Bug fixed: statement level triggers

Guys,

Just to let you know, we applied patch for bug1185276 and it fixed the problem on our 8.1.6.1.1 db. This was a generic bug that occurred on all platforms. Thanks to those who tried it on NT, Linux, Netware etc just to confirm.

BTW, the bug also exists on 8.1.5.1 contrary to what's documented in the patch!

Rgs,

> Leng.
> -----------------
> Leng Kaing
> Tel: +61-3-9843-8440
> Email: leng.kaing_at_hancorp.com.au
>
> -----Original Message-----
> From: Vadim Gorbunov [SMTP:dim_at_protek.ru]
> Sent: Friday, 9 June 2000 17:25
> To: Leng Kaing
> Subject: Re: Bug Confirmed: statement level triggers
>
> Leng, thanx for the info. Setting up an alert on 8.1.6.2 with Oracle
> support.
> Vadim
>
>
> > Guys,
> >
> > After a week with Oracle support, they have confirmed that I'm hitting a
> > known bug.
> >
> > Here's a snippet of reply from my tar:
> > - ct is most likely hitting <Bug:936193> marked as fixed in 8.1.5.1 and
> > 8.1.6.2& 8.1.7 --> only 8.1.5.1 is available at
> > the moment - 8.1.6.2 due out shortly.
> >
> > So we're stuck. Don't want to downgrade. Gotta wait for the upgrade.
> Argh!
> >
> > Leng.
> >
> > ----------------------
> >
> > From: Vadim Gorbunov <dim_at_protek.ru>
> > Date: Wed, 07 Jun 2000 13:22:50 +0400
> > Subject: Re: 8.1.6 Bug - statement level triggers
> >
> > Hello, Leng
> >
> > Have the same problem. Both on UnixWare and NT. Looks like generic bug.
> > Master/detail and delete triggers make detonating mix. One way triggers
> do
> > not
> > fire (see metalink), another crash. :-(. I appreciate very much if you
> let
> > me
> > know when you find solution
> >
> > Vadim Gorbunov
> > Oracle DBA
> >
> > Leng Kaing wrote:
> >
> > > Hello everyone,
> > >
> > > If you're on Oracle 8.1.6 (any o/s will do) can you please try this
> and
> > let
> > > me know if it's causing the same error as I'm encountering.
> > >
> > > <<trig.sql>>
> > > Essentially the scenario is this:
> > > 1. table SB1 is the parent of SB2
> > > 2. There are statement level triggers on both tables
> > > 3. If the statement level trigger on both tables are enabled, and I
> issue
> > > the statement DELETE * FROM SB1;, the client get ORA-3113. A trace
> file is
> > > generated in the udump area with the following dump:
> > >
> > > Exception signal: 11 (SIGSEGV), code: 1 (Address not mapped to
> object),
> > > addr: 0x90, PC: kglpnp()+92
> > > *** 2000-06-06 15:27:11.280
> > > ksedmp: internal or fatal error
> > > ORA-07445: exception encountered: core dump [kglpnp()+92] [SIGSEGV]
> > [Address
> > > not mapped to object] [144] [] []
> > > Current SQL statement for this session:
> > >
> > > If the statement level triggers on SB2 are disable, the delete
> statement
> > is
> > > processed without any errors.
> > >
> > > I've tried the same scenario on 8.0.5 and it is ok. But the errors are
> > there
> > > on 8.1.6.1.1.0 and 8.1.6.0.0.
> > >
> > > Thanks,
> > >
> > > > Leng.
> > > > -----------------
> > > > Leng Kaing
> > > > Tel: +61-3-9843-8440
> > > > Email: leng.kaing_at_hancorp.com.au
> > > >
> > >
> >
> > > Leng.
> > > -----------------
Received on Wed Jun 14 2000 - 19:35:19 CDT