+1 to Mark's comments. On Fri, Aug 5, 2016 at 4:57 AM Mark Waite <[email protected]> wrote:
> Replies inline > > > On Fri, Aug 5, 2016 at 1:51 AM Manuel Jesús Recena Soto <[email protected]> > wrote: > >> Hello Mark, >> >> In advance, thanks for your feedback. >> >> (replies inline) >> >> 2016-08-04 21:27 GMT+02:00 Mark Waite <[email protected]>: >> > >> > On Thu, Aug 4, 2016 at 12:46 PM Manuel Jesús Recena Soto < >> [email protected]> wrote: >> >> >> >> Hello Oliver, >> >> >> >> I like the idea. >> >> >> >> General speaking, ATH and PCT are very important and we should look >> after its healthy. >> >> >> >> IMHO, PCT results are a nice place to figure out bad symptoms on some >> plugins. >> >> >> > >> > Can you describe the types of symptoms you envision detecting with PCT? >> > >> > I am interested in running the plugin tests with a different >> jenkins.version value, after compiling with an unmodified jenkins.version >> value. That seems like an interesting test alternative that might surface >> interesting differences in the test harnesses or the core code. >> > >> > I'm not really interested in having some external automation tell me >> that the current code does not compile with the latest Jenkins core. >> JENKINS-36646 seems like the same condition I see, failure to compile with >> the newer Jenkins version because code changes are required before the >> plugin can depend on the newer version. >> > >> >> 1. You can always ignore that information. There may be developers >> interested on it. >> 2. I totally agree with you that JENKINS-36646 is not a bug, but >> something that it should be investigate. I'd say: a task. >> >> > I update the Jenkins core dependency rarely and with great care. A bug >> report like JENKINS-36646 is a distraction rather than a help, since I will >> see the compilation failure as soon as I prepare the change for the plugin >> to support the new Jenkins core version. >> > >> > For example, the git plugin 3.0.0-beta already has the necessary change >> to compile with more recent Jenkins core, but the automation probably won't >> know that it needs to evaluate a beta version on a different branch than >> master. >> > >> >> IMHO, when we find this situation: a plugin based on an old Jenkins >> base line + not recent releases > bad symptom (a red flag in my >> production environments). >> >> PCT helps on these cases. >> >> > Sorry that I'm not making myself clear. The git client plugin (1.19.7 and > it predecessors) matches the situation you describe (old Jenkins baseline, > not recent release, does not compile without changes against current > core). It has intentionally been kept at that older release in order to > not create unnecessary compatibility issues for users of its APIs. > > When git client plugin 2.0 releases (planned by mid-September) it will > update to depend on JDK 7, Jenkins 1.625, and JGit 4. That will be an > intentional major version of the plugin to switch from JDK 6 to JDK 7. > Even then, there is a risk that it may not compile against Jenkins 2.7.2. > Acceptance tests are run regularly against Jenkins 2.7.2, but it is not > regularly checked that it compiles against 2.7.2 because I've not found > that check especially helpful. > > I really have no problem ignoring infrastructure that tells me irrelevant > things are failing, but when irrelevant failures are reported to me, I > develop the habit of ignoring more and more information from that source. > Sometimes, that means I miss a real problem because I've been ignoring the > irrelevant. > > Mark Waite > > >> Regards, >> >> > Mark Waite >> > >> >> >> >> Please, find here two issues as result of PCT: JENKINS-36646 and >> JENKINS-36623 >> >> >> >> What we maybe need is make more usable and accessible this information. >> >> >> >> Regards, >> >> >> >> >> >> 2016-08-03 15:31 GMT+02:00 Oliver Gondža <[email protected]>: >> >>> >> >>> Hi, >> >>> >> >>> I created an issue[1] to verify if all recommended plugins has >> passing unittest when run against latest core. I would like to use that as >> a GA criteria for releases, LTS ones at least. >> >>> >> >>> I see people are still updating plugin-compat-tester so I guess it is >> not dead, though I did not manage to find any recent documentation or >> running CI job. Can it be used to implement that? Is somebody working on >> this? >> >>> >> >>> [1] https://issues.jenkins-ci.org/browse/JENKINS-37145 >> >>> -- >> >>> oliver >> >>> >> >>> -- >> >>> You received this message because you are subscribed to the Google >> Groups "Jenkins Developers" group. >> >>> To unsubscribe from this group and stop receiving emails from it, >> send an email to [email protected]. >> >>> To view this discussion on the web visit >> https://groups.google.com/d/msgid/jenkinsci-dev/08d33372-c328-71d5-a2c6-870976e3b688%40gmail.com >> . >> >>> For more options, visit https://groups.google.com/d/optout. >> >> >> >> >> >> >> >> >> >> -- >> >> Manuel Recena Soto >> >> * manuelrecena.com [/blog] >> >> * linkedin.com/in/recena >> >> >> >> -- >> >> You received this message because you are subscribed to the Google >> Groups "Jenkins Developers" group. >> >> To unsubscribe from this group and stop receiving emails from it, send >> an email to [email protected]. >> >> To view this discussion on the web visit >> https://groups.google.com/d/msgid/jenkinsci-dev/CABa-UocUg78nQkNas9iyKK0JLKZvv%3DuT9eDoZLn1f0YYfsTPSA%40mail.gmail.com >> . >> >> For more options, visit https://groups.google.com/d/optout. >> > >> > -- >> > You received this message because you are subscribed to the Google >> Groups "Jenkins Developers" group. >> > To unsubscribe from this group and stop receiving emails from it, send >> an email to [email protected]. >> > To view this discussion on the web visit >> https://groups.google.com/d/msgid/jenkinsci-dev/CAO49JtFWnPvcmjCXE3ikhe9cFUSjNbnoLzuBq9AQU49gwB7zYw%40mail.gmail.com >> . >> > >> > For more options, visit https://groups.google.com/d/optout. >> >> >> >> >> -- >> Manuel Recena Soto >> * manuelrecena.com [/blog] >> * linkedin.com/in/recena >> >> -- >> You received this message because you are subscribed to the Google Groups >> "Jenkins Developers" group. >> To unsubscribe from this group and stop receiving emails from it, send an >> email to [email protected]. >> To view this discussion on the web visit >> https://groups.google.com/d/msgid/jenkinsci-dev/CABa-UocPDEZ%3DzX1DZbbO3pOc0W86n3tbPA%2B%3DW7ShVqPQSMzMTg%40mail.gmail.com >> . >> For more options, visit https://groups.google.com/d/optout. >> > -- > You received this message because you are subscribed to the Google Groups > "Jenkins Developers" group. > To unsubscribe from this group and stop receiving emails from it, send an > email to [email protected]. > To view this discussion on the web visit > https://groups.google.com/d/msgid/jenkinsci-dev/CAO49JtH5mNWCVt9T3vwTOb%2Buv1wpizRjWg6dAjboZ98ioMnd0A%40mail.gmail.com > <https://groups.google.com/d/msgid/jenkinsci-dev/CAO49JtH5mNWCVt9T3vwTOb%2Buv1wpizRjWg6dAjboZ98ioMnd0A%40mail.gmail.com?utm_medium=email&utm_source=footer> > . > For more options, visit https://groups.google.com/d/optout. > -- You received this message because you are subscribed to the Google Groups "Jenkins Developers" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. To view this discussion on the web visit https://groups.google.com/d/msgid/jenkinsci-dev/CAPiUgVcvjVPZy9Kva_oxvyjY0wYcEiLO7K1Oy%3DOAiAZ5U1owgg%40mail.gmail.com. For more options, visit https://groups.google.com/d/optout.
