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
Right now, when using Paima, resetting your database also resets the START_BLOCKHEIGHT. This makes sense in two ways:
It avoids replaying any history from previous testing which might contain data you want to ignore
You want to avoid spending time waiting for historical data to sync
However, this feature was built at a time before localhost testing was a thing. It makes both of these points mostly useless because
You can just restart the chain entirely if you want to remove previous data
You're just syncing locally, so it's really fast anyway
That is to say, this feature is not only unnecessary in the average case, it's also dangerous because of timers. By resetting your START_BLOCKHEIGHT, it will not affect standard transactions (which still trigger during the presync), but it will not trigger time-based actions (which don't run during the presync, and probably can't run then)
Ways we can solve this
Remove the START_BLOCKHEIGHT reset (maybe make it optional behind a flag, since it is very situationally useful)
Add a warn/error if a time-based action gets added to Paima while presync is happening
The text was updated successfully, but these errors were encountered:
Thematically related to #414 and #413
Right now, when using Paima, resetting your database also resets the
START_BLOCKHEIGHT
. This makes sense in two ways:However, this feature was built at a time before localhost testing was a thing. It makes both of these points mostly useless because
That is to say, this feature is not only unnecessary in the average case, it's also dangerous because of timers. By resetting your
START_BLOCKHEIGHT
, it will not affect standard transactions (which still trigger during the presync), but it will not trigger time-based actions (which don't run during the presync, and probably can't run then)Ways we can solve this
START_BLOCKHEIGHT
reset (maybe make it optional behind a flag, since it is very situationally useful)The text was updated successfully, but these errors were encountered: