Operational Defect Database
BugZero found this defect 48 days ago.
Data sources
All data on this page is proprietary to BugZero® or gathered from public sources
MongoDB | 2627697
Marker OpTime indicating optime list was truncated must be fast to retrieve
Last update date:
4/1/2024
Affected products:
MongoDB Server
Affected releases:
No affected releases provided.
Fixed releases:
No fixed releases provided.
Description:
Info
The downgrade code for replicateVectoredInsertsTransactionally currently uses a marker OpTime value of "OpTime::max()" to indicate the oplog chain for a transaction table has been truncated. OpTimes with a timestamp with either component (seconds or inc) with the high bit set result in a full collection scan when searched for, resulting in poor performance with a long oplog. We can just make this marker optime {{1,0),-1} instead; this will never exist in the oplog because we never write optimes with an inc value of 0.
Top User Comments
Steps to Reproduce
