Discussion:
VSAM Release 2 (was Since it's Friday)
Add Reply
Mike Kerford-Byrnes
2017-12-05 11:42:27 UTC
Reply
Permalink
Raw Message
Going back a looong way, I worked on VSAM Release 1 (OS/VS1 Rel 2.6-ish) in
1974. It was not the happiest of times, with a significant number of errors
and problems, especially when VSAM limits were tested. Eventually peace
returned, and then VSAM Release 2 was announced.(I think late 1975 or early
1976). It supported Alternate Indexes, Improved CI Processing, GET-Previous
(Read Backwards) and introduced RRDS and spanned records.



Having endured a "busy" time dealing with Release 1, we were not best
pleased at having to demonstrate to our masters that Release 2 would not
visit yet further challenges to our production systems. The effort of
proving that was almost as great as that of the original problems, which
included a genuine Severity 1 (NO production at all) for a period of over 24
hours.



The LISTC output indicated which release had created the file and hence the
processing that was (or was not) permitted. VSAM Catalogs could describe
files created by either or both versions. Of course once a
REPRO/DELETE/DEFINE/REPRO or EXPORT/IMPORT was carried out on the new
system, it would be a 'Version 2' file and the check on how to process the
file was un-necessary.



I have just realised that was 43 years ago! It must have hurt - I can still
remember. The phone calls to STL - with the 8-hour time difference - and
the dump 'lost' by one of the couriers necessitating an IBM SE personally
transporting its replacement to California in his carry on bag!



MKB


----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to ***@listserv.ua.edu with the message: INFO IBM-MAIN
Steve Smith
2017-12-05 15:14:10 UTC
Reply
Permalink
Raw Message
I've occasionally wondered why LISTCAT has RELEASE-------------2 on
every entity listing (1974 is "before my time"). It's amazing how
release & version numbers deteriorate in various ways (Browser
Identity strings are the worst (maybe - hideous examples abound)).

I guess we can give up waiting for "Release 3".

sas
Post by Mike Kerford-Byrnes
Going back a looong way, I worked on VSAM Release 1 (OS/VS1 Rel 2.6-ish) in
1974. It was not the happiest of times, with a significant number of errors
and problems, especially when VSAM limits were tested. Eventually peace
returned, and then VSAM Release 2 was announced.(I think late 1975 or early
1976). It supported Alternate Indexes, Improved CI Processing, GET-Previous
(Read Backwards) and introduced RRDS and spanned records.
Having endured a "busy" time dealing with Release 1, we were not best
pleased at having to demonstrate to our masters that Release 2 would not
visit yet further challenges to our production systems. The effort of
proving that was almost as great as that of the original problems, which
included a genuine Severity 1 (NO production at all) for a period of over 24
hours.
The LISTC output indicated which release had created the file and hence the
processing that was (or was not) permitted. VSAM Catalogs could describe
files created by either or both versions. Of course once a
REPRO/DELETE/DEFINE/REPRO or EXPORT/IMPORT was carried out on the new
system, it would be a 'Version 2' file and the check on how to process the
file was un-necessary.
I have just realised that was 43 years ago! It must have hurt - I can still
remember. The phone calls to STL - with the 8-hour time difference - and
the dump 'lost' by one of the couriers necessitating an IBM SE personally
transporting its replacement to California in his carry on bag!
MKB
----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
--
sas

----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to ***@listserv.ua.edu with the message: INFO IBM-MAIN
Loading...