| T.R | Title | User | Personal
Name | Date | Lines |
|---|
| 634.1 | Yes | FILTON::BIRCH | It's not a bug, it's a creature | Wed May 02 1990 18:25 | 7 |
| Volume shadowing (if by that you mean the Digital product) is
transparent to applications; therefore you may shadow an Ingres
database, just as you may an Rdb database, RMS files, or any other
software package. VS worries about disk blocks, not
files/databases/other software structures.
PDB
|
| 634.2 | Thanks, and "official" words about Ingres relationship? | ODIXIE::RIDGWAY | Florida Native | Wed May 02 1990 23:40 | 14 |
| Thanks,
I wasn't sure how I wanted to ask the question.....I seem to remember
some type of competitive knock off about Ingres and Oracle not being
able to take advantage of VS.
Also, does anyone know the party line about what officially our
relationship is with Ingres? I am responding to an RFP that asks about
our relationship with Ingres. Any verbage that I can pass along in
response to the customer would be appreciated.
Thanks,
Keith R>
|
| 634.3 | My 2 cents | MAIL::DUNCANG | Gerry Duncan @KCO - DTN 452-3445 | Fri May 04 1990 00:17 | 29 |
| There is a minor problem with Oracle V6 and volume shadowing but this
only appears when we have a disk failure so I'm not sure you want to
bring this up. I don't know if it has been fixed or not.
Anyway, it goes like this.
If you have a shadow set which contains an Oracle database file AND
you MOUNT another disk into the set, VMS will initiate a shadow copy in
order to make sure that both (or all three) disks will eventually be
the same. At the time the shadow copy is started, there is a brief
period of time (2-3 seconds I think) were all members of the set are
shown as not available (or something like that).
The problem we experienced was that IF Oracle was attempting to write
to that shadow disk and received some sort of not ready/not available
message, the Oracle code didn't trap the return code and retry and
the writing process aborted.
I'm sorry I can't remember all the details but I do remember going
round 'n round with the cusomter and finally told them to buzz off and
have Oracle fix their code. As far as I know, the problem has not
happened again.
Anyway, from a logical point of view, Oracle and Ingres (and the rest)
cannot avoid volume shadowing if it is being used. Also, the term
"taking advantage" really doesn't have much meaning since all
readers of the DUS shadow disk will received the same advantage without
doing anything special.
-- gerry
|
| 634.4 | Gimme that Laser...gotta a Oracle to bust! | DPDMAI::DAVISGB | Escapee from New Hampshire... | Tue May 08 1990 17:40 | 13 |
| Except in the case of Oracle, which doesn't have code to handle this
situation on VMS....but of course, they're portable....
I spoke with a sales rep in California yesterday who told me about
another DEC sales rep who's wife is a sales rep for ...guess who? the
big O in California. I can imagine the hidden agenda when I go in and
try to convince him that Rdb is the best solution for his customer,
when his wife is "partnering" with DEC so sell Oracle...
C'est La Vie!
Gil
|