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
In order to create data (checkpoint file + protocol snapshot file) for dynamic bootstrapped execution node, currently we have to make a copy of the entire data disk of an execution node, and run state extraction on it. And usually we use an old snapshot , which usually is about 1 day old, for this data preparation task. Since the data is about 1 day old, it will take time for the dynamic bootstrapped execution node to catch up. And running state extraction would need dedicated machine resource and also takes time.
Idea
Since the execution node generates checkpoint file from time to time, we could make use the latest checkpoint file for dynamic bootstrapping. The only thing missing is to generate the corresponding protocol snapshot file for a given checkpoint file. If this could work, then creating data for dynamic bootstrapping would be much faster and cheaper.
Solution
Since a checkpoint could contain up to 500 trie root hashes, we need to pick one and find a block which produces a statecommitment matches the trie root hash, and then find a finalized block that contains the seal to that block.
Add a command to admin tool or a util command that takes a checkpoint file and generates a protocol snapshot file for it.
The text was updated successfully, but these errors were encountered:
Problem Definition
In order to create data (checkpoint file + protocol snapshot file) for dynamic bootstrapped execution node, currently we have to make a copy of the entire data disk of an execution node, and run state extraction on it. And usually we use an old snapshot , which usually is about 1 day old, for this data preparation task. Since the data is about 1 day old, it will take time for the dynamic bootstrapped execution node to catch up. And running state extraction would need dedicated machine resource and also takes time.
Idea
Since the execution node generates checkpoint file from time to time, we could make use the latest checkpoint file for dynamic bootstrapping. The only thing missing is to generate the corresponding protocol snapshot file for a given checkpoint file. If this could work, then creating data for dynamic bootstrapping would be much faster and cheaper.
Solution
The text was updated successfully, but these errors were encountered: