Assuming the new machine(s) were brought online as I see mine schedule and built just fine. ;)
-- Jimmy On Fri, Nov 6, 2015 at 3:53 PM, Jimmy Berry <[email protected]> wrote: > I see the updated Blender package has been published. Thanks for > knocking this out. Just tested and I can drag mp4 into video editor > which I cannot do with base one. I'll go ahead and delete mine. > > Thanks again! > > -- > Jimmy > > > On Wed, Nov 4, 2015 at 6:38 AM, Dave Plater <[email protected]> wrote: >> On 11/4/15, Stefan Botter <[email protected]> wrote: >>> Hi Jimmy, >>> >>> sorry, I hit "send" prematurely... >>> >>> On Tue, 3 Nov 2015 00:42:47 -0600 >>> Jimmy Berry <[email protected]> wrote: >>> >>>> I branched obs package, merged in previous packman build changes to >>>> enable ffmpeg, and cleaned up a bit. I was able to successfully build >>>> for Tumbleweed x86_64. When I installed locally the package worked >>>> great and I was able to use features that depend on ffmpeg which I am >>>> unable to do in the obs version. >>>> >>>> See for successful build: >>>> https://pmbs.links2linux.de/package/binaries/home:boombatower:branches:Extra/blender-update?repository=openSUSE_Tumbleweed >>>> >>>> Unfortunately, the majority of the time the packman obs instance seems >>>> to just loop trying to build the package since it ends up terminating >>>> the build process at some random percentage through the build. It >>>> seems like it is running out of ram. >>> >>> Seems possible, as most workers are running chroot builds, where there >>> are no guaranteed resources, and hence builds with contraints are not >>> scheduled to build. ATM only swkj11-swkj14 are workers building in KVM, >>> and all but 11 had a configuration problem throughout October - so >>> nothing was built there. They are fixed now. >>> >>>> The previous build on packman had >>>> the following memory _constraint: >>>> >>>> <physicalmemory> >>>> <size unit="M">1700</size> >>>> </physicalmemory> >>>> >>>> I upped to 2000, but then it seems to never be scheduled. I am >>>> assuming none of the packman workers have that much ram? >>> >>> swkj11 has 1897MB RAM available for build, swkj12-14 have 1950MB. A >>> _constraint file requesting more than these values will block the >>> package from being scheduled to build. >>> >>>> If so does anyone have suggestions for different build flags to >>>> possibly reduce ram usage during compile or acquire a more powerful >>>> worker or two? >>> >>> As mentioned some days ago, I was able to secure some more hardware, >>> and am currently - as time permits - in the phase to repair and enable >>> it for use - just Yesterday evening I was able to restart the workers >>> 15-19 (virtual machines doing chroot builds), which before were running >>> on old desktops, which I have to return to their proposed use soon. >>> There is a test machine swkjt01 currently providing 4 processes in KVM, >>> but providing 1GB RAM, only. >>> >>> I have some more machines, and provided I can fix them, I will at least >>> one of them configure to run with more than 2 GB instance memory. >>> >>> As said, as time permits, perhaps over the weekend. >>> >>> >>> Greetings, >>> >>> Stefan >>> -- >>> Stefan Botter zu Hause >>> Bremen >>> >> >> I've got maintainer rights for Extra/blender and have synced it with >> openSUSE but it Factory and Tumbleweed have multiple ffmpeg related >> unresolvables, Leap42 builds though. I'm going to wait for it to >> settle for a while, it built for Factory and Tumbleweed in home:davepl >> before I submitted to Extra. >> Regards >> Dave >> >> _______________________________________________ >> Packman mailing list >> [email protected] >> http://lists.links2linux.de/cgi-bin/mailman/listinfo/packman _______________________________________________ Packman mailing list [email protected] http://lists.links2linux.de/cgi-bin/mailman/listinfo/packman
