You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
After pruning was moved into the background in v1.x and v0.38.x-experimental, tests relying on information obtained by querying a node for the first and last available height are becoming flaky. A majority of these tests are gRPC and RPC related.
While there is no 100% certain way to impose determinism, we have already fixed a number of these issues. This issue serves as a reminder on the cause for these failures in case we start seeing an increase in test failures related to blocks not being found at certain heights.
The issues and PRs addressing these problems that have been closed and merged are:
After pruning was moved into the background in v1.x and v0.38.x-experimental, tests relying on information obtained by querying a node for the first and last available height are becoming flaky. A majority of these tests are gRPC and RPC related.
While there is no 100% certain way to impose determinism, we have already fixed a number of these issues. This issue serves as a reminder on the cause for these failures in case we start seeing an increase in test failures related to blocks not being found at certain heights.
The issues and PRs addressing these problems that have been closed and merged are:
Some of these tests might not require checking the first height in the first place.
Bug Report
Setup
CometBFT version (use
cometbft versionorgit rev-parse --verify HEADif installed from source):Have you tried the latest version: yes/no
ABCI app (name for built-in, URL for self-written if it's publicly available):
Environment:
node command runtime flags:
Config
What happened?
What did you expect to happen?
How to reproduce it
Logs
dump_consensus_stateoutputAnything else we need to know
The text was updated successfully, but these errors were encountered: