Re: Oracle 9i Latching Issue
Date: Fri, 28 Mar 2008 14:38:17 -0700 (PDT)
On Mar 28, 12:11 pm, sybra..._at_hccnet.nl wrote:
> On Fri, 28 Mar 2008 11:13:21 -0700 (PDT), hpuxrac
> <johnbhur..._at_sbcglobal.net> wrote:
> >On Mar 28, 10:24 am, "Ana C. Dent" <anaced..._at_hotmail.com> wrote:
> >> It appears you then must use the undocumented parameter in initSID.ora
> >> _make_database_faster_=TRUE-
> >Looks like Johne is well on the way to identifying problems and
> >potential alternatives. Nice job!
> >The contributions of Ana Dent in this thread ... well not so much
> At the same time you got on your periodical 'Holier than thou
> ' chair, you usually misuse to bash Daniel Morgan, at the same time
> not providing anything useful.
> Mr. Hurley, are you an authority of moral wisdom, or just a plain
> hypocrite? I would bet the latter: It begins to show quite a bit!!!
> Sybrand Bakker
> Senior Oracle DBA
I gotta say, I'm with hpuxrac on this one. Ana is possibly half-wrong (or fully half-unfunny), the OP and the people helping him have identified that it is not necessarily slow SQL, but quite possibly a classic hot block issue that could be worked around. I would agree fixing the SQL would be the _correct_ thing to do, if it is indeed as David said.
Johne, be aware you can start traces from other sessions, too. Sometimes being able to show exactly what is wrong can help the lead time of a fix.
Ana, if this is a known problem with Discoverer or you were joking, all apologies! Of course, the classic cause of the classic hot block problem is classically bad programming. Just because something is from Oracle doesn't mean it is immune.
-- @home.com is bogus. http://www.sfgate.com/cgi-bin/article.cgi?f=/c/a/2008/03/27/MNUAVQUK2.DTLReceived on Fri Mar 28 2008 - 16:38:17 CDT