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

Re: --initdb




On Dec 1, 2006, at 8:13 AM, James Olin Oden wrote:

On 11/30/06, Jeff Johnson <n3npq jbj gmail com> wrote:
On 11/30/06, Tony Nelson <tonynelson georgeanelson com> wrote:
> Does "rpm --initdb" ever delete database files, specifically the Packages > file? Is it safe to do "rpm --initdb ; rpm --rebuilddb"? I want to get
> straight what "rpm --initdb" does, so I don't give bad advice.
>
> I had read that --initdb created a fresh, new database, as implied by "man > rpm", wiping out the old database. Others have disagreed, and it seems > that they are right. I looked casually at the RPM source code for --initdb > (stopping when it called through a function pointer), and it looks like > --initdb only tries to open / create the database file(s), and won't
> replace existing files.  I ran "rpm -vv --initdb --dbpath
> /my/broken/database", and it only mentioned the Packages file.

The only current use in rpm for --initdb is to conform to the doco
so that I don't have to waste time and energy explaining to users why --initdb
hasn't been needed for years and years.

Which is to say rpm uses a lazy algorithm to create the initial rpm
database, such that you never have to explicitly call --initdb which
historically was used to create the initial rpm database.   In other
words, rpm creates the initial rpm database automatically.


Thanks for the translation.

If up to me, I would have removed --initdb a long long time ago. rpm
has way too many options, --checksig is another nearly useless option.

73 de Jeff


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