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
Currently teku has 3 different pruners (Block, blob and state) all relying on timers that trigger their execution at a frequency defined by the used. To avoid running those non-critical task during critical periods (like epoch transition) we could either:
Implement a validation to avoid pruning during these periods. This will likely end up being called / used in 3 different places.
Implement a mechanism that is slot based and only prune when we're in a "safer window" and will likely not affect the other critical tasks.
Other suggestion are accepted. @tbenr dumped some ideas here and here
The text was updated successfully, but these errors were encountered:
Currently teku has 3 different pruners (Block, blob and state) all relying on timers that trigger their execution at a frequency defined by the used. To avoid running those non-critical task during critical periods (like epoch transition) we could either:
Other suggestion are accepted. @tbenr dumped some ideas here and here
The text was updated successfully, but these errors were encountered: