| Oracle FAQ | Your Portal to the Oracle Knowledge Grid | |
 |
 | |||
| Oracle FAQ | Your Portal to the Oracle Knowledge Grid | |
|
| |||
Home -> Community -> Usenet -> c.d.o.server -> Re: OracleServiceSID - shutdown question
OK, let's add one more point to this to clarify the discussion.
Oracle on NT in the earlier releases (prior to 8.0? Someone please correct me if this only came in 8i) had no option than to shutdown abort. The reason for this is the shutdown service capability at the OS level only gave you something like 20 seconds to do a clean shutdown, and we simply couldn't guarantee that in 20 seconds we could cleanly shut down Oracle. By "cleanly" I mean kick off all the users, roll back their uncommitted work, release the locks etc. etc.
In 8.0 (or 8i if my memory is failing me!), there were some registry entries added to NT that allowed you to specify (and again I'm going from memory here so I could be wrong on the specifics) both the type of shutdown and the amount of time to take before proceeding on the assumption it had finished.]
HTH. Pete
miriam wrote:
> Hi Sybrand,
>
> Thanks again...
>
> Sybrand Bakker wrote:
>
> > 1 A proper shutdown abort would reflect in the alert log file and this
> > shutdown doesn't. The only thing you see is the crash recovery. This brings
> > me to the assumption that stopping a service simply kills all threads
> > associated with them.
> Hmmm, that doesn't sound very funny to me... I'll try to find this one
> out for sure.
>
> > 2 Oracle decided to implement their software as an NT-service and NT
> > services work this way. Call them both accomplishes in this crime. Or better
> > still, while Oracle (at least this is known fact about Larry) definitely
> > dislikes NT, they simply can't afford not to make their software available
> > on NT.
> The last thing I know. I can't afford to say that I don't want to work
> with MS as well.. ;-)
>
> > 3 No, as far as I know there isn't. I didn't do any research because I have
> > always been administering Oracle on NT servers, and that server simply was
> > never shut down. Of course NT stops the services, which in turn.. etc...
> > I always considered the effect of this 'crash' as marginal.
> Well, as a mentioned before, for me now it isn't a big deal (I didn't
> even notice it until I looked in some logfiles). But it can be dangerous
> if there are other users working on the database.
>
> > 4 A while ago I came across someone making fun of the shutdown procedure of
> > Windows by means of the startup button. But frankly, the only really stable
> > operating systems are Unix and especially VMS. We all know what the fate of
> > VMS is going to be. Sic transit gloria mundi!
> Ok, ok. Don't ask me things about Unix. I would like to know about it,
> but I did not jet end up with it yet. But then again, I'm just a
> starter...
> And I don't want to start a discussion like the one about MS-SQL server
> vs. Oracle (that was the point in the first please, wasn't it?).
> (And don't throw latin on me please, I always regretted it not choosing
> latin on my VWO ('highschool'))
>
> B.t.w. you don't accidently work with Microstation (from Bentley
> systems) yourself, do you?
>
> Groetjes,
>
> Miriam
--
Regards
Pete
Pete Sharman Email: psharman_at_us.oracle.com Project Manager Phone: +1.650.607.0109 (int'l) Worldwide Internal Services Education (650)607 0109 (local)Oracle Corporation
SQL> select standard_disclaimer, witty_remark 2 from company_requirements;
Opinions are mine and do not necessarily reflect those of Oracle Corporation
"Controlling application developers is like herding cats."
Kevin Loney, ORACLE DBA Handbook
"Oh no it's not! It's much harder than that!"
Bruce Pihlamae, long term ORACLE DBA
Received on Fri May 21 1999 - 02:44:55 CDT
 |
 |