On 5 August 2015 at 14:21, Takashi Iwai <tiwai@suse.de> wrote:
In theory, all regressions should be curable by package reverts :) User may try to revert relevant packages if a regression occurs by the update.
And, having old packages allows developer to check whether the regression comes from which package more easily. Currently, even just for reproducing the reported problem, you'll have to build the old package manually at each time.
Of course, this is no silver bullet. A massive update like Plasma 5 or gcc update is a totally different story.
My very strong personal opinion on the topic is, if you want rollback like that, use snapper Like you said, Individual package reverts are not a silver bullet, but I'd argue they're worse than that. Everything in Tumbleweed is tested as a collective unit, and that's how it should be used. Mixing and matching from one snapshot to another is a dangerous messy affair If someone upgrades to a new snapshot and finds problems, rolling back to using snapper should be the safe and sane option they should consider. -- To unsubscribe, e-mail: opensuse-factory+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse-factory+owner@opensuse.org