You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Unable to find SchemaNode for path 'foo.bar.ts' in the schema of file 's3a://goes-se-sandbox01/warehouse/tablespace/external/hive/foo.db/bar/data/00101-0-data-hive_20221201122048_e2f79a77-e78a-4dc6-b528-6e9bc2e728dd-job_16698906862860_0009-8-00001.parquet'.
Impala query plan always shows 1/1 partition scans from partitioned iceberg table
CREATETABLEfoo.bar (c1 INT) PARTITIONED BY SPEC (c1) STORED AS ICEBERG;
INSERT INTOfoo.barVALUES (1), (2), (3);
SELECT*FROMfoo.barWHERE c1 BETWEEN 1AND2;
observed behaviour
00:SCAN S3 [foo.bar, RANDOM]
S3 partitions=1/1 files=2 size=722B <-- 1/1 partitions even though table has 3 partitions, number of files and bytes scanned is correct
predicates: c1 <= CAST(2 AS INT), c1 >= CAST(1 AS INT)
stored statistics:
table: rows=3 size=1.06KB
expected behaviour: query plan shows 1/3 partitions scanned
Hive in-place migration with timestamp type missing bounds in manifest files