On Sun, Feb 23, 2014 at 10:58 PM, Reindl Harald <h.reindl@xxxxxxxxxxxxx> wrote: > > > Am 23.02.2014 22:40, schrieb Theodore Lee: >> On 24/02/14 06:29, Susi Lehtola wrote: >>> On Sun, 23 Feb 2014 18:12:55 +0100 >>> Reindl Harald <h.reindl@xxxxxxxxxxxxx> wrote: >>>> https://admin.fedoraproject.org/updates/FEDORA-2014-2922/libreoffice-4.2.1.1-1.fc20?_csrf_token=a6a024f6e2d35ad3f3333b8666c1244e215a6aa2 >>>> >>>> how can people pretend "installation went smoothly, no issue detected during basic >>>> document manipulation" for packages which are not installable at all due >>>> dependencie problems? >>> >>> People *couldn't* know there were problems, because all the positive >>> reports were from the time the update was in updates-testing. All who >>> tried the update, also had the dependency available in updates-testing. >> >> For what it's worth, my report (the first with the dependency issue) and >> a subsequent one were also from updates-testing, and both did not have >> the dependency available. > > they never did > >> I did do a manual check of Koji and Bodhi to try and figure out why my >> results were different from the previous testers, and could only find >> the necessary build in Koji, which quite frankly left me very confused >> and unsure if I was experiencing some kind of mirror sync issue and/or >> chronic lack of coffee syndrome. I now understand the initial positive >> karma results had something to do with a buildroot override > > which never hit updates-testing > > 2014-02-21 13:59:06 This update has been submitted for testing by dtardo > 2014-02-22 09:35:15 will be pushed to the stable updates repository > > this is *unacceptable* in case of broken deps and buildroot overrides > while this is not a secuity update and people are pushing such things > to stable refuse to understand the the dependecy error may result in > *not get whatever SECURITY UPDATE* for ordinary users for no gain That's not true. packagekit / gnome-software / dnf use an "--skip-broken" equivalent by default iirc. > and no "you have to apologize" from the maintainer does change that > > if the maintainers would run a baisc virtual machine consuming > ordinary repos without manual overrides such mistakes would be > recognized by them.............. Well the proper way to fix is to have automated and enforced dep checks ... <insert reply from AdamW here telling me that it is not ready for that yet ;) > -- devel mailing list devel@xxxxxxxxxxxxxxxxxxxxxxx https://admin.fedoraproject.org/mailman/listinfo/devel Fedora Code of Conduct: http://fedoraproject.org/code-of-conduct
