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 the segfetcher loop we first resolve segments and check the NextQuery table to see what is cached. Then we request missing data. If multiple core request are done and revocations are involved it can happen that the NextQuery entry for an already requested core segment is deleted. So in the next resolve phase the request is again identified as to be fetched.
To prevent this we should make sure that for each segment type we only fetch once per path request.
The text was updated successfully, but these errors were encountered:
Instead of deleting next query entries when receiving revocation, be smarter when querying:
That means if a query returns no result after being filtered with revocations, try to refetch, otherwise return cached result to clients.
Fixes#3234
In the segfetcher loop we first resolve segments and check the NextQuery table to see what is cached. Then we request missing data. If multiple core request are done and revocations are involved it can happen that the NextQuery entry for an already requested core segment is deleted. So in the next resolve phase the request is again identified as to be fetched.
To prevent this we should make sure that for each segment type we only fetch once per path request.
The text was updated successfully, but these errors were encountered: