Hi,

For case of statistics there is made a copy from a database some time
ago. I produced a SQL- Statement, that worked fine on the actual
database, but when I use the same statement on the other DB I got an
Oracle Error 600, followed by some values (the 1st is 733).
I do not know the differences between both DBs exactly, but of course
the admin changed some preferences for this one way DB. When I looked
for the reason of this error to find out, which changes are to be made
to get the results of my statement, I read, that there is/was a
metalink- Page, that describes some/all 600- errors depending on the 1st
parameter.
I could not connect to metalink (I did it last time many month ago),
because it did not only want to get to know very much of my personal
state, but any kind of service number. Assuming that there is no free
way to go there anymore I like to ask you the following:
- Is there someone, who knows about this web form? Does it exist
anymore, so that it is useful to buy such service number?
- Does anybody know the problem OE 600, [733]?

Best regards
Andreas

Oracle 9.2.0.7.0

--
wenn email, dann AndreasMosmann <bei> web <punkt> de

On Feb 22, 5:10*am, Andreas Mosmann <mosm... (AT) expires-28-02-2010 (DOT) news-
group.org> wrote:
I suggest you look at fixing your Oracle support access.

Oracle support note 146580.1 ORA-600 [733] "memory request size is too
big for this port (o/s)" lists about 10 bug reports related to this
error. The note suggests reducing the sort_area_size and
hash_area_size and re-running the query. It is not 100% clear in the
note but this solution may only apply if a hash_join is being used in
the plan.

This note may not apply to your issue.

HTH -- Mark D Powell --

On 22 Feb, 10:10, Andreas Mosmann <mosm... (AT) expires-28-02-2010 (DOT) news-
group.org> wrote:
Try patching to 9.2.0.8 and get back to us.

</Oracle support>

-g

Mark D Powell schrieb am 22.02.2010 in
<38e0b443-2f67-4a18-9a3f-3f1781354cb4 (AT) g11g2000yqe (DOT) googlegroups.com>:

nearly the same value as the original database and retried the query.
I harvested a similar error Ora 600, with Parameters [KGHALP1],
[0x0]...
When I looked for it at metalink I got a result like known bug, there
could be the following symptoms:
... Ora 600 [733], Ora 600 [KGHALP1] ...
Should be fixed in 9.2.0.6

But I use Oracle 9.2.0.7 AND on the original version it works fine.
Does anyone has an idea what to do except patching to 9.2.0.8?


Andreas


--
wenn email, dann AndreasMosmann <bei> web <punkt> de

gazzag schrieb am 23.02.2010 in
<8eeb5157-7099-4400-813c-8b5e67b210b0 (AT) s17g2000vbs (DOT) googlegroups.com>:

all the questions outside including questions for the wheather of the
next 10 years, but I can not do this (neither I nor the admin) because
this has to be kept unchanged (If I would believe that an upgrade would
solve this problem I probably could built an independent new database
with the same data, but I am afraid I don't. Metalink told, that such
bug should be handled in 9.2.0.5 so I suppose I have a similar but not
the same. And I don't believe that especially this bug is fixed in
9.2.0.8 if a solve is mentioned for .5). So I have to find a work around
for this.
Do you have the 2nd best answer for me?


--
wenn email, dann AndreasMosmann <bei> web <punkt> de

On Feb 25, 8:48*am, Andreas Mosmann <mosm... (AT) expires-28-02-2010 (DOT) news-
group.org> wrote:
You might consider opening a SR. It is possible the fix requries a
one off patch or was delivered via a CPU patch set and was not
automatically included in the base patch level.

HTH -- Mark D Powell --