RE: x$kglna - memory corruption ?

From: yong huang <yong321_at_yahoo.com>
Date: Thu, 21 Dec 2000 09:44:34 -0800 (PST)
Message-Id: <10717.125141@fatcity.com>

Based on the workaround you told us about, it looks like a bad index on the underlying x$ table, even though there's no difference shown in v$indexed_fixed_table between my 8.1.5 and 8.1.6 databases for x$kglna.

Anyway, the following query is also a good workaround:

select * from x$kglna where to_number(kglnahsh) = 4049165760;

But this will not:

select /*+ FULL */ * from x$kglna where kglnahsh = 4049165760;

Interesting!

Yong Huang
yong321_at_yahoo.com

you wrote:

> This sounds like it's related to bug 904088 ("select from v$sqltext
> hangs") which manifests itself by causing queries of the form :
>
> select sql_text from v$sqltext where hash_value = n;
>
> to spin and eat CPU.
>
> What do you get if you run the statement as :
>
> select * from x$kglna WHERE KGLNAHSH+0 = nnnnnnnn;
>
> This bug is fixed in 8.1.6.
>
> Regards,
>
> Mike Hately

---

Do You Yahoo!?
Yahoo! Shopping - Thousands of Stores. Millions of Products. Received on Thu Dec 21 2000 - 11:44:34 CST