On Fri, Sep 17, 2010 at 8:56 PM, Jonathan Lange <[email protected]> wrote: > On Fri, Sep 17, 2010 at 5:23 AM, Robert Collins > <[email protected]> wrote: >> This has been growing in importance over the last 2 cycles but as we >> get closer and closer to switching workflows its now actually >> important and relevant. >> >> The key message is: >> >> *Please do not land -anything- on devel unless it is ok to be deployed >> on lpnet*. > > Just to be clear, that means: > * don't ever land a thing on devel that requires cronscripts, db, or > updates to non-appservers
db: land on db-devel cronscripts: - if deploying it to the production cronscripts is unwanted until the next release, land it on db-devel (or guard it with a feature flag) other non-appservers: - if deploying it to $wherever is not wanted until the next release, land it on db-devel (or guard it with a feature flag) > * it's ok to land stuff that might fail QA (although obviously we > shouldn't land crap) > > Is that right? Yes. Note that stuff which fails QA will need to be backed out or fixed before we can resume deployments, so its a pipeline stall (but OTOH we want to have a very small pipeline..) -Rob _______________________________________________ Mailing list: https://launchpad.net/~launchpad-dev Post to : [email protected] Unsubscribe : https://launchpad.net/~launchpad-dev More help : https://help.launchpad.net/ListHelp
