+1 on Kevin's suggestions, but I prefer we not even consider testing debug builds unless we have extra time.
Thanks, Josh On Tue, Sep 27, 2016 at 1:51 PM, Kevin Kane via Allseen-core < [email protected]> wrote: > Probably not. If you find an issue in release, the best course at that > point is for someone to start triaging the issue, which may then lead to > re-running the failing test cases with trace logging enabled and/or using a > debug build. I don’t expect re-running the test case on a debug build will > be useful frequently enough to have it be standard procedure for every > failure. > > > > I consider running all test cases in debug a nice to have since it means > assertions will get exercised, but if we don’t have the resources for it, > that’s that. > > > > *From:* Hertel, Carrie Lynn [mailto:[email protected]] > *Sent:* Tuesday, September 27, 2016 11:37 AM > *To:* Kevin Kane <[email protected]>; allseen-core@lists. > allseenalliance.org > > *Subject:* RE: Debug vs. Release build testing - 16.10 > > > > Thank you for your input Kevin! Would running in release and then > retesting in debug if we find an issue be helpful? With the IPv6 addition, > our testing is going to be tight with the extra validation. > > > > Thanks, > > Carrie > > > > *Carrie Hertel, PMP* > > Software Dev Group & Test Lab Director > > Center for Applied Systems & Software > > Oregon State University > > > > *From:* Kevin Kane [mailto:[email protected] <[email protected]>] > *Sent:* Tuesday, September 27, 2016 10:56 AM > *To:* Hertel, Carrie Lynn <[email protected]>; > [email protected] > *Subject:* RE: Debug vs. Release build testing - 16.10 > > > > My personal opinion: > > > > There’s value in running test scenarios on debug builds because code > assertions are compiled in and checked, and test cases might catch some > violated assertions we aren’t catching elsewhere. > > > > If we have to choose only one, absolutely run on release build since that > is what will be shipping in product. > > > > This then becomes a resourcing decision on whether or not we want to run > both, and I’ll let others with knowledge of our testing resources weigh in > on that. > > > > *From:* [email protected] [ > mailto:[email protected] > <[email protected]>] *On Behalf Of *Hertel, > Carrie Lynn > *Sent:* Tuesday, September 27, 2016 10:19 AM > *To:* [email protected] > *Subject:* [Allseen-core] Debug vs. Release build testing - 16.10 > > > > Hi All – > > > > I was informed by my testers that unless a test case lists a specific > build, that they need to run it on both debug *and* release variants. > > > > I would expect that if we are in release testing, that we could test on > release variant only unless the test calls for debug specifically. > > > > Any objections? Currently we are running a lot of tests twice when I > don’t think we need to… > > > > Thanks! > > Carrie > > > > *Carrie Hertel, PMP* > > Software Dev Group & Test Lab Director > > Center for Applied Systems and Software > > Oregon State University – School of EECS > > 209 Milne CC > > Corvallis, OR 97331 > > office: 541-737-6012 > > cell: 541-619-0849 > > [email protected] > > CASS.oregonstate.edu > <https://na01.safelinks.protection.outlook.com/?url=cass.oregonstate.edu&data=01%7C01%7Ckkane%40microsoft.com%7Cc3b215a466b14dc938ee08d3e6fa69ed%7C72f988bf86f141af91ab2d7cd011db47%7C1&sdata=6hmaiA5jvC827esi2c%2BcKlIGnU%2BbrgH6vIr1ZaJpSSs%3D&reserved=0> > > > > > > _______________________________________________ > Allseen-core mailing list > [email protected] > https://lists.allseenalliance.org/mailman/listinfo/allseen-core > >
_______________________________________________ Allseen-core mailing list [email protected] https://lists.allseenalliance.org/mailman/listinfo/allseen-core
