Re: Reuse vs. Reorg

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

On 2010-09-24 08:39, daparnin@xxxxxxxxxxxxxx wrote:

We have a file that is somewhat dynamic. Records are generated, they are
kept for awhile, some are deleted, others are generated, and so on. The
deleted records increases by about 100,000 per day. We run a reorg once a
week to get the deleted records back. I had thought about changing the
file to reuse deleted records to minimize the need to do reorgs. Is there
any downside to this? My boss thought that if the default for files was
*NO then there might be a reason that we wouldn't want to do it. This
isn't a file where we would want/need to reactivate the deleted records.
It would be easier to restore from a backup or just regenerate them.
Thoughts?


Dave Parnin

My recollection is that the reason is just historical. I don't believe
reusing deleted records was available when sliverlake became AS400 way
back when we had to crank the silly buzzards. I don't remember when the
support was added, but I'd guess it was around release 2. Anyway, as is
IBM's usual procedure, the previous behavior became the default, so as
not to break anything. As long as you don't have any processes that rely
on the physical sequence of the records (nobody who does THAT anymore,
right?) there's no reason to be afraid of it.

- --
Pete Hall
pete@xxxxxxxxxxxxxx
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.10 (GNU/Linux)

iEYEARECAAYFAkydQ8EACgkQXczQcKdXKg7cmQCfTs3OuEkhK1AtQj6jf3HbWi22
yCAAoIhlQ/4/jzA9DuBgOzH69XjyWX5x
=ijk2
-----END PGP SIGNATURE-----