Re: journal record reduction issue using SQL

[Vernon Hamberg <vhamberg@xxxxxxxxxxxxxxx>]

I see there is more info at this URL -

https://www.ibm.com/developerworks/community/wikis/home?lang=en#!/wiki/IBM%20i%20Technology%20Updates/page/DISPLAY_JOURNAL%20%28easier%20searches%20of%20Audit%20Journal%29

This may be an update of the information in the other URL.

There are many "services" available now for getting information through 
SQL that is not really coming from tables.

Enjoy!

On 12/11/2013 12:38 PM, Vernon Hamberg wrote:
> How about taking a look at the DISPLAY_JOURNAL UDTF that is available
> since sometime in 6.1? Here's a URL to an example of its use -
>
> http://www.code400.com/forum/showthread.php/9974-Display_Journal-UDTF-%28V6R1%29
>
> On 12/11/2013 10:07 AM, Stone, Joel wrote:
> > After DSPJRN command to an outfile named "jrnflALL", I want to reduce ALL the journal records to only the oldest and newest pair (for each JOOBJ/RRN).
> >
> > So if Sally changed "Vanilla extract" item #123 from 2 oz to 201 oz to 20.01 oz to 2.01 oz throughout the weekend, I want to reduce that to the oldest BEFORE journal image and the newest AFTER image when extracting the journal Sunday evening - to identify NET changes to that record.
> >
> > I am using the following SQL and it works well.
> >
> >
> >         INSERT INTO JRNFLB4AF
> >                                                         
> >          select * from jrnflALL
> >          where joseqn in(
> >                     
> >          select min(joseqn) from jrnflALL
> >          group by joobj, joctrr
> >                         
> >          union
> >                            
> >          select max(joseqn) from jrnflALL
> >          group by joobj,joctrr)
> >
> >                     
> >          order by joobj,joctrr,joseqn
> >
> >
> >
> >          
> > So IT ops did some maintenance Saturday night and IPL'ed.  This reset the JOSEQN back to 1 for new journal writes.
> >
> > Now the SQL stmt returns misleading info.  The min JOSEQN is no longer the EARLIEST journal record.
> >
> > Questions:
> >
> >    - How does JOSEQN reset - does an IPL cause this?
> >
> >    - What is a good workaround?  Should I always add 1,000,000,000 to JOSEQN when processing?  Will this do the trick (forcing the post-IPL sequence numbers to the end)?  Or is that problematic?
> >
> >    - Should I add JODATE || JOTIME to the min and max stmt?
> >
> >
> > Any ideas would be appreciated.
> >
> > Thanks!
> >
> >
> > ______________________________________________________________________
> > This outbound email has been scanned for all viruses by the MessageLabs Skyscan service.
> > For more information please visit http://www.symanteccloud.com
> > ______________________________________________________________________