Focal Point
Multiple session id's / userid on reportserver......

This topic can be found at:
https://forums.informationbuilders.com/eve/forums/a/tpc/f/7971057331/m/4461029122

April 05, 2007, 09:33 AM
Pete
Multiple session id's / userid on reportserver......
Hi,

We are testing a situation in which the reportcaster launches several reports , which we can see appear in the reportserver, under the same username , but when we launch multiple reports under the same userid, it seems that it causes exclusive row locks ?

Any clues?

Peter


D: WF 7.6.2 P. : WF 7.6.2 on W2003
------------------------------------------------------------------
I see myself as an intelligent, sensitive human, with the soul of a clown which forces me to blow it at the most important moments.

-Jim Morrison-

April 05, 2007, 10:27 AM
hammo1j
Pete

I'm guessing that you are getting row locks on the rcaster repository BOT* TABLEs and not the report application tables.

Is this right?

John



Server: WF 7.6.2 ( BID/Rcaster) Platform: W2003Server/IIS6/Tomcat/SQL Server repository Adapters: SQL Server 2000/Oracle 9.2
Desktop: Dev Studio 765/XP/Office 2003 Applications: IFS/Jobscope/Maximo
April 05, 2007, 12:30 PM
Darin Lee
Our experience with ReportCaster has been that we have had to limit the number of agent available to RC to only three or four. Otherwise we run into similar issues. Don't know exactly why, but with queueing available in RC, we didn't see the need to use up a bunch of available agents with RC all at once. We just let them queue up and run as agents become available.
This will vary among sites, depending on what is scheduled and such, but just wanted to share what we've found works best.


Regards,

Darin



In FOCUS since 1991
WF Server: 7.7.04 on Linux and Z/OS, ReportCaster, Self-Service, MRE, Java, Flex
Data: DB2/UDB, Adabas, SQL Server Output: HTML,PDF,EXL2K/07, PS, AHTML, Flex
WF Client: 77 on Linux w/Tomcat
April 06, 2007, 01:13 AM
Pete
Hammo,

I can only that there is exclusive row locking , but i dont know on which tables



P.


D: WF 7.6.2 P. : WF 7.6.2 on W2003
------------------------------------------------------------------
I see myself as an intelligent, sensitive human, with the soul of a clown which forces me to blow it at the most important moments.

-Jim Morrison-

April 06, 2007, 01:21 AM
Pete
So basically Darin, if you use too many agents at once it the reports lock eachother?


D: WF 7.6.2 P. : WF 7.6.2 on W2003
------------------------------------------------------------------
I see myself as an intelligent, sensitive human, with the soul of a clown which forces me to blow it at the most important moments.

-Jim Morrison-

April 06, 2007, 11:31 AM
Darin Lee
That was kind of a big generalization on my part. It's going to vary depending of which jobs are being run simultaneously, how often, data being accessed, etc. I don't know if there is record locking in the RC tables (BOT*), but it is entirely possible for large volumes of frequent RC jobs.
Since we have basically two sets of data (warehouse data in DB2 and finance data in ADABAS), most of the jobs that run are hitting the same data, and when several kicked off at once, we ran into resource issues and locking problem in the databases. So we've limited the agents available to RC to three, and just let the rest queue up and run as agents become available. The number you use will depend on quantity and frequency of scheduled jobs.


Regards,

Darin



In FOCUS since 1991
WF Server: 7.7.04 on Linux and Z/OS, ReportCaster, Self-Service, MRE, Java, Flex
Data: DB2/UDB, Adabas, SQL Server Output: HTML,PDF,EXL2K/07, PS, AHTML, Flex
WF Client: 77 on Linux w/Tomcat
April 07, 2007, 11:05 AM
Diptesh Patel
Pete: Your DBAs should be able to check and see which database tables have exclusive row-locking. We have 10 agents set p for Report Caster, and schedule about 1000 jobs. The rest go on the queue. We do not encounter any row-locking on the RC BOT* tables. For the most part, the BOT* tables are read (not updatd), except for a couple of them, tha get updated after the job finishes.
Does your scheduled Focexec do any updating of the database? What is the DBMS? Have the DBA check the database to verify that locking is at the lowest possible level. (row versus page versus tablespace). That might be the problem.


Diptesh
WF 7.1.7 - AIX, MVS
April 10, 2007, 01:33 AM
Pete
Disptech,

We try to manage both the dba en webfocus ourselves.

However , on friday we found out that - we work on Oracle 10g - there were some bitmap indexes defined on a very large table , and it turns out that those indexes caused the problems , because once we changed them into normal indexes , the reports were fast as ****.

Thanks for all the input , guys


D: WF 7.6.2 P. : WF 7.6.2 on W2003
------------------------------------------------------------------
I see myself as an intelligent, sensitive human, with the soul of a clown which forces me to blow it at the most important moments.

-Jim Morrison-

April 10, 2007, 04:27 AM
hammo1j
Spent about 2 months trying to tell Oracle their bit map indexes didn't work.

Honestly folks dont spend 2 million on Oracle dw - there's so much of it like bitmaps and partitioning that doesn't work properly...



Server: WF 7.6.2 ( BID/Rcaster) Platform: W2003Server/IIS6/Tomcat/SQL Server repository Adapters: SQL Server 2000/Oracle 9.2
Desktop: Dev Studio 765/XP/Office 2003 Applications: IFS/Jobscope/Maximo