This is an automated email from the ASF dual-hosted git repository.
yiguolei pushed a commit to branch branch-4.1
in repository https://gitbox.apache.org/repos/asf/doris.git
The following commit(s) were added to refs/heads/branch-4.1 by this push:
new bb41e674edc [fix](case) test_partition_instance_query_cache: gate
plan-shape checks to single BE (#64520)
bb41e674edc is described below
commit bb41e674edcd20e711cce5b0b3b34f659e03fad3
Author: shuke <[email protected]>
AuthorDate: Tue Jun 16 19:23:42 2026 +0800
[fix](case) test_partition_instance_query_cache: gate plan-shape checks to
single BE (#64520)
## Problem
`test_partition_instance_query_cache` is flaky/failing on the branch-4.1
P0 regression. It asserts the query-cache **partition-parallelism** plan
shape: `scanInstanceCount == partitionCount` and that all of a
partition's tablets land in **one** scan instance. That shape only holds
when the scan runs on a **single BE**.
On a multi-BE cluster the optimization
(`UnassignedScanSingleOlapTableJob`) legitimately produces **one
instance per (partition, BE)** — or, when `totalTablets <=
parallelExecInstanceNum * numBE`, falls back to default per-tablet
parallelization. So on the multi-BE P0 cluster the case fails for the
wrong reason (verified: `partitions=3/3` but `scanInstanceCount=24`).
## Fix
Gate the single-BE-only plan-shape assertions on a single-BE topology
(`show backends`). The result-correctness and cache-consistency checks
above still run on **any** topology. Multi-BE behavior should be covered
by a dedicated test.
## Verification
Verified on a branch-4.1 cluster: on the multi-BE topology the gated
branch is skipped and the suite passes; the data/cache assertions still
execute.
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-authored-by: Claude Opus 4.8 (1M context) <[email protected]>
---
.../cache/test_partition_instance_query_cache.groovy | 14 ++++++++++++++
1 file changed, 14 insertions(+)
diff --git
a/regression-test/suites/query_p0/cache/test_partition_instance_query_cache.groovy
b/regression-test/suites/query_p0/cache/test_partition_instance_query_cache.groovy
index 6d2281e3ef6..83963777f51 100644
---
a/regression-test/suites/query_p0/cache/test_partition_instance_query_cache.groovy
+++
b/regression-test/suites/query_p0/cache/test_partition_instance_query_cache.groovy
@@ -93,6 +93,20 @@ suite("test_partition_instance_query_cache") {
contains("DIGEST")
}
+ // The "one scan instance per partition" plan shape asserted below only
holds when the
+ // scan runs on a single BE. On a multi-BE cluster the query-cache
partition parallelism
+ // legitimately produces one instance per (partition, BE) -- or, when
+ // totalTablets <= parallelExecInstanceNum * numBE, falls back to default
per-tablet
+ // parallelization (see UnassignedScanSingleOlapTableJob). The
result-correctness and
+ // cache-consistency checks above already ran on any topology; only the
single-BE-only
+ // plan-shape checks are gated here. Multi-BE behavior should have a
dedicated test.
+ def aliveBeNum = sql("show backends").count { it[9].toString() == "true" }
+ if (aliveBeNum != 1) {
+ logger.info("Skip partition-instance plan-shape assertions: requires a
single BE, found ${aliveBeNum}.")
+ sql "DROP TABLE IF EXISTS ${tableName}"
+ return
+ }
+
def distributedRows = sql("EXPLAIN DISTRIBUTED PLAN ${querySql}")
def distributedPlan = distributedRows.collect { it[0].toString()
}.join("\n")
assertTrue(distributedPlan.contains("UnassignedScanSingleOlapTableJob"))
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
