Hi all,
Upgrading DB schema from 2.0.0 to 2.1.0 is causing an error. Did anyone
experience similar issues?
Below we leave the command and stacktrace.
Thanks,
*./schematool -dbType mysql -upgradeSchemaFrom 2.0.0*
Starting upgrade metastore schema from version 2.0.0 to 2.1.0
Upgrade script upgrad
Hi Gopal,
I saw the log files and the hash table information in it. Thanks.
Also, I enforced shuffle hash join. I had a couple of questions around it:
1. In the query plan, it still says Map Join Operator (Would have expected
it to be named as Reduce side operator).
2. The edges in this query pl
Hi Dudu,
I tried the same on same table which has 6357592675 rows. See response of
all three.
*I tried 1st one , its giving duplicates for rows. *
> CREATE TEMPORARY TABLE INTER_ETL_T AS
> select *
> ,cast (floor(r*100) + 1 as bigint) + (100 * (row_number () over
> (partition by cast (
1.
This is strange.
The negative numbers are due to overflow of the ‘int’ type, but for that reason
exactly I’ve casted the expressions in my code to ‘bigint’.
I’ve tested this code before sending it to you and it worked fine, returning
results that are beyond the range of the ‘int’ type.
Please
I don’t have answers for you, except for #1 – mapreduce are the new classes in
Hadoop, from my understanding. They’ve been out for a while, but the Hive
storage handler API hasn’t been updated to make use of them. Which leads me to
my very related question: When might hive provide a storage ha
In case someone else encounters this issue, it looks like this was due to
encoding differences between the hourly and daily table. The hourly table
often had the same values stored consecutively for certain columns, but the
group by on multiple columns caused them to be shuffled around to different
