HP3000-L Archives

November 2001, Week 1

HP3000-L@RAVEN.UTC.EDU

Options: Use Monospaced Font
Show Text Part by Default
Show All Mail Headers

Message: [<< First] [< Prev] [Next >] [Last >>]
Topic: [<< First] [< Prev] [Next >] [Last >>]
Author: [<< First] [< Prev] [Next >] [Last >>]

Print Reply
Subject:
From:
"Atwood, Tim (DVM)" <[log in to unmask]>
Reply To:
Atwood, Tim (DVM)
Date:
Thu, 1 Nov 2001 11:57:19 -0800
Content-Type:
text/plain
Parts/Attachments:
text/plain (43 lines)
Thanks to everyone both on and off line.

It appears "jiggered DBPUT in XL" answer was correct. As far as I can
determine the behaviour is being caused by the Superdex (Bradmark) third
party indexing.

I am only 80% sure on this. Making it 100% sure would take more time than I
have right now. I would need to do such things as re-install the Superdex
modified DBPUT to the SL to be sure it really was not anyone else modifying
the SL routine. And I would then need to build a test database and test the
scenario that seems to be the case. Since this is only happening in CM mode
SL and does not really hurt much, I won't be able to justify the time.

The scenario I am 80% sure about is:
1. Superdex traps all DBPUT calls to determine if it needs to update the
"SI" Superdex index dataset. If Superdex is enabled it of course needs to
lock the SI dataset before updating the indices.
2. The "bug" (feature) in the Superdex DBPUT in the SL seems to occur when:
        a. The database has Superdex indices defined. Therefore creating an
SI dataset.
        b. Adager is used to add dataset(s). Using the "<next" syntax to add
the dataset prior to the "SI" dataset. The dataset number of the "SI"
dataset changes.
        c. SIMAINT is run to update Superdex definitions to new database
structure. This should fix everything to account for the SI dataset now
being a different dataset number. And everything appears to have been
updated as expected.
3. The CM (SL) Superdex appears to work fine. But it also appears to now
apply a set level lock to the dataset which is now in the numeric position
the SI dataset used to occupy.
4. The XL version does not exhibit this behaviour.

So I guess I won't need to start peach farming in the Okanogan. Though I
still think Carly Fiorina, the HP/Compaq merger, OpenVMS, and all the rest
are an alien plot to destroy MPE/iX because it is the only operating system
good enough to stop the alien invasion.




* To join/leave the list, search archives, change list settings, *
* etc., please visit http://raven.utc.edu/archives/hp3000-l.html *

ATOM RSS1 RSS2