BOB: Business Objects Board
Not endorsed by or affiliated with SAP

Register | Login 

Want to sponsor BOB? 
Want to sponsor BOB? (Opens a new window)  

General Notice: No events within the next 45 days.

Experience with BOBJ on Hadoop
1 members found this topic helpful
Goto page Previous  1, 2
 
Search this topic... | Search General Discussion... | Search Box
Register or Login to Post    Forum Index -> General Discussion  Previous TopicPrint TopicNext Topic
Author Message
Nick Daniels
Forum Aficionado
Forum Aficionado



Joined: 15 Aug 2002

Posts: 13816
Location: England


flag
PostPosted: Wed Jan 25, 2017 2:16 pm 
Post subject: Re: Experience with BOBJ on Hadoop

Andreas wrote:
Would you not use rather IMPALA or VORA to access Hadoop I wonder?..
We use Impala. Fast but flakey. BO doesn't make it easy to connect.
_________________
My latest idea....
Back to top
Andreas
Forum Advocate
Forum Advocate



Joined: 20 Jun 2002

medal_silver.gif*2medal_gold.gif
Posts: 17326
Location: *** BEEP ...Dreaming of Africa... leave No message ; ) BEEP ***


flag
PostPosted: Wed Jan 25, 2017 2:18 pm 
Post subject: Re: Experience with BOBJ on Hadoop

And please which SAP BI BusObject version, Nick?..
_________________
Follow me on Twitter
Reading "The Design Of Everyday Things" by Don Norman
Focusing on Data Visualization, Design Thinking, SAP DesignStudio + scripting, SAP BI 4.x platform & architecture, SAP connectivity, Data Modeling, and SAP HANA Certified Associate
Back to top
Nick Daniels
Forum Aficionado
Forum Aficionado



Joined: 15 Aug 2002

Posts: 13816
Location: England


flag
PostPosted: Wed Jan 25, 2017 2:21 pm 
Post subject: Re: Experience with BOBJ on Hadoop

4.1 SP8 - it is a supported solution but .sbo files.....don't get me started.
We did it on 4.1 SP6 for a while which was not supported as we're on Impala 2.x and we managed to get that working but oh boy was it hard.

[edit]Oh and Lumira 1.32 - I've pointed that to it as well...again flakey[/edit]

We use SSL - maybe it would be more robust connecting to non-SSL.

_________________
My latest idea....
Back to top
nwdb
Principal Member
Principal Member



Joined: 26 Oct 2005

Posts: 112
Location: Birmingham


flag
PostPosted: Thu Jan 26, 2017 10:41 am 
Post subject: Re: Experience with BOBJ on Hadoop

We have Impala/Hadoop working on BO 4.1 SP3 and flakey is the kindest way to put it.

We had to edit a prm file and then it was fairly simple to connect, but performance is unreliable, and there are some characters that cannot be read using either the JDBC on the Linux server or the ODBC on our PCs

I would not recommend trying on an unsupported version

Also using the Java Web Client results in terminally corrupted reports especially when you have a mix of parquet and mosaic

David

_________________
Now working on 4.2 SP03 WebI HTML getting better, but interface needs improvement
Back to top
Nick Daniels
Forum Aficionado
Forum Aficionado



Joined: 15 Aug 2002

Posts: 13816
Location: England


flag
PostPosted: Thu Jan 26, 2017 3:43 pm 
Post subject: Re: Experience with BOBJ on Hadoop

I don't just find it flakey through BO - I think it is a flakey product, great potential but....I have an ever growing list of tricks and workarounds to just keep moving. But the upside is reducing queries that were Oracle originally to running in a tenth of the time in Impala. I have seen problems with the Java client but also experience them with the Rich Client tool as well. I find creating a universe against in a fairly bad experience - the schema browser is awful in IDT on Impala.
_________________
My latest idea....
Back to top
malteser
Principal Member
Principal Member



Joined: 21 Feb 2006

Posts: 116
Location: Manchester, UK



PostPosted: Thu Jul 13, 2017 11:12 am 
Post subject: Re: Experience with BOBJ on Hadoop

Just read through this thread with interest. We're on 4.1 SP6 and trying to connect to Hive on top of Hadoop using Hortonworks Hive ODBC driver.

We've got a universe working (IDT) okay apart from the fact that it doesn't recognise outer joins i.e. they're in the universe but when you view the resulting SQL it just comes up as an equi-join.

We logged a ticket with SAP and they said they didn't support the Hortonworks drivers and that we should use SAP OEM Hive Driver with source version 0.13 instead. However when we looked for that we could only find a 32 bit version but we need both 32 bit and 64 bit versions.

Any suggestions? Is this something that can be fixed by editing an .sbo file or .prn file instead?

All suggestions welcome icon_smile.gif
Back to top
charlie
Forum Associate
Forum Associate



Joined: 20 Aug 2002

Posts: 757
Location: Woodinville, WA USA


flag
PostPosted: Thu Dec 07, 2017 2:38 pm 
Post subject: Re: Experience with BOBJ on Hadoop

I've got an IDT universe (4.2 SP3) built on Cloudera using the Hive ODBC driver. Connection details say it's Apache Hadoop Hive 1.0. I can query an entire table ok, but as soon as I set a filter and run the query in WRC I get [Simba][HiveODBC] (35) Error from server: error code: '0' error message: ExecuteStatement finished with operation state: ERROR_STATE.

I tried limiting the query to a single column (the same as the filter) but got the same result. I'm a total newbie with hadoop so any help is appreciated.

_________________
Regards,

Chaz
Back to top
Display posts from previous:   
Register or Login to Post    Forum Index -> General Discussion  Previous TopicPrint TopicNext Topic
Page 2 of 2 All times are GMT - 5 Hours
Goto page Previous  1, 2
 
Jump to:  

Index | About | FAQ | RAG | Privacy | Search |  Register |  Login 

Get community updates via Twitter:

Not endorsed by or affiliated with SAP
Powered by phpBB © phpBB Group
Generated in 0.0151 seconds using 18 queries. (SQL 0.0036 Parse 0.0003 Other 0.0112)
CCBot/2.0 (http://commoncrawl.org/faq/)
Hosted by ForumTopics.com | Terms of Service
phpBB Customizations by the phpBBDoctor.com
Shameless plug for MomentsOfLight.com Moments of Light Logo