github-actions[bot] commented on code in PR #67161:
URL: https://github.com/apache/doris/pull/67161#discussion_r3870451033
##########
fe/fe-core/src/main/java/org/apache/doris/qe/ShortCircuitQueryContext.java:
##########
@@ -115,6 +116,7 @@ public ShortCircuitQueryContext(Planner planner, Queriable
analzyedQuery) throws
this.tbl = this.scanNode.getOlapTable();
this.tableName = this.scanNode.getTableNameInPlan();
this.schemaVersion = this.tbl.getBaseSchemaVersion();
+ this.partitionTopologyVersion = this.tbl.getPartitionTopologyVersion();
Review Comment:
**[P1] Bind the topology generation to the pruner snapshot**
This samples the generation whenever a `ShortCircuitQueryContext` is
constructed, but the first prepared execution constructs two contexts around
the same planner scan. A concrete interleaving is: the temporary context in
`StmtExecutor` executes and `lazyEvaluateRangeLocations()` builds the old range
map; that execution returns; another connection commits `ADD PARTITION`; then
`ExecuteCommand` constructs the persistent context from the same
planner/`OlapScanNode`, so this line stamps the stale, recently populated
pruner with the new generation. The next execution passes `isReusable()`,
`update()` observes its ten-second throttle, and the newly matching row is
still missed. No table lock spans execution completion and cache publication.
Please cache the exact context/generation used to build the pruner (or record
and compare the pruner's own topology generation) and add a latch-based test
for DDL between lazy pruning and context publication.
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: [email protected]
For queries about this service, please contact Infrastructure at:
[email protected]
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
