[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [Bacula-devel] libdbi backend to catalog database


On Saturday 23 February 2008 03.07:05 João Henrique Freitas wrote:
> Very good notice
>
> On Fri, Feb 22, 2008 at 2:42 PM, Kern Sibbald <kern@xxxxxxxxxxx> wrote:
> > Hello João,
> >
> >  I had a few problems integrating your patch into the SVN trunk, but it
> > is now in, and I would appreciate it if you could test it to make sure it
> > got in correctly.
> >
> >  Things that are different:
> >  - Most of your changes to autoconf/config.h.in failed, but I think the
> >   essential ones got in.
> >  - Some of your changes to configure.in were rejected, but it seems to
> >   be OK.
>
> Really, autoconf/config.h and configure.in is two files much boring to
> do. But I need read and learning more about this.

Yes, it is boring and insanely complicated. It took me more than 3 years 
before I felt reasonably comforatble modifying it ...

>
> >  - You had included changes to cats.h and sql.c that I aready had in the
> > svn so they caused failures.
>
> Yes, in svn (sql.c and cats.h) SQL_TYPE_POSTGRE. I changed to
> SQL_TYPE_POSTGRESQL. Nothing more.

Those changes were lost due to the fact that the patch was based an old 
version of the code.  However, I have put it back.

>
> >  - I had to turn off the change that you made to dird/bacula-dir.conf.in,
> > since your change would probably break the default behavior, and would be
> > a bit confusing to most users.
>
> Ok
>
> >  - I had to move the port option from -p to -t.  You cannot simply change
> > the meaning of an option -- that could break users scripts, even though
> > the -p option is not often used.  In addition, the -p is used in other
> > programs to have the same meaning.  If necessary we can discuss changing
> > this, but we need to look at it carefully.
>
> No problem.
>
> >  I have run the regression scripts using MySQL here, and it works, now I
> > am running the full tape regression tests, but all seems OK -- that is
> > nothing is broken.  I haven't yet tried to configure or run a dbi driver.
> >   Hopefully I can get to that next week.
> >
> >  Thanks for all the hard work.  Please take a look at how the code went
> > in, and tomorrow I will take a look at your list of questions or
> > remaining things to be done ...
>
> In svn trunk, everything is Ok. compiling and run without problems.
>
> >  > The following itens need some work or review:
> >  >
> >  > 1- Implement bath insert with dbi
>
> Now, I will starting this item.

Yes, this is a bit complicated, but it is an important item for performance.
If you want to understand better how it works, you can read the document:

 docs/techlogs/batch_insert_documentation.odt

>
> >  > 2- Review the options ./configure --with-dbi

What needs reviewing here?

> >  > 3- To solve/review the make install when ./configure --with-dbi.
> >  > Currently,the database scripts don't  is installed on system by make
> >  > install command.

Can you explain the make install problem a bit more?  i.e. exactly what 
scripts need to be installed?

> >  > 4- Review in bscan patchs
> >  > 5- Documentation to install, configure and compiling bacula and libdbi


Kern


-------------------------------------------------------------------------
This SF.net email is sponsored by: Microsoft
Defy all challenges. Microsoft(R) Visual Studio 2008.
http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/
_______________________________________________
Bacula-devel mailing list
Bacula-devel@xxxxxxxxxxxxxxxxxxxxx
https://lists.sourceforge.net/lists/listinfo/bacula-devel


This mailing list archive is a service of Copilot Consulting.