Oracle FAQ Your Portal to the Oracle Knowledge Grid

Home -> Community -> Mailing Lists -> Oracle-L -> Re: Seeking advice on potential 10g upgrade

Re: Seeking advice on potential 10g upgrade

From: Charles Schultz <>
Date: Mon, 28 Aug 2006 09:04:22 -0500
Message-ID: <>

From my own experience, I would have to say 10g is not ready yet. I cannot speak for Oracle Text, but the other 3 reasons your client wants to upgrade do not warrant the effort. "Overall easier DBA" is a pure PR hoax, do NOT swallow
it! It would be interesting to upgrade to 10g and have your client administrate it just so he can see how "easy" it is. *grin*

Granted, 10g is the birth of some really cool concepts, and I look forward to future releases when these concepts mature. But, being new, they do not work 100%. We have turned off many of the "new" features, not to mention patch which turns off various other bells and whistles simply because they are broken.

Speaking from my own limitted experience, one thing that I have benefitted from is the AWR (especially the dba_hist* views). There may be other gadgets we use unknowingly, but the AWR we use actively. Also, as the sword cuts two ways, we have learned a lot from 10g (from trying to solve and reason out all the inexplicable, ineffable bugs) - if you have the time and patience, it would be an excellent learning opportunity. But if you, as a consultant, want something that is relatively low maintenance, stick with 9i (latest patch).

On 8/27/06, Janine Sisk <> wrote:
> Hi all,
> I used to be a regular reader of this list, but work took me in
> another direction and I've been away the last few months. However,
> life is circular and so here I am back again with another Oracle
> consulting job.
> I have kept all of my clients on 9i up to now, because the last time
> I was reading this list there was still considerable skepticism about
> the stability of 10g, especially on Linux which is what all our
> installations use. But this new client wants to use 10g, so I have
> many questions. I tried to search the archives, but I came up with
> way too many matches and many were several years old and so probably
> out of date.
> My questions... any insight would be much appreciated, as would links
> to resources that might help.
> - is 10g on Linux ready for production use yet? If not, where are
> the problem areas?
> - have people experienced any significant problems when upgrading
> from 9i to 10g?
> - is there an independent source to read about the actual
> improvements in 10g over 9i (ie, not just what Oracle says has been
> improved)?
> Basically, I have a situation where I'm telling the client "it's
> running well, let's leave it alone and put our efforts into other
> areas" and the client is chomping at the bit to upgrade to 10g
> because he thinks he is going to see lots of improvements. Areas
> where he says he expects to see improvement are:
> - Oracle Text
> - materialized views
> - automatically managed tablespaces
> - overall easier DBA
> I've looked at the "new feature guide" for 10g Release 2 and I see
> some of these things mentioned, but of course that doesn't mean the
> improvements are actually significant or that they outweigh the
> stability risks of upgrading.
> So, any and all suggestions, war stories, encouragement, etc
> welcomed! Pointers on good books for someone who is most familiar
> with 8i to get up to speed with 10g are needed also.
> Thanks!!!
> janine
> --

Charles Schultz

Received on Mon Aug 28 2006 - 09:04:22 CDT

Original text of this message