My favorite pet bug (2004): Yum mishandles Ctrl-C

[Frank Ch. Eigler]

"Patrick W. Barnes" <nman64 at n-man.com> writes:

> [...]  Going further on the database terms, most of the work you do
> on a transaction is *preparing* the transaction.  Once the
> transaction is *ready*, you have a final opportunity to cancel the
> transaction before you *commit* the transaction.  The commit stage
> is where the real work of the transaction is done, and it is the one
> stage that you don't want to interrupt.  [...]  or there's no
> telling what sort of condition the database will be left in.  [...]

Maybe baby databases without proper logging have problems with this
sort of thing (interruption during a commit), but adult ones can roll
back even then.  If they couldn't, they wouldn't be atomic in the face
of abrupt system failures.

If RPM maintained a proper log of its actions during installation, it
could be more of a ... let's say teenager database than a baby one.

- FChE