github-actions[bot] commented on code in PR #68196:
URL: https://github.com/apache/doris/pull/68196#discussion_r4078759641
##########
fe/fe-core/src/main/java/org/apache/doris/datasource/hive/HiveMetadataOps.java:
##########
@@ -346,8 +346,10 @@ public void afterTruncateTable(String dbName, String
tblName, long updateTime) {
if (tbl.isPresent()) {
Env.getCurrentEnv().getRefreshManager()
.refreshTableInternal(db.get(), (ExternalTable)
tbl.get(), updateTime);
+ return;
}
}
+
Env.getCurrentEnv().getExtMetaCacheMgr().invalidateTable(catalog.getId(),
dbName, tblName);
Review Comment:
[P1] Canonicalize this cold-TRUNCATE fallback before routing it.
`TruncateTableInfo` stores the caller's DB/table spelling, while Hive cache
keys use canonical local names and invalidation compares those names exactly
(and hashes them for file entries). With `lower_case_database_names=2` or
`lower_case_table_names=2`, a follower whose DB is cached but table object was
evicted reaches this fallback with alternate casing: the row-count fallback
widens to the DB, but schema/partition/file invalidation misses the canonical
keys, so post-TRUNCATE reads can reuse stale file listings until expiry. Carry
the resolved local names (or invalidate the canonical DB scope), and add a
mixed-case cold-replay test with primed Hive caches.
--
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]
