Investigate Prysm Errors During Sync

[siladu]

During 24.10.0-RC1 burn-in, I received many prysm errors during sync, of two types...
(1)

{"error":"received 0 payload bodies from engine_getPayloadBodiesByRangeV1 with count=1 (start=20932021): engine api payload body response is invalid","handler":"beacon_blocks_by_root","level":"error","msg":"Could not get reconstruct full block from blinded body","prefix":"sync","time":"2024-10-10T01:48:25Z"}

(2)

{"error":"processPastLogs: Internal error","level":"error","msg":"Unable to process past deposit contract logs, perhaps your execution client is not fully synced","prefix":"execution","time":"2024-10-10T01:42:26Z"}

I am currently confirming if this also happens for the previous Besu release or if it's a new regression.

I suspect these should be warnings and not errors. At least (2) doesn't make sense as an error to me since it is expected to wait until the client is synced.

[siladu]

Tested on holesky and 24.9.1 and 24.10.0-RC1 both get these errors...

{"lastBlockNumber":0,"level":"error","msg":"Beacon node is not respecting the follow distance. EL client is syncing.","prefix":"execution","time":"2024-10-11T05:42:02Z"}

{"error":"received 0 payload bodies from engine_getPayloadBodiesByRangeV1 with count=1 (start=2509611): engine api payload body response is invalid","handler":"beacon_blocks_by_root","level":"error","msg":"Could not get reconstruct full block from blinded body","prefix":"sync","time":"2024-10-11T04:53:40Z"}

So I think it's a preexisting issue.
Not sure exactly why holesky gets a different error to mainnet though.