Thanks,
Daniel
If the files are large enough, you will benefit by doing a RMVM on each of the
logicals (not a DLTF), do the RGZPFM, then add back in the logicals using
ADDLFM (not CRTLF).
You never really delete the logicals, you just remove any and all members from
them. The indexes for the logicals will then be updated when the ADDLFM's are
executed (and you can execute more than one at a time), rather than the
logicals being updated as part of the RGZPFM.
Scott Lindstrom
Zenith Electronics
Make certain to display (print) the logical member attributes before
removing unless you can see that they're all simple and direct. And even
then it's not a bad idea to have a listing.
Tom Liotta
In article <20000919215620...@ng-fz1.aol.com>,
--
Tom Liotta, AS/400 Systems Programmer
The PowerTech Group, Inc.; http://www.400security.com
...and for you automated email spammers out there:
rhu...@fcc.gov jqu...@fcc.gov sn...@fcc.gov rch...@fcc.gov
Sent via Deja.com http://www.deja.com/
Before you buy.
Peter
=========================================
In article <8q9931$957$1...@nnrp1.deja.com>,
In order to run RGZPFM, you need an exclusive lock on the file.
I've just been given a free copy (not eval) of MiMiX/ Promotor... I think
it's for this month only - they're just giving it away whether you're a
MiMiX customer or not.
Works like this: (You run command MIMIX/RGZACTF.)
Without taking the users off your system (releasing locks on those files),
MP makes a COPY (temporary DASD implication) of the PF and associated
logicals (only if they're in the same Library as the PF).
He copies the data over to the new file, without the deleted records.
Using OS/400 journaling, he catches any changes your users make to your live
PF and applies them to the copy (in a ## library) - using MiMiX/Replicator
logic.
While this is happening, he's watching the locks on your live file - and
when he's happy he can copy back, he does a quick move of the file from the
## lib to your prod lib (not sure of the exact steps - all cloak and dagger
stuff)... Hence you drop your downtime from hours to minutes.
Hey presto - it all happens automatically and you don't need to worry about
controlling reorgs anymore.
--
Regards,
Bryan Douglas-Henry
br...@mediscor.co.za
Send your spam to: rhu...@fcc.gov; jqu...@fcc.gov; sn...@fcc.gov;
rch...@fcc.gov
"Daniel" <rome...@bigfoot.com> wrote in message
news:ooRx5.35260$XT1.5...@news5.giganews.com...
> We want to reorganize our files, recapture the deleted record space, but
I'm
> not sure which way is best to do this. We have files with 4, 5 and more
> logicals built on them and I would like some info from others that have
done
> this before.
>
> Thanks,
> Daniel
>
>
Why? When the reorg starts, doesn't the system invalidate all the
indexes and rebuild them when it finishes?
Richard Jackson
http://www.richardjacksonltd.com/
Mailto://richard...@richardjackson.net
Voice: 303 808 8058
Fax: 303 663 4325
When I reorganise a file with several logicals, I see status messages
'Building access path ...' flashing by. This would suggest that OS/400 is
smart enough to postpone the index building until after the physical has
been reorganised (remember that for using RGZPFM you'll need an exclusive
lock on the file, explicitly or implicitly).
Joep Beckeringh
"ScLind" <scl...@aol.com> wrote in message
news:20000919215620...@ng-fz1.aol.com...
But the reorg takes longer this way, so be careful...
--
Simon Brunning
You're bound to be unhappy if you optimize everything. - Donald E. Knuth
In article <8qa9qk$fsn$1...@ctb-nnrp2.saix.net>,
"Bryan Douglas-Henry" <br...@mediscor.co.za>
wrote:
> Free advertising for Lakeview.
(www.lakeviewtech.com)
>
> In order to run RGZPFM, you need an exclusive
> > not sure which way is best to do this. We
have files with 4, 5 and more
> > logicals built on them and I would like some
info from others that have
> done
> > this before.
> >
> > Thanks,
> > Daniel
Are there benchmark results somewhere that show this? If not, what is the
basis for your statement? Thanks!
--
Dave Shaw
MAPICS-L Moderator
Spartan International, Inc.
Spartanburg, SC
"Simon B." <bru...@my-deja.com> wrote in message
news:8qciju$4gg$1...@nnrp1.deja.com...
> In article <8q99fo$9mo$1...@nnrp1.deja.com>,
> In for a penny <hibidd...@yahoo.com> wrote:
> > And another thing to consider is maybe specifying the KEYFILE as the
> > most frequently used access path...
>
> But the reorg takes longer this way, so be careful...
>
> --
> Simon Brunning
> You're bound to be unhappy if you optimize everything. - Donald E. Knuth
>
>
I have no idea.
> If not, what is the basis for your statement? Thanks!
Bitter experience. ;)
--
Simon Brunning
We had better get coding straight away, because we are going to have
lots of debugging to do. - Steve McConnell, Code Complete
We're still on the older V3R2 version of mimix, and now we have to stop our
entire SAP R/3 environment in order to do a RGZPFM.
<sma...@my-deja.com> wrote in message news:8qd91j$uju$1...@nnrp1.deja.com...
You'll still need some downtime, but only enough for MiMiX to copy the temp
file back into production.
The actual 'reorg' of the +10Gb file may take a good many hours (days ?),
but you'll be able to use the production copy during that time.
--
Regards,
Bryan Douglas-Henry
br...@mediscor.co.za
Send your spam to: rhu...@fcc.gov; jqu...@fcc.gov; sn...@fcc.gov;
rch...@fcc.gov
"tom" <tom....@advalvas.be> wrote in message
news:8qo2mt$dsv$1...@news0.skynet.be...