Oracle FAQ | Your Portal to the Oracle Knowledge Grid |
![]() |
![]() |
Home -> Community -> Usenet -> c.d.o.server -> Re: Oracle 9i DB Security Hole
I think that your judgement on this case may be a bit harsh. Given that it took about 24 hours for the patch to appear from the moment the post hit the newsgroup, it clearly wasn't a case of:
"It's too difficult / dangerous / expensive to fix, let's hope no-one else notices before 9.2"
I think this was just a case of a front-end Metalink analyst not realising the significance of what he was seeing, and the developer who picked it up fixing it being focused on code-lines, not on the real world and customers.
I'd bet on it being a junior employee failing to follow procedures or, as you put it
"the an institutional discipline that makes it possible to deal with problems like this in a timely, straightforward and plain-spoken manner"
I don't think this was anything to do with a conspiracy of silence.
-- Jonathan Lewis http://www.jlcomp.demon.co.uk Author of: Practical Oracle 8i: Building Efficient Databases Next Seminar - Australia - July/August http://www.jlcomp.demon.co.uk/seminar.html Host to The Co-Operative Oracle Users' FAQ http://www.jlcomp.demon.co.uk/faq/ind_faq.html m. fowler wrote in message ...Received on Thu Apr 18 2002 - 11:36:48 CDT
>The dialogue on this Usenet group is pathetic, to put it mildly - I
>quit reading the Oracle Usenet groups years ago but was directed back
>by the article on theinquirer.net( other than there and in a few
>discussion groups there has been not a peep about all this ). Anyone
>with half a brain, upon reading the original post on Mon., and with a
>few minutes of testing would have grasped the awful and awesome truth
>- and at that point would have shut down their 9.0.1 db. Oracle has
>released a patch using the same bug # that was logged back in Dec. (
>thus announcing the problem to the world ).
> You may know( on second thought you probably don't ) that 9.0.1
>installs with username: dbsnmp / password: dbsnmp - account unlocked -
>this un has connect role by def., which includes create view - thus
>the global R/W access.
> I'm rather sorry to see what happened here - I've suffered for
>years with Oracle, turning me into one of their many critics but also
>one of their biggest fans( especially considering the competition ).
>I perceive that there are many competing camps within any large corp.
>- they don't speak as one monolithic entity and what they do say is
>skewed from the top and amongst their willing water-boys, the media.
>So there needs to be an institutional discipline that makes it
>possible to deal with problems like this in a timely, straightforward
>and plain-spoken manner. Was that the case here - you be the judge.
![]() |
![]() |