Re: What is the SYS password?

From: Daniel A. Morgan <damorgan_at_exesolutions.com>
Date: Thu, 25 Apr 2002 20:38:48 +0100
Message-ID: <3CC85B47.4013684B@exesolutions.com>

Pete Sharman wrote:

> In article <3CC89293.E2096823_at_exesolutions.com>, Daniel says...
> >
> >Pete Sharman wrote:
> >
> >> In article <3CC8716C.A4F46AD1_at_exesolutions.com>, Daniel says...
> >> >
> >> >Daniel Astillero wrote:
> >> >
> >>>> That's not true in all operating systems/Oracle versions. If you use Oracle
> >> >> for Windows NT, you can choose the SYS password during the instance
> >> >> creation, but if you use Oracle for Solaris you should change it after the
> >> >> instance creation.
> >> >>
> >> >> "NorwoodThree" <norwoodthree_at_my-deja.com> escribió en el mensaje
> >> >> news:ba03e2c.0204241303.ab11d4f_at_posting.google.com...
> >> >> > It is whatever you set it to be.
> >> >
> >> >And this is not correct as a generic statement.
> >> >
> >> >What is absolutely true is that Oracle always install with default passwords
> >> >and, not just on Solaris, you should change them.
> >> >
> >> >Daniel Morgan
> >> >
> >>
> >>Unfortunately this is not absolutely true. As of 9i, if you're using the DBCA
> >> to create a database, the last screen tells you that the database has been
> >> successfully created, that the passwords for SYS and SYSTEM are the defaults,
> >>and asks you to click on the Password Management button to go to a screen that
> >>allows you to change them. SCOTT is left with the default password of TIGER for
> >> regression testing (this is what the DBCA uses to test that the database has
> >> been created successfully - it attempts to log on as scott/tiger). All the
> >>other accounts are left locked and with non-default passwords. So if you unlock
> >>the sample schema accounts but don't change their passwords, you cannot log on
> >> e.g. you can't connect as OE/OE.
> >>
> >> HTH. Additions and corrections welcome.
> >>
> >> Pete
> >>
> >> SELECT standard_disclaimer, witty_remark FROM company_requirements;
> >
> >Are you saying that as of 9i the CREATE DATABASE command is been removed?
> >
> >Daniel Morgan
> >

>

> No, I'm not saying that at all. What I'm saying is that if you use the DBCA,
> your point is not valid. You stated as 100% fact that Oracle ALWAYS installs
> with default passwords. That is not correct.

>

> Having said that, let's make things a bit clearer. The CREATE DATABASE command
> is still there for you to use manually. In point of fact, all the DBCA does is
> put a nice interface around that command, so you don't need to remember all the
> syntax for the command. As more and more functionality is added to the CREATE
> DATABASE command (such as the UNDO TABLESPACE and DEFAULT TEMPORARY TABLESPACE
> functionality) it becomes more likely that people will use the DBCA rather than
> having to remember to add the new functionality in as they type an increasingly
> long CREATE DATABASE command. Yes there will still be people that use the
> CREATE DATABASE command manually. If you do so (and there are valid reasons for
> doing this such as scripted installs), then you have to remember to change the
> default passwords.

>

> To me it simply boils down to this - using the DBCA is more and more sensible as
> the way to create a database if you can. But don't make any leaps to things
> like commands disappearing simply because I point out different functionality.

>

> HTH. Additions and corrections welcome.

>

> Pete

>
> SELECT standard_disclaimer, witty_remark FROM company_requirements;

Either you misread what I wrote ... Or I wrote it very poorly. I was not trying to communicate that the passwords are always C_ON_I and MANAGER. I was trying to communicate that they should always be changed from the default. The original posting stated that they only needed to be changed with Solaris.

Let me quote the posting I was responding to: "but if you use Oracle for Solaris you should change it after the instance creation."

Implying that if you use HP/UX, AIX, NT, etc. etc. there was no need to. I think that is a very correct and reasonable thing to state.

Daniel Morgan Received on Thu Apr 25 2002 - 14:38:48 CDT