In general, when making changes that we expect to break a large number of other people, we try to minimize the time of the breakage by either (a) providing patches that will fix things or (b) engaging with downstream consumers to help them create the patch. In particular, I'd encourage you to go look at these best practices which the TSC formally adopted after some significant cross-project breakage pain: https://wiki.opendaylight.org/view/TSC:Main#Best_practices_for_preventing.2Fhandling_cross-project_breakages
--Colin On Mon, Nov 14, 2016 at 3:56 AM, Jozef Bacigál <[email protected]> wrote: > Why we have already weather process ? I thought if I made this everyone > should stay alert and fix if an issue in his projects. Here in ODL is some > unwritten rule if bigger change is coming in one project then we > should/need/must to create patches for all touched projects ? Why the > projects itself won’t fix any issue or react for weather or changes. And > even then if you create an patch for some project the project won’t merge > it because they don’t read mailing list. > > > > Jozef > > > > *From:* Colin Dixon [mailto:[email protected]] > *Sent:* Thursday, November 10, 2016 7:53 PM > *To:* Robert Varga <[email protected]> > *Cc:* openflowplugin-dev <[email protected]>; > Release ([email protected]) <[email protected]>; > Daniel Malachovsky -X (dmalacho - PANTHEON TECHNOLOGIES at Cisco) < > [email protected]> > *Subject:* Re: [openflowplugin-dev] [release] [WEATHER] - topology id in > openflowplugin > > > > Something something. Reality vs. practice. Something something. > > More seriously, if we have a suggested best practice to avoid hard coding > this in and then can search through all the projects and help them move > away from it, I'm sure that would be appreciated by lots of people. > > > > --Colin > > > > On Thu, Nov 10, 2016 at 1:12 PM, Robert Varga <[email protected]> wrote: > > On 11/10/2016 06:44 PM, Colin Dixon wrote: > > I think flow:1 is hard code a lot of places even inside OpenDaylight... > > That is a *really* bad idea. This is equivalent of assuming a particular > implementation of an interface. > > Whoever is using this hard-coded value should be getting a reference to > a topology instead (and check whether the injected topology conforms to > expectations by advertising a topology-type). > > Bye, > Robert > > > > JozefBacigál > > Software Engineer > > > Sídlo / Mlynské Nivy 56 / 821 05 Bratislava / Slovakia > R&D centrum / Janka Kráľa 9 / 974 01 Banská Bystrica / Slovakia > +421 908 766 972 / [email protected] > reception: +421 2 206 65 114 / www.pantheon.sk > > [image: logo] > > >
_______________________________________________ openflowplugin-dev mailing list [email protected] https://lists.opendaylight.org/mailman/listinfo/openflowplugin-dev
