Re: application problems

From: Ben <benalvey_at_yahoo.com>
Date: Mon, 11 Feb 2008 10:13:02 -0800 (PST)
Message-ID: <bc80578c-6879-4d54-94a4-35650440924b@s12g2000prg.googlegroups.com>


On Feb 11, 12:25 pm, gazzag <gar..._at_jamms.org> wrote:
> On 11 Feb, 16:51, Ben <benal..._at_yahoo.com> wrote:
>
>
>
>
>
> > 10.2.0.2 Ent Ed AIX 5.2 64 bit
>
> > We have a vendor app that we recently upgraded and in turn moved it
> > from SQLServer to an Oracle database environment. All was well for
> > about a month then we started seeing random application failures. I've
> > looked at everything I know to look at in the database and don't see
> > any kind of indication that it is something within the database
> > causing the issues. In my opinion the app wasn't written well for an
> > Oracle db to begin with, they don't use any bind variables. It is a
> > fairly small (600M) oltp type of application but does have quite a few
> > transactions. The SGA is about 600M total and I haven't seen any
> > errors in the alert log to make me think that we are hitting any kind
> > of memory errors related to SQL cursors using up all the sga, or core
> > dumps, or anything for that matter to make me think there should be an
> > issue. The parse percentages are ridiculously bad, but even with that
> > I don't see extreme waits related to parse times.
>
> > Wouldn't I be seeing alert log errors if something in the sga was a
> > miss and causing client side problems?
>
> > Their client side logging is pretty lack luster as well, just cryptic
> > ole errors.
>
> In my experience, most database applications are pretty poorly
> written.  This is largely down to the software house's desire to
> create "database unaware" programs.  The rationale here being that it
> doesn't matter whether a potential customer wants to use Oracle, DB2,
> SQL Server or even mySQL, for that matter, as their backend, the
> applicaton itself won't need a rewrite; it'll behave just as poorly
> across the board.
>
> </rant>
>
> Anyhow, you say "random application failures".  I'm afraid that this
> doesn't give us too much to go on.  What do you mean when you say
> "failure"?
>
> HTH
>
> -g- Hide quoted text -
>
> - Show quoted text -

the application "runs" as a service on a windows machine with 10 or so threads. These threads seem to be getting killed or just dying off. The problem compounds and gets faster as more of them die. So it might take 10 hours for the first one to die, then 9 hours more for a 2nd one, then 6 hours more for the 3rd, then 2 hours for the 4th, 30 minutes more for the next, etc...

Once the last thread has died, all the users (40 or 50) lock up and the service ( and / or ) server must be restarted. Received on Mon Feb 11 2008 - 12:13:02 CST

Original text of this message