Description of problem: When a package is moved from updates_testing to updates, and an older was alredy in here, we have two . So we should just keep the latest one. Why ? 1.It make rpmdrake easier to read and example with vlc (without tainted + 32bits ) 1.1.12-3.1.mga1 // core-updates (Mga, 1, x86_64) 1.1.11-0.1.mga1 // core-updates (Mga, 1, x86_64) So for x86_64 user it's about 6 rpm just for vlc. (without release) 2. It can also decrease the data. example with firefox: 0:7.0.1-0.1.mga1 0:6.0.2-0.1.mga1 0:6.0.1-1.2.mga1 0:6.0-1.1.mga1 each srpm is about 65Mo.
Hi, +1 for this one Other example un Core Updates Debug - kdelibs4-debug 4.6.5-1.1.mga1 - kdelibs4-debug 4.6.5-1.2.mga1 - kdelibs4-debug 4.6.5-1.3.mga1 and if you don't take attention (a long list of updates by example) you select several updates for one package and you have a lovely message like "kdelibs4-debug-4.6.5-1.3.mga1.i586 (en raison de conflit avec kdelibs4-debug-4.6.5-1.2.mga1.i586)". I had this one (very) recently and I first thought a dependancy problem and occupied to open a ticket on bugzilla ... ;-) A+ Raph
CC: (none) => mailinglistsduraph
ping for repoctl :p
So I have check in mdv, only kernel ( and the relative fglrx-kernel-*, etc) are keep. So what about have the same ?
dup *** This bug has been marked as a duplicate of bug 2382 ***
Status: NEW => RESOLVEDResolution: (none) => DUPLICATE