"{Darkavich} Steven Misrack" <[EMAIL PROTECTED]> wrote:
> I figured it out, and it was actually worse than this. I want to
> open a bug report on the problem.
>
> Here is what I did.
>
> primary host was X
> primary was run as:
> utreplica -p -a Y
>
> host Y was run as:
> utreplica -p -a X
> ^^^^^^^^^^^^ Big Mistake.
>
> This forced the LDAP sever on X to change it's DN in the ldap
> directory to be cn=utadmin,Y
>
> So after I fixed my mistake on Y and tried to re-run the utreplica
> on Y, it failed because the DN was no longer valid on the primary.
>
> The bug is that it really allowed me to do this. I tested this over
> and over in the lab and could corrupt a running primary every
> time by running "utreplica -p -a X". Every priamry instance is at
> risk of a - p typo from a secondary host.
>
> I would suggest that if a primary is getting a request to load the
> secondary config, that the command should throw an error or a
> warning message.
>
> Did that all make sense?
Makes sense to me. I agree that it shouldn't be that easy to trash
a running primary. I'd say you should at least be asked to confirm
that you really do want to convert the existing primary into a
secondary. I could even see just failing the command and requiring
you to manually go and unconfigure the primary before you're allowed
to convert it into a secondary.
OttoM.
__
ottomeister
Disclaimer: These are my opinions. I do not speak for my employer.
--
___________________________________________________
Play 100s of games for FREE! http://games.mail.com/
_______________________________________________
SunRay-Users mailing list
[email protected]
http://www.filibeto.org/mailman/listinfo/sunray-users