RE: Disaster Recovery

["MITTELHEISSER Jean Claude" <MITTELJC@xxxxxxx>]

400 miles is the distance between production and backup AS/400. If primary 
AS/400 goes down, we must be able to restart as quick as possible on backup 
one. So when business starts at 8 on primary AS/400, it would be better we have 
all data we need to restart on backup side if necessary.

I know what it means when safes are not complete. And you're right : SAVSECDTA, 
SAVCFG, SAVDLO and SAV must be added to SAVCHGOBJ. We do it today.

As soon as we have the new system, tests will be done between the 2 machines : 
Restart on backup from safes, software tests and user tests, return to primary 
machine,... It must be organized at company level. Who does what, when and 
where...

Thanks.
Jean-Claude

-----Message d'origine-----
De : midrange-l-bounces@xxxxxxxxxxxx
[mailto:midrange-l-bounces@xxxxxxxxxxxx]De la part de Dan Bale
Envoyé : mercredi 15 septembre 2004 15:57
À : Midrange Systems Technical Discussion
Objet : RE: Disaster Recovery


I must have lost something in the translation.  What does 400 miles have to
do with anything?  Is that the distance between the two AS/400's?  So what
happens if your primary production AS/400 goes down?  Or, perhaps better
put, what do you want to accomplish in that scenario?

You probably already know this, but consider if you're missing anything by
not doing SAVDLO and SAV backups as well.

Disaster recovery is not a trivial task.  There's a lot of up-front planning
and analysis required so that you don't find yourself in a hole that you
can't climb out of when disaster strikes.  And testing.

db

> -----Original Message-----
> From: midrange-l-bounces@xxxxxxxxxxxx / MITTELHEISSER Jean Claude
> Sent: Wednesday, September 15, 2004 9:29 AM
>
> I agree with you if backup system is close to production one. My
> Save-21 takes about 45 minutes (50 GB to a Ultrium-1 tape unit),
> much shorter than 14 hours. But in my case backup system will be
> 400 miles away. So I have 13 hours to sent tape to backup. It's
> to short. But as we have no backup system today, we do every
> night SAVSECDTA, SAVCFG and SAVLIB *ALLUSR.
>
> Thanks.
> JCM
>
> -----Message d'origine-----
> De : midrange-l-bounces@xxxxxxxxxxxx
> [mailto:midrange-l-bounces@xxxxxxxxxxxx]De la part de Dan Bale
> Envoyé : mercredi 15 septembre 2004 14:49
> À : Midrange Systems Technical Discussion
> Objet : RE: Disaster Recovery
>
>
> If you've got 14-hour window every night and if a Save-21 doesn't require
> any operator intervention to change tapes, why not just do a Save-21 every
> night?  Or at least the SAVLIB *ALLUSR, SAVDLO, & SAV sets?  You'll have a
> FAR better recovery experience than if you have to rely on restoring up to
> several SAVCHGOBJ sets.  Oh, and if you use SAVCHGOBJ, you need
> to consider
> what happens when objects are deleted after the last full save.  In
> recovery, you'd restore the last full save, which restores the object you
> deleted, but restoring the SAVCHGOBJ sets won't tell the system about the
> fact that the object was deleted, will it?  Al?  Others?
>
> Again, I just think that you'd make it much more harder on yourself if you
> go with SAVCHGOBJ when you've got the time & resources to do full saves.
>
> Just my .02,
> db
>
> > -----Original Message-----
> > From: midrange-l-bounces@xxxxxxxxxxxx / MITTELHEISSER Jean Claude
> > Sent: Wednesday, September 15, 2004 8:20 AM
> >
> > We are currently thinking about solutions to recover one or more
> > AS/400. We will probably buy a new AS/400, but we don't want to
> > use mirroring or external software. Knowing that AS/400 has no
> > activity between 6 p.m. and 8 a.m., I'd thought using following method :
> > - A total system safe ("go save" 21) on tape once a week on
> > saturday morning. A copy of the tape will be sent to the backup
> > by postal mail.
> > - At end of each production day, save all modified objects
> > (SAVCHGOBJ) and send it to backup by using file transfer.
> > - 4 or 5 times a day, save the JRN files and send it to backup by
> > using file transfer too.
> >
> > Do you think that this solution will work ? I don't have too much
> > doubt about the first 2 safes, it's only a matter of volume of
> > data and of network speed. But is it possible to save JRN files
> > without stopping corresponding tasks and to restore them on
> > backup side ? Our OS is a V4R5 on a 820 and we will surely
> > migrate to a new 520 in V5R3 of course. 820 will become our
> > backup machine. Does V5R3 offer new possibilities in saving and
> > restoring ? Pehaps by making the things easier ?
> >
> > Thanks for your assistance.
> > Jean-Claude Mittelheisser
>
> --
> This is the Midrange Systems Technical Discussion (MIDRANGE-L)
> mailing list
> To post a message email: MIDRANGE-L@xxxxxxxxxxxx
> To subscribe, unsubscribe, or change list options,
> visit: http://lists.midrange.com/mailman/listinfo/midrange-l
> or email: MIDRANGE-L-request@xxxxxxxxxxxx
> Before posting, please take a moment to review the archives
> at http://archive.midrange.com/midrange-l.
>
>
> --
> This is the Midrange Systems Technical Discussion (MIDRANGE-L)
> mailing list
> To post a message email: MIDRANGE-L@xxxxxxxxxxxx
> To subscribe, unsubscribe, or change list options,
> visit: http://lists.midrange.com/mailman/listinfo/midrange-l
> or email: MIDRANGE-L-request@xxxxxxxxxxxx
> Before posting, please take a moment to review the archives
> at http://archive.midrange.com/midrange-l.
>

--
This is the Midrange Systems Technical Discussion (MIDRANGE-L) mailing list
To post a message email: MIDRANGE-L@xxxxxxxxxxxx
To subscribe, unsubscribe, or change list options,
visit: http://lists.midrange.com/mailman/listinfo/midrange-l
or email: MIDRANGE-L-request@xxxxxxxxxxxx
Before posting, please take a moment to review the archives
at http://archive.midrange.com/midrange-l.