[
https://issues.apache.org/jira/browse/SPARK-32672?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17181868#comment-17181868
]
Robert Joseph Evans commented on SPARK-32672:
---------------------------------------------
So I am able to reduce the corruption down to just a single 10,000 row chunk,
and still get it to happen. I'll post a new parquet file soon that will
hopefully make debugging a little simpler.
{code}
scala> val bad_order =
spark.read.parquet("/home/roberte/src/rapids-plugin-4-spark/integration_tests/bad_order.snappy.parquet").selectExpr("b",
"monotonically_increasing_id() as id").where(col("id")>=70000 and col("id") <
80000)
bad_order: org.apache.spark.sql.Dataset[org.apache.spark.sql.Row] = [b:
boolean, id: bigint]
scala> bad_order.groupBy("b").count.show
+-----+-----+
| b|count|
+-----+-----+
| null| 619|
| true| 4701|
|false| 4680|
+-----+-----+
scala> bad_order.cache()
res2: bad_order.type = [b: boolean, id: bigint]
scala> bad_order.groupBy("b").count.show
+-----+-----+
| b|count|
+-----+-----+
| null| 618|
| true| 4701|
|false| 4681|
+-----+-----+
{code}
> Data corruption in some cached compressed boolean columns
> ---------------------------------------------------------
>
> Key: SPARK-32672
> URL: https://issues.apache.org/jira/browse/SPARK-32672
> Project: Spark
> Issue Type: Bug
> Components: SQL
> Affects Versions: 2.4.6, 3.0.0, 3.0.1, 3.1.0
> Reporter: Robert Joseph Evans
> Priority: Blocker
> Labels: correctness
> Attachments: bad_order.snappy.parquet
>
>
> I found that when sorting some boolean data into the cache that the results
> can change when the data is read back out.
> It needs to be a non-trivial amount of data, and it is highly dependent on
> the order of the data. If I disable compression in the cache the issue goes
> away. I was able to make this happen in 3.0.0. I am going to try and
> reproduce it in other versions too.
> I'll attach the parquet file with boolean data in an order that causes this
> to happen. As you can see after the data is cached a single null values
> switches over to be false.
> {code}
> scala> val bad_order = spark.read.parquet("./bad_order.snappy.parquet")
> bad_order: org.apache.spark.sql.DataFrame = [b: boolean]
>
> scala> bad_order.groupBy("b").count.show
> +-----+-----+
> | b|count|
> +-----+-----+
> | null| 7153|
> | true|54334|
> |false|54021|
> +-----+-----+
> scala> bad_order.cache()
> res1: bad_order.type = [b: boolean]
> scala> bad_order.groupBy("b").count.show
> +-----+-----+
> | b|count|
> +-----+-----+
> | null| 7152|
> | true|54334|
> |false|54022|
> +-----+-----+
> scala>
> {code}
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
---------------------------------------------------------------------
To unsubscribe, e-mail: issues-unsubscr...@spark.apache.org
For additional commands, e-mail: issues-h...@spark.apache.org
