Skip to content
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

tracing regression in erigon 3 #10444

Closed
banteg opened this issue May 22, 2024 · 4 comments
Closed

tracing regression in erigon 3 #10444

banteg opened this issue May 22, 2024 · 4 comments
Assignees
Milestone

Comments

@banteg
Copy link
Contributor

banteg commented May 22, 2024

System information

Erigon version: erigon version 3.00.0-dev-be889f62

OS & Version: Linux

Commit hash: be889f6

Erigon Command (with flags/config):

Consensus Layer: erigon internal

Consensus Layer Command (with flags/config):

Chain/Network: ethereum

Expected behaviour

correct traces are returned

Actual behaviour

rpc errors out with "insufficient funds for gas", as well as "nonce too low".

this also affects receipts since erigon now generates them ad hoc.

Steps to reproduce the behaviour

i ran cryo traces -a on a full block range, it has reported errors in 23 chunks (23k blocks).

then i further isolated the issue by running trace_block against these block ranges, and narrowed it down to 6105 blocks.

then i ran trace_transaction against every transaction in these blocks and wrote down the ones that have returned an error. however, i noticed that all the txs after the first failing transaction are failing. so i'm only providing the first failing tx for each block i got from trace_block.

the full list can be found here. it should be helpful for producing and verifying the fix.

@awskii
Copy link
Member

awskii commented Jun 6, 2024

thank you for valid test, trying to reproduce now, could take day or two.
Previously such errors happened during HisotryExecution and most of the time meant that bad history file provided. But i assume this is happening on available snapshots on webseeds, so i am syncing them now.

UPD: issue persists and roots down to incorrect snapshot .ef files for given region. Log was helpful - some blocks are going by long increasing ranges which gave a clue.

@AskAlexSharov
Copy link
Collaborator

fixed by #10746 but still we need need release 2 new .ef files.
also removing historical state cache reduced perf, created ticket to re-introduce cache #10747

AskAlexSharov added a commit that referenced this issue Jun 18, 2024
for #10444
it also removing use of `state.NewCachedReader` seems need re-introduce
it in E3 - to see perf impact. Creating ticket for it:
#10747
@AskAlexSharov
Copy link
Collaborator

fixed. but need (git pull and re-sync). and we didn't work on #10747 yet

@banteg
Copy link
Contributor Author

banteg commented Jun 27, 2024

this appears fixed as of 900f730

@banteg banteg closed this as completed Jun 27, 2024
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

4 participants