Oracle FAQ Your Portal to the Oracle Knowledge Grid
HOME | ASK QUESTION | ADD INFO | SEARCH | E-MAIL US
 

Home -> Community -> Usenet -> c.d.o.server -> Re: Undefined Materialized View

Re: Undefined Materialized View

From: Antoine BRUNEL <antoinebrunel/yahoo.fr>
Date: Wed, 7 May 2003 13:45:39 +0200
Message-ID: <3eb8f1e4$0$14841$79c14f64@nan-newsreader-02.noos.net>


This must be a kernel bug.

You should open an iTar on http://metalink.oracle.com to find the relevant bug (if it already exists).

Cdlt

"Donna" <dhoffman0917_at_yahoo.com> a écrit dans le message de news:79808a91.0305070338.322d69f6_at_posting.google.com...
> Version : Oracle 8.1.7.4
> OS: Solaris 8
> Architecture: Star Schema Data Mart with Materialized View with
> underlying partitioned fact table and aggregate table.
>
> As stated in several Oracle reference books, our Materialized View
> becomes "UNDEFINED" or "Unusable" each time DML is used against it for
> refreshing it. This is expected and is resolved by "ALTER
> MATERIALIZED VIEW <viewname> compile;" and works fine manually.
>
> It has begun to cause a problem when introduced in our automated load
> process and locks the underlying tables (the two facts) during the
> ALTER MV statement. The problem is that it hangs the tables during
> this lock transaction and never releases them in turn, hanging the
> database for any additional processes waiting to query the underlying
> fact tables. During this lock, the text in V$SQLTEXT displays the
> transaction command as "ALTER SUMMARY ....." as the syntax for the MV
> command (assuming it is referring to the ALTER MATERIALIZED VIEW
> <viewname> COMPILE command.
>
> Has anyone seen this lock scenario happen before? Do I need an
> additional COMMIT in my auto load process to release the transaction?
> Any ideas/suggestions would be greatly appreciated.
>
> Thanks
> Donna
Received on Wed May 07 2003 - 06:45:39 CDT

Original text of this message

HOME | ASK QUESTION | ADD INFO | SEARCH | E-MAIL US