Re: O/S Choice for Database Servers

> - or when they need to reboot the DB server because it's been up more
> than 90 days straight... well, that's when you know the platform you've
> chosen is probably not the wisest choice.
>
> nope that's when you know that the admin doesn't understand the platform. I
> must reboot every 90 days is an admission that something that I don't
> understand is happening.

While the person(s) you reference apparently did not make references for their arguments, I will.

If the server happens to be a 32-bit flavor of Windows Server because that's all the vendor supports, the dreaded "feature" of low System Page Table Entries can certainly be a factor -- it definitely for me. See:

http://blogs.technet.com/b/cotw/archive/2008/04/07/symptoms-lack-of-free-system-page-table-entries-ptes-system-wide-delays-i-o-request-failures-and-low-on-paged-pool-memory-and-or-non-paged-pool-memory-on-32-bit-windows.aspx

The "solutions" given in the article are as vague and poorly thought out as the posts you allude to. The title of this article might as well be "Let's Play Games With Our Production Server and Pray That Something Works".

I file the article under the theory given to us by Cary Millsap from his blog at
http://carymillsap.blogspot.com/2011/01/axiomatic-approach-to-algebra-and-other.html

"Theory: Redoing an n-step math problem instead of learning how to propagate a correction to an error made in step n – k through step n is how we get to a society in which our support analysts know only two solutions to any problem: (a) reboot, and (b) reinstall."

...which I happen to have hanging on my cube wall. Unfortunately, given my situation, my only cost-effective solution is (a). I hate (a). It's 20-freaking-11 for crying outside.

No, I don't know Win administration. I know a little of Unix/Linux. I think your rant is certainly valid, given the apparent poor evidence of the poster(s) to back up the suppositions. But there is also truth to *some* of their posts.

If someone could just point me to where I can turn off file caching of Oracle datafiles -- with proof -- that'd be a great start.

I like this ranting, Niall! :) Although I'll stop here, since dealing with SQL Server this week, which apparently does not enforce dependencies between objects, making it impossible to enforce. Fun! (mini rant) ;)

Cheers!

Rich

--
http://www.freelists.org/webpage/oracle-l