very latest version, you are fine.
once all nodes are updated everything will get fine again.
Seeing the same behavior. Root cause is a mixed-version set: some validators on v1.7.15 (faster KVM / tighter timeouts) and others on v1.7.14. With >25% of consensus still on 1.7.14, those slower nodes were timing out on a specific tx path and rejecting it, while the 1.7.15 nodes accepted it. That split caused prevote/precommit churn, outOfSync on 1.7.14, and consensus instability (blocks proposed then bounced).
Fix: upgrade all validators to v1.7.15 and ensure primary + fallback run the exact same version. Don’t mix versions.
It is weird as KUNAIKASH-4 shows fallback at 1.7.14, but it is running on 1.7.15 and still got jailed even with no mixed versions. I have checked all logs and bot mainnode and fallback runs on 1.7.15
Not totally true, even @CryptoJaeger had updated all nodes and have some jailed on the newest version so there is more on going
Guys, do you have any updates? The problem is still here.
What problem do you face, can you show?
Nodes are periodically running out of sync, even on the latest version.
