Hi,

On Thu, Aug 28, 2008 at 4:10 AM, Adriano Marques <[EMAIL PROTECTED]> wrote:
> Loading xml in memory is not reasonable. Nmap can *really* output some
> *really* verbose and *really* huge xml files, which we wouldn't ever
> mind to have then loaded in memory. That's why we moved from dom
> parsing to sax.

This is true indeed.

>
> So, I'm not fully aware of how are things working on the proposed
> umitdb, but it looks reasonable to mix the files with the database. We
> could maybe keep a summary of the xml file content inside the
> database, and decide to look inside the xml if the summary is
> promissing based on the search term used.

There is some confusion here, probably because of the similar umitdb
and umitDB names (that is why I used to refer to umitdb-ng when I
talked about mine). I think it is a good time to quickly talk about
them, since people that joined umit this year may not be aware of them
at all, so here it goes (although I may miss something since there has
been some time since I looked at them):

umitdb:
  - The current one used by umit
  - It keeps the entire nmap xml (with fields added by usr format)
file in the database.
  - umit uses this, plus the completed scans in the open tabs, to search scans;

umitDB:
  - The current one used by network inventory
  - Breaks the nmap xml output and store in database, optionally the
entire xml file can be stored too;
  - doesn't include usr fields in the broken format.
  - network inventory uses this to tell differences between scans;
  - network inventory also uses this to search scans made;

I dislike very much this option of storing the entire xml in the
database, because we miss the point of using a database. Instead we
should improve it and store all the other fields we intend to look
for. This then introduces the problem of recreating the xml, so I
would need to think about this before dropping support for storing the
entire xml.

>
> Or, maybe I'm just wrong and we can keep the xml files inside the
> database and search them from there directly.

That is what umit does right now.

>
> Guilherme, do you remember what was the problems you were facing when
> you decided to move the files outside the database? Maybe that could
> help us make a better decision.
>

I think I never decided that, except if I forgot about it. It seems,
somehow, Luis introduced some confusion here (or the db names were the
reason).

Me and Luis had a short talk in IRC, where he started talking about
merging umitdb and umitDB. Then I told him how umitdb works, and why
it could potentially stop existing.
What I suggested was dropping umitdb and storing the scans in
~/.umitdb/scans/, for example. umitdb currently doesn't do much as a
database, that is why this suggestion took place.

>
> Cheers!
>
> --
> Adriano Monteiro Marques
>
> http://adriano-marques.blogspot.com
> http://umit.sourceforge.net
> [EMAIL PROTECTED]
>
> "Don't stay in bed, unless you can make money in bed." - George Burns
>
> -------------------------------------------------------------------------
> This SF.Net email is sponsored by the Moblin Your Move Developer's challenge
> Build the coolest Linux based applications with Moblin SDK & win great prizes
> Grand prize is a trip for two to an Open Source event anywhere in the world
> http://moblin-contest.org/redirect.php?banner_id=100&url=/
> _______________________________________________
> Umit-devel mailing list
> [email protected]
> https://lists.sourceforge.net/lists/listinfo/umit-devel
>



-- 
-- Guilherme H. Polo Goncalves

-------------------------------------------------------------------------
This SF.Net email is sponsored by the Moblin Your Move Developer's challenge
Build the coolest Linux based applications with Moblin SDK & win great prizes
Grand prize is a trip for two to an Open Source event anywhere in the world
http://moblin-contest.org/redirect.php?banner_id=100&url=/
_______________________________________________
Umit-devel mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/umit-devel

Reply via email to