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
Even though every loop has called the update_safe_head method to put block info of 7134334 into the state.l2_refs, it will drop the span batch with the message 'prev L2 block not found'.
If Magi gets the previous L2 block from the op-geth, it will run correctly.
The text was updated successfully, but these errors were encountered:
The l2_refs only caches block info in the range of [safe_head - (max_seq_drift / block_time), safe_head].
In the optimism sepolia testnet, the max_seq_drift is 600, and block_time is 2.
When the safe_head is 7134334, the result is [7134034, 7134334]
Unexpectedly, the span batch is valid which timestamp is 1706070328 and refs to 7133894 but the l2 ref info of 7133894 has been removed from l2_refs when the safe head is 7134334.
Oh this is interesting. I was under the impression that after the sequencer drift you are supposed to start injecting empty blocks so this caching length would be acceptable. We'll need to figure out what the correct number to increase it to is.
@GrapeBaBa do you have any idea what it should be?
Oh this is interesting. I was under the impression that after the sequencer drift you are supposed to start injecting empty blocks so this caching length would be acceptable. We'll need to figure out what the correct number to increase it to is.
@GrapeBaBa do you have any idea what it should be?
No, @thinkAfCod current fixed it by qurey op-geth when can't find it in the cache.
Even though every loop has called the update_safe_head method to put block info of 7134334 into the state.l2_refs, it will drop the span batch with the message 'prev L2 block not found'.
If Magi gets the previous L2 block from the op-geth, it will run correctly.
The text was updated successfully, but these errors were encountered: