feat(engine): set eth_syncing = true if pruner is active
#4063
Conversation
Codecov Report
... and 11 files with indirect coverage changes
Flags with carried forward coverage won't be shown. Click here to find out more.
|
|
@mattsse I think it's about time to detach txpool from the syncing state and create another control flag for it, because it's not ideal that we don't accept incoming pool transactions whenever we prune |
| } | ||
| EnginePruneEvent::TaskDropped => { | ||
| error!(target: "consensus::engine", "Failed to receive spawned pruner"); | ||
| return Some(Err(BeaconConsensusEngineError::PrunerChannelClosed)) | ||
| } | ||
| EnginePruneEvent::Finished { result } => { | ||
| trace!(target: "consensus::engine", ?result, "Pruner finished"); | ||
| self.sync_state_updater.update_sync_state(SyncState::Idle); |
There was a problem hiding this comment.
this is okay, and will controlled by the pipeline if we need to resync again
| @@ -1599,13 +1599,15 @@ where | |||
| EnginePruneEvent::Started(tip_block_number) => { | |||
| trace!(target: "consensus::engine", %tip_block_number, "Pruner started"); | |||
| self.metrics.pruner_runs.increment(1); | |||
| self.sync_state_updater.update_sync_state(SyncState::Syncing); | |||
There was a problem hiding this comment.
I'd like a note here why we're doing this, basically, because prune+fcu/engine are mutually exclusive so we're not making progress while we're pruning
agree, will look into it |
Resolves #4056