LFS201 Ch 06: Clarification on rebuilding RPM database + More and more questions :)
here it says
You can use the --rebuilddb option to rebuild the database indices from the installed package headers
What exactly is meant by 'installed package headers' ?
Are these files in specific install directories? And where are they exactly?
Also, i'm reading from fedora docs that we should back up the rpm database before running
rpm --rebuilddb. Is that done with rpm, or just
What situations in real life can result in corrupted rpm database, and why would rebuilding the database have risks? If the database is based on the header entirely, wouldn't backing up the database be superfluous? I mean, you would back up the source of information, not result of information
It take it the database isn't intrinsically necessary (they could have programmed rpm & dnf etc...to not use/need one) but rather is used for performance reasons, correct?