[pacman-dev] (Most likely) final RC before going to testing
Alessio 'mOLOk' Bolognino
themolok.ml at gmail.com
Sun Feb 25 12:08:02 EST 2007
On 11:34 Sun 25 Feb , Dan McGee wrote:
> On 2/25/07, Alessio 'mOLOk' Bolognino <themolok.ml at gmail.com> wrote:
> > On 08:59 Sun 25 Feb , James Rosten wrote:
> > > >
> > > > When I'm doing a system upgrade or installing a package and i run
> > > > makepkg3, it complains for pacman already running and tell that can
> > > > not
> > > > do the dependency check or so.
> > > > Is this a bug or an intended behaviour?
> > >
> > > Could you tell us what makepkg3's output was because I think
> > > stonecrest mentioned this on IRC and knowing if it is the same thing
> > > would be helpful.
> >
> > here it is:
> >
> > (12000)$> makepkg3
> > ==> Entering fakeroot environment
> > ==> Making package: gpomme 1.1-3 (Sun Feb 25 16:34:25 CET 2007)
> > ==> Checking Runtime Dependencies...
> > error: unable to lock database
> > ==> ERRORe: pacman3 returned a fatal error (1):
> > if you're sure a package manager is not already running,
> > you can remove /tmp/pacman.lck
> > ==> Checking Buildtime Dependencies...
> > ==> Retrieving Sources...
> > -> Found gpomme-1.1.tar.gz in build dir
> > ==> Validating source files with md5sums
> > gpomme-1.1.tar.gz ... Passed
> > ==> WARNING: Integrity checks (sha1) are missing or incomplete.
> > ==> Extracting Sources...
> > -> tar -xf gpomme-1.1.tar.gz
> > ==> Removing existing pkg/ directory...
> > [...]
> >
>
> Did you try doing another pacman operation that locks the DB right
> after? That would have likely failed too.
Of course it does.
> If you read the message, it
> indicates that the lock file exists, which happens if pacman is
> running an operation that requires a locked DB. Try following the
> instructions given- if you are sure pacman is not running, remove the
> lock file.
Indeed pacman was running. I got that output while pacman was running,
but I think that an operation like dependency check that only reads the
db should be allowed. Maybe it's a design choice for pacman3, because
pacman2 doesn't complain for it.
(Now Andreas Radke opened a thread for this behaviour)
--
Alessio 'mOLOk' Bolognino
Arch Linux Trusted User
http://www.archlinux.org
More information about the pacman-dev
mailing list