spark-user mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From sim <>
Subject Re: Schema change on Spark Hive (Parquet file format) table not working
Date Sun, 09 Aug 2015 03:47:47 GMT
Yes, I've found a number of problems with metadata management in Spark SQL. 

One core issue is  SPARK-9764
<>  . Related issues are 
SPARK-9342 <>  ,  SPARK-9761
<>   and  SPARK-9762
<>  .

I've also observed a case where, after an exception in ALTER TABLE, Spark
SQL thought a table had 0 rows while, in fact, all the data was still there.
I was not able to reproduce this one reliably so I did not create a JIRA
issue for it.

Let's vote for these issues and get them resolved.

View this message in context:
Sent from the Apache Spark User List mailing list archive at

To unsubscribe, e-mail:
For additional commands, e-mail:

View raw message