backups on AS/400, objects -- MIDRANGE-L

Al may correct me if I'm wrong, but I believe the GO BACKUP menu is old and
kept there purely for the sake of backwards compatibility.  The new, preferred
menu to use for backup is GO SAVE.  Whether or not the backup options from the
BACKUP menu are as complete as the SAVE menu's options, I have no idea.  "SAVE
21" is the lazy man's way to indicate option 21 on the SAVE menu, which is
used to save the ENTIRE system.  It is my general understanding that this is
the backup that saves ABSOLUTELY everything on the system that can be saved.
If your drives melted one night right after a SAVE 21, you could use the SAVE
21 backup to get everything back up and running (well, you'd also need your
MULIC tape in that situation, but it's pretty much assumed that everyone knows
where theirs is, right? <g>)

>This was interesting, turns out the backup was really just using the command
> on the BACKUP menu (see my other post today..) running thru Job Scheduler

How did you determine this?  Does the option on the BACKUP menu put the job on
the Job Scheduler?

I know there's a way to get at the source code behind the SAVE 21 app, and
maybe the other options on the SAVE menu, but I forgot where it is (jeez, Al
just now told me that it is retrievable via RTVCLPGM QSYS/QMNSAVE; how's that
for service, eh?)  You may also want to check out the Backup & Recovery guide
for a section titled "What the Save Menu Options Do" (section 2.2.1 in the
V4R4 softcopy); it shows you the names of the IBM-supplied programs that you
can retrieve the CL source for, as well as the major commands used for each
option.  Al reminded me that the source for these is not "CL 101" class
material.

I apologize if I have misunderstood your writings (they are a little too
rambling for my short attention span <g>), but I would strongly emphasize
that, even if you are God's gift to S/36 OCL, AS/400 CL programming is
different enough that you can get in trouble if you don't know what you're
doing, especially with an extremely important function such as backup.  You
were a guru on saving and recovering on a S/36?  Congratulations, but you
wouldn't touch my backup & recovery strategy with a ten-foot pole in my shop
with the level of AS/400 experience which you professed to earlier.  I
appreciate that you've got an "accomplished" programmer to help you gauge your
backup, but quite frankly, programmers who are not *responsible* for systems
management (you didn't say whether he was) do not normally have the requisite
knowledge for dependable AS/400 backup and recovery.

All in all, I would suggest you (and everybody with that responsibilty) to
commit to memory the Backup and Recovery guide.  Well, of course no one has,
but you get the idea.  Ignorance on the subject of backup can kill you when it
comes time to use it and you haven't done it properly.  On *any* platform.

- Dan
Dan Bale says "BAN DALE!"
IT - AS/400
Handleman Company
248-362-4400  Ext. 4952
D.Bale@Handleman.com
  Quiquid latine dictum sit altum viditur.
  (Whatever is said in Latin seems profound.)

-------------------------- Original Message --------------------------
Hi Midrange,

Dan Bale
>Second, if the CL program is observable (also shown on DSPPGM), you can run
>RTVCLPGM on the program object to produce the source code  <snip>

This was interesting, turns out the backup was really just using the command
  on the BACKUP menu (see my other post today..) running thru Job Scheduler

>Opinion:  I'm getting the idea from your posts that you're not a programmer
>experienced on the AS/400.  Quite frankly, if you don't know what you're
 doing
>in CL, you may be setting your company up for a disaster fiddling with your
>backup.

Ahh... let's assuage your concerns here a bit..
I had originally fine-tuned the backup procedures for this company in the
90's on the
SSP S36...

(They were sold a Y2K bill of goods, spent a ton of money on completely
   unnecessary work, and they called me back in a year + later to
   get things working again...)

Now, granted, I let the backup setup of the previous company stand on the
AS/400, and didn't check much (just an occasional QS36F to another library
name restore) until this situation arose where objects did NOT restore and
they were not on the tape.. (and I am assured that there were no user jobs
running at the time)

BTW, for some operations issues on OS/400 through all this I was counting
on another
programmer who backs me up and comes in occasionally and is
fairly accomplished, and he gave no holler previously about the save..
.. at this point I am combining his expertise with this forum :-)

(My only other CL needs have been some Menu type stuff essentially, all
  the application code is still in good ol OCL and RPG 3)

So if you see my other post today, I think all the basic issues are hashed
out.. (SAVLIB being the better command, saving configuration and security
too.. saving while libs active AND the error log being sent to a simple
permanent file ... ) ... granted this should have been done months ago,
this non-saving of objects on the AS/400 with minimal notice did sorta
blindside me ..
.. oh I should also touch the issue of what is needed for an OS/400 and
utility and PTF restore in case of a full crash.. to close the loop... (like
your disaster recovery questions below..) my clients S36 had maybe
about 3 crashes thru the 90's so I am not unaware of these issues...

(I will have to figure out what to do if I still have user objects
sometimes not
  saving, which  hopefully will not be the case with SAVLIB and the right
parameters..
  ... I would probably trouble shoot why this occurs and/or add the S36 Save
  commands then.. )

My point is that this is not essentially a CL issue, this is more a "what
constitutes
a full and proper and verified save.." ..  there are many instances that
accomplished
programmers who know all the Op Sys commands let things go by on this..
... once 10+ years ago I changed the name of a physical file and didn't add
it the new name to the file by file save (didn't have a utility for saving
file by file
at the time) and ooppsss.. fortunately it was only a difficulty not a
disaster..

>  Can your company survive not having a complete backup if something
>happens to your system?  If you say yes, you must be also able to say that
 you
>have tested your backups to verify that they are complete.  Why not just do a
>SAVE 21 and be safe with that?  Having said all that, most AS/400 shops I've
>seen have never tested their backups in a DR situation, mine included.  Our
>company has been starting to get aggressive in this area and I expect that we
>will soon be asked to test our DR strategy.

Oh .. what is a SAVE 21 ?

The issues are basically the current theme of this post and the other one
 today
where I talk about the SAVLIB command and some parameter and logging
and initialization issues..

Granted, initially I mixed in some other OS/400 neophyte issues, so I
appreciate your concern..

Steven in newyawkl

+---
| This is the Midrange System Mailing List!
| To submit a new message, send your mail to MIDRANGE-L@midrange.com.
| To subscribe to this list send email to MIDRANGE-L-SUB@midrange.com.
| To unsubscribe from this list send email to MIDRANGE-L-UNSUB@midrange.com.
| Questions should be directed to the list owner/operator: david@midrange.com
+---