This is an automated email from the ASF dual-hosted git repository.
stigahuang pushed a commit to branch master
in repository https://gitbox.apache.org/repos/asf/impala.git
The following commit(s) were added to refs/heads/master by this push:
new f092646f8 IMPALA-13993: waitForHmsEvent should check table events
under missing dbs
f092646f8 is described below
commit f092646f840a0ffe087051479a2948ef425e376d
Author: stiga-huang <[email protected]>
AuthorDate: Fri Apr 25 21:04:29 2025 +0800
IMPALA-13993: waitForHmsEvent should check table events under missing dbs
When a db is missing in the catalog cache, waitForHmsEvent request
currently just check if there are pending database events on it,
assuming processing the CREATE_DATABASE event will add the db with
the table list. However, that's wrong since processing CREATE_DATABASE
event just adds the db with an empty table list. We should still wait
for pending events on underlying tables.
Tests:
- Added e2e test which consistenly fails when running concurrently with
other tests in TestEventSyncWaiting without the fix.
Change-Id: I3fe74fcf0bf4dbac4a3584f6603279c0a2730b0c
Reviewed-on: http://gerrit.cloudera.org:8080/22817
Reviewed-by: Impala Public Jenkins <[email protected]>
Tested-by: Quanlong Huang <[email protected]>
---
.../impala/catalog/events/MetastoreEventsProcessor.java | 10 +++-------
tests/metadata/test_event_processing.py | 13 +++++++++++++
2 files changed, 16 insertions(+), 7 deletions(-)
diff --git
a/fe/src/main/java/org/apache/impala/catalog/events/MetastoreEventsProcessor.java
b/fe/src/main/java/org/apache/impala/catalog/events/MetastoreEventsProcessor.java
index b24b741e7..ba4223f54 100644
---
a/fe/src/main/java/org/apache/impala/catalog/events/MetastoreEventsProcessor.java
+++
b/fe/src/main/java/org/apache/impala/catalog/events/MetastoreEventsProcessor.java
@@ -1679,15 +1679,11 @@ public class MetastoreEventsProcessor implements
ExternalEventsProcessor {
} else if (catalogObject.isSetTable()) {
TTable table = catalogObject.getTable();
if (catalog_.getDb(table.db_name) == null) {
- // We will check existence of missing dbs. Once the missing db is
added,
- // the underlying tables are also added (as IncompleteTables). So
don't
- // need to check table events. I.e. there are no stale metadata on
those
- // tables.
+ // We will check existence of missing dbs.
dbNames.add(table.db_name);
- } else {
- db2Tables.computeIfAbsent(table.db_name, k -> new ArrayList<>())
- .add(table.tbl_name);
}
+ db2Tables.computeIfAbsent(table.db_name, k -> new ArrayList<>())
+ .add(table.tbl_name);
}
}
// Step 2: Check DB events
diff --git a/tests/metadata/test_event_processing.py
b/tests/metadata/test_event_processing.py
index f04a4ccfb..51bf2cf49 100644
--- a/tests/metadata/test_event_processing.py
+++ b/tests/metadata/test_event_processing.py
@@ -599,3 +599,16 @@ class TestEventSyncWaiting(ImpalaTestSuite):
"alter table {} partition(p=0)compact 'minor' and wait".format(tbl))
res = self.execute_query_expect_success(client, "show files in " + tbl)
assert len(res.data) == 1
+
+ def test_hms_event_sync_on_missing_db(self, vector, unique_name):
+ client = self.create_impala_client_from_vector(vector)
+ db = unique_name + "_db"
+ try:
+ self.run_stmt_in_hive("""create database {0};
+ create table {0}.tbl(i int);
+ create table {0}.tbl_2(i int);""".format(db))
+ self.execute_query_expect_success(
+ client, "insert into {0}.tbl values (0)".format(db))
+ finally:
+ self.run_stmt_in_hive(
+ "drop database if exists {0} cascade".format(db))
