"Tim Young [Elevate Software]" <timyoung@elevatesoft.com> wrote in message
news:6E4D74F7-FC6D-4D57-B993-6DC7C652ED66@news.elevatesoft.com...
> Robert,
>
> << Look Tim, I did not "criticize" anything, simply stated that given what
> I observed (the fact that if I mapped the database as a drive it works,
> but if it a drive plus directories, or UNC, then it does not) your theory
> appeared to be incorrect. >>
>
> It's not a "theory", and I'm not trying to be "correct" about anything.
> I'm trying to go through the possible scenarios to explain what you're
> seeing, and you seem very displeased with the fact that I'm not telling
> you what you want to hear.  What do you want me to say ?  That it's a bug
> when there's no evidence that this is the case ?

There's plenty of evidence, I think, that something is not right in server
admin, at least with the version I am running here (4.25 Build 4). The fact
that what is apparently the same code (DirectoryExists and ForceDirectories)
return true in one program and false in another makes me think that the
input string is not exactly the same. Somehow we managed to corrupt it in
the server admin program, but I don't know how.

What do I want you to say? "The way you set up a directory on another
machine is as follows:" followed by something that works.

>
> <<  D6. >>
>
> The source code for both D5 and D6 are identical for both DirectoryExists
> and ForceDirectories.
>

So then why does the same directory cause an error in your program, and not
in my program? Have you tried the server admin using my scenario? (dbserver
and server admin on computer A, database on computer B). If it works for
you, then the problem is something else. If it does not work, then there is
something in server admin that is corrputing the directory name string and
causing the tests to fail.

Robert

PS: Don't take these things personally. It is just software.

Robert,

<< There's plenty of evidence, I think, that something is not right in
server admin, at least with the version I am running here (4.25 Build 4).
The fact that what is apparently the same code (DirectoryExists and
ForceDirectories) return true in one program and false in another makes me
think that the input string is not exactly the same. Somehow we managed to
corrupt it in the server admin program, but I don't know how. >>

Exactly.  So, you really *don't* have any evidence that there's a bug.  And
how exactly would one "corrupt" the input string, yet still have it work in
every other possible scenario ?  The only thing that is different here is
the use of a mapped drive for the database directory and nothing is changing
in the Server Administration Utility and/or Database Server code.  So how
exactly could it work for a local drive but fail for a mapped drive when
DBISAM is simply passing the exact string that you input on to the
DirectoryExists and ForceDirectories functions in both cases ?  The answer
is that it wouldn't work in either case if it were a bug in our code or
Delphi's code.  So it must be something environmental.

<< What do I want you to say? "The way you set up a directory on another
machine is as follows:" followed by something that works. >>

There's nothing magical here.  You specify the database directory in mapped
or UNC form, and that's it.  Whether the OS or machine returns the correct
resource information for the OS calls is a whole different matter.

<< So then why does the same directory cause an error in your program, and
not in my program? Have you tried the server admin using my scenario?
(dbserver and server admin on computer A, database on computer B). >>

Of course I tried it.  It's the first thing I always do before I even reply,
provided that I have all of the information needed to try and replicate the
issue.

<< PS: Don't take these things personally. It is just software.  >>

I'm not taking the fact that DBISAM may have flaws personally.  What I'm
taking personally are your continual attacks on my responses to you as
inadequate, in cases where there is no evidence that I done anything to
warrant such a response from you.

--
Tim Young
Elevate Software
www.elevatesoft.com

Robert

Are you running the dbsrvr as a service?  If you are can you stop the
service and try running the server as an application using the same user as
is running the svradmin program?

In my tests I was only getting the error if the privileges for the dbsrvr
process were not sufficient to "see/access" the share on the second machine

John

John,

<< In my tests I was only getting the error if the privileges for the dbsrvr
process were not sufficient to "see/access" the share on the second machine
>>

Good pickup.  I always forget about the service account restrictions.

--
Tim Young
Elevate Software
www.elevatesoft.com

"John Hay" <j.haywithoutdot@crbsolutionsremoveallthis.co.uk> wrote in
message news:049A5FDD-E8D1-48E1-A176-C3EEE43484C1@news.elevatesoft.com...
> Robert
>
> Are you running the dbsrvr as a service?  If you are can you stop the
> service and try running the server as an application using the same user
> as
> is running the svradmin program?
>
> In my tests I was only getting the error if the privileges for the dbsrvr
> process were not sufficient to "see/access" the share on the second
> machine
>

Thanks, that "solved" the problem. I'm still not sure why the dbsrvr process
does not have the privileges, but that's an issue for another day.

Robert

"Tim Young [Elevate Software]" <timyoung@elevatesoft.com> wrote in message
news:ECC1163C-B2DF-4A48-A772-C267FB01659A@news.elevatesoft.com...
>
> << PS: Don't take these things personally. It is just software.  >>
>
> I'm not taking the fact that DBISAM may have flaws personally.  What I'm
> taking personally are your continual attacks on my responses to you as
> inadequate, in cases where there is no evidence that I done anything to
> warrant such a response from you.
>

Tim, a response is "inadequate" (your word, not mine) when it does not point
to a solution to the problem. You posted a piece of code, I reproduced that
code in a program and got different results. But, as the kids say,
"whatever"....

Robert

Robert,

<< Tim, a response is "inadequate" (your word, not mine) when it does not
point to a solution to the problem. You posted a piece of code, I reproduced
that code in a program and got different results. But, as the kids say,
"whatever".... >>

What exactly do you expect ?  Are you insisting that the first response from
me always be the exact solution to your problem, with no exceptions ?  If
so, you're going to be very disappointed.

Also, did you realize that you asked this exact same question almost exactly
a year prior, and that I gave you the correct response immediately ?  You
even used the exact same title for the post:

http://www.elevatesoft.com/newsgrp?action=openmsg&group=12&msg=6996&page=1#msg6996

So, it appears that we both forget things over time and aren't perfect.

--
Tim Young
Elevate Software
www.elevatesoft.com