Hi
            Ernesto,
         
        Papyrus-RT is hosted on HIPP4 [1], which is
            barely used [2]. So server issues are very unlikely in this
            case
         
        However, since Papyrus-RT moved to
            Target-Platform-based builds, it is now downloading all
            Eclipse dependencies via HTTP (And much more dependencies
            than before, as well, because it is downloading Eclipse SDK
            and Papyrus SDK, including lots of documentation). So this
            is probably the main issue
         
        In Papyrus, we solved this by duplicating each
            target platform, to provide an Eclipse-server specific
            version (Using file:/ URIs instead of http://). The drawback
            being that the build is then harder to maintain (But we
            already have some tools for that in Papyrus, so it works).
            Reusing these Releng Tools in Papyrus-RT may be an option,
            although they have been quite specifically designed for
            Papyrus. Anyway, this may be part of Céline’s remaining
            tasks on the build (I don’t know the exact plans)
         
        [1]
            https://hudson.eclipse.org/
        [2]
            https://dev.eclipse.org/committers/help/status.php
         
        Regards,
            Camille
         
        De :
            papyrus-rt-dev-bounces@xxxxxxxxxxx
            [mailto:papyrus-rt-dev-bounces@xxxxxxxxxxx]
            De la part de Ernesto Posse
            Envoyé : vendredi 20 mai 2016 00:44
            À : papyrus-rt developer discussions
            <papyrus-rt-dev@xxxxxxxxxxx>
            Objet : [papyrus-rt-dev] Hudson jobs taking much
            longer than usual
         
        
          
            Hi. Several codegen+rts jobs have been
              automatically aborted recently due to timeouts (which
              happen after 30min). This is really strange, as typically
              those jobs take less than 10 minutes.
           
          
          
            Is this due to a heavy load on the
              server or something else?
           
          
          
            I remember we had a similar issue
              before and it had something to do with getting the p2
              repos hosted in the eclipse server via http. Could that be
              the problem again?
           
          
          
          --