impala-reviews mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Csaba Ringhofer (Code Review)" <>
Subject [Impala-ASF-CR] IMPALA-8980: Remove functional*.alltypesinsert from EE tests
Date Wed, 01 Apr 2020 14:54:31 GMT
Csaba Ringhofer has posted comments on this change. (

Change subject: IMPALA-8980: Remove functional*.alltypesinsert from EE tests

Patch Set 3:

File testdata/workloads/functional-query/queries/QueryTest/insert_null.test:
PS3, Line 11: LOCATION 'hdfs://localhost:20500/test-warehouse/nullinsert';
The path should be created inside the uniquedb, so it should be /test-warehouse/{uniquqdbname}.db/nullinsert
. In the current setup running the same test with different parameters can interfere due to
using the same path.

Another issue is that the current path would only work on hdfs but not with s3/adsl

There are some macros to help with this situation:
would instantiate the correct values.
File tests/common/
PS3, Line 895:   def __drop_partitions(self, db_name, table_name):
This can be also removed.
PS3, Line 907:   def __reset_table(self, db_name, table_name):
This can be also removed.
File tests/query_test/
PS3, Line 161:   @pytest.mark.execute_serially
             :   @SkipIfNotHdfsMinicluster.tuned_for_minicluster
             :   def test_insert_mem_limit(self, vector):
             :     if (vector.get_value('table_format').file_format == 'parquet'):
             :       vector.get_value('exec_option')['COMPRESSION_CODEC'] = \
             :           vector.get_value('compression_codec')
             :     self.run_test_case('QueryTest/insert-mem-limit', vector,
             :         multiple_impalad=vector.get_value('exec_option')['sync_ddl'] == 1)
             :     # IMPALA-7023: These queries can linger and use up memory, causing subsequent
             :     # tests to hit memory limits. Wait for some time to allow the query to
             :     # be reclaimed.
             :     verifiers = [MetricVerifier(i.service)
             :                  for i in ImpalaCluster.get_e2e_test_cluster().impalads]
I think that unique_database is better here, but execute_serially is still needed due to the
comment below.

To view, visit
To unsubscribe, visit

Gerrit-Project: Impala-ASF
Gerrit-Branch: master
Gerrit-MessageType: comment
Gerrit-Change-Id: I257e936868917a2fcc6c030f6c855b247e8a0eea
Gerrit-Change-Number: 15529
Gerrit-PatchSet: 3
Gerrit-Owner: Adam Tamas <>
Gerrit-Reviewer: Adam Tamas <>
Gerrit-Reviewer: Csaba Ringhofer <>
Gerrit-Reviewer: Gabor Kaszab <>
Gerrit-Reviewer: Impala Public Jenkins <>
Gerrit-Comment-Date: Wed, 01 Apr 2020 14:54:31 +0000
Gerrit-HasComments: Yes

  • Unnamed multipart/alternative (inline, 8-Bit, 0 bytes)
View raw message