On Sat, Nov 22, 2008 at 9:40 PM, Eric Bélanger <[EMAIL PROTECTED]> wrote: > On Sat, 22 Nov 2008, Aaron Griffin wrote: > >> On Sat, Nov 22, 2008 at 8:06 PM, Eric Bélanger >> <[EMAIL PROTECTED]> wrote: >>> >>> Hi, >>> >>> The db script for extra x86_64 is stuck: >>> >>> $ /arch/db-extra64 >>> error: db generation is already in progress (started by andyrtr) >>> rm: cannot remove `/srv/tmp/.repolock.extra.x86_64': Operation not >>> permitted >>> >>> That was started hours ago. Can someone with root power remove the lock >>> file >>> to fix it? >> >> Fixed it. I'm fixing some oopses I made with the dbscripts now. If >> this happens again, let me know - I commented out the chmod crap for >> now >> >> > > I'm not sure if it's part of the oopses, but it seems that the problem is > that the script for x86_64 doesn't remove the lock file because it can't > find it. Error message on running db-extra64: > error: repo lock doesn't exist... something went terribly wrong! > > A /srv/tmp/.repolock.extra.x86_64 lockfile was left with 644 permissions for > my user. I removed it manually to not block updates for other users.
I think this is related to bug 10888 - where sourcing the PKGBUILD overwrites the $arch local var, so it's always i686 (as $foo of an array, foo, evaluates to the first array item, which is usually i686). I will have this fixed in a few hours or so
