skunert
commented
3 months ago Hi,
I investigated this issue a bit today. From the collator side everything seems to be okay. However, for some reason the submitted blocks fail to validate on the rococo validators.
I extracted this from one validators log:
2024-08-06 18:09:12.752 INFO tokio-runtime-worker parachain::candidate-validation: Failed to validate candidate para_id=Id(2043) error=Invalid(WorkerReportedInvalid("execute: Execution aborted due to trap: wasm trap: wasm `unreachable` instruction executed
WASM backtrace:
error while executing at wasm backtrace:
0: 0x620fb6 - neuroweb_runtime.wasm!rust_begin_unwind
1: 0xc76a - neuroweb_runtime.wasm!core::panicking::panic_fmt::h8b4135fb81936008
2: 0x3466d3 - neuroweb_runtime.wasm!core::panicking::panic_display::h287c324f965c61f8
3: 0x346147 - neuroweb_runtime.wasm!frame_executive::Executive<System,Block,Context,UnsignedValidator,AllPalletsWithSystem,COnRuntimeUpgrade>::apply_extrinsics::{{closure}}::panic_cold_display::h87e29365917e1814
4: 0x34410c - neuroweb_runtime.wasm!frame_executive::Executive<System,Block,Context,UnsignedValidator,AllPalletsWithSystem,COnRuntimeUpgrade>::execute_block::h307560923d65cd10
5: 0x594ddd - neuroweb_runtime.wasm!<cumulus_pallet_aura_ext::BlockExecutor<T,I> as frame_support::traits::misc::ExecuteBlock<Block>>::execute_block::h17a9647f9efaccce
6: 0x1a0440 - neuroweb_runtime.wasm!environmental::local_key::LocalKey<T>::with::hce9d98b3bbf43917
7: 0x29dbcf - neuroweb_runtime.wasm!cumulus_pallet_parachain_system::validate_block::implementation::validate_block::h69a5a40d817377d2
8: 0x4a6723 - neuroweb_runtime.wasm!validate_block
9: 0x6d94dd - neuroweb_runtime.wasm!<wasm function 8475>"))What the reason for this validation failure is exactly I am not sure. Maybe something went wrong with the runtime upgrade? When was it deployed?
Mihajlo-Pavlovic
alexggh
sandreim
SBalaguer
NZT48
bkchr
Is there an existing issue?
Experiencing problems? Have you tried our Stack Exchange first?
Description of bug
At NeuroWeb we are experiencing an issue with NeuroWeb Testnet connected to the Rococo relay chain.
As per @skunert request, we are opening a new issue. Initially, we reported this to issue #1202.
After runtime upgrade where we changed dependencies from v0.9.40 to v1.9.0 (https://github.com/OriginTrail/neuroweb/pull/86), upgrade was successful but after 30+ blocks, the chain could not produce a block.
All collators have block 4238138 as their best block and 4238137 as finalized, but they are constantly trying to create 4238138 block again and are stuck in a loop. With following logs, with
aura::cumulus=trace,parachain=debug,txpool=debug:We assume that it is caused by forks on relay chain (Rococo) which are not handled properly by collators on NeuroWeb Testnet. But we would like to get confirmation and better understanding of this issue, so we are sure how to avoid having it on NeuroWeb Mainnet, alongside finding the way to unblock NeuroWeb Testnet.
Steps to reproduce
No response