[Mondrian] Infobright Distinct Count

Tom Barber Tom.Barber at ecommera.co.uk
Wed Feb 9 11:18:14 EST 2011

Always helps if I actually list the error :)

[java] Caused by: java.sql.SQLException: Row 4 was truncated; it contained more data than there were input columns

From: mondrian-bounces at pentaho.org [mondrian-bounces at pentaho.org] On Behalf Of Tom Barber [Tom.Barber at ecommera.co.uk]
Sent: 09 February 2011 16:08
To: jhyde at pentaho.com; 'Mondrian developer mailing list'
Subject: RE: [Mondrian] Infobright Distinct Count

When was IB last tested? Cause the loader keeps freaking out at:

3100,3000,"Gross Sales","Income","+","LookUpCube("[Sales]","(Measures.[Store Sales],"+time.currentmember.UniqueName+","+ Store.currentmember.UniqueName+")")"

[java] java.lang.RuntimeException: Error while executing statement: LOAD DATA INFILE '/tmp/tmpfile2200554152399396504.csv' INTO TABLE account FIELDS TERMINATED BY ',' OPTIONALLY ENCLOSED BY '"'

We've had issues with the IB loader before, but I was wondering if this was in the test suite before or this is post last IB test run.


From: Julian Hyde [jhyde at pentaho.com]
Sent: 08 February 2011 18:13
To: Tom Barber; 'Mondrian developer mailing list'
Subject: RE: [Mondrian] Infobright Distinct Count

> Tom wrote:
> Interestingly, the mysql client says there is a warning. But
> "show warnings" returns null, and the result is of course
> correct so I think this may be a legacy thing, I am running
> 3.4.2 and I think 3.5 is now available as well.
> I shall run the mondrian test suite and let you know the results.

Thanks. Looking forward to those results. Attach a patch to the jira case
when you have one.

If I remember rightly, in the previous version there was an outright error,
not just a warning.

In previous versions of Infobright the concern was not correctness. If the
feature was not handled by the Infobright engine it would fall through to
MySQL and performance would be abyssmal.

It looks like every version of Infobright has significant improvements. Keep
a close eye on that log4j.category.mondrian.sql output and let me know if
you see anything else that is odd.


More information about the Mondrian mailing list