Where communities thrive


  • Join over 1.5M+ people
  • Join over 100K+ communities
  • Free without limits
  • Create your own community
People
Repo info
Activity
    Timothy Perrett
    @timperrett
    which is still an order of magnitude better than what most people are doing
    Rob Norris
    @tpolecat
    interesting, i had no idea everything was so terrible. that make me feel a little better
    ;-)
    Timothy Perrett
    @timperrett
    oh yeah, operations is pretty bad
    at scale its just a really hard problem
    i mean, think about the kinds of workload most people have: either idle or spikey is probally most common
    very few workloads are banging the cpu 24/7
    Rob Norris
    @tpolecat
    yep.
    Timothy Perrett
    @timperrett
    so my changes to the release process dont seem to take affect, will need to look at this more another time
    as its getting late
    Rob Norris
    @tpolecat
    Alright, no worries. Now I know how to survive a busted master, so that was a useful exercise.
    Timothy Perrett
    @timperrett
    haha well i wouldnt recomend that course of action in the general case, as its a bit nuclear
    but whatever
    :D
    it works!
    Rob Norris
    @tpolecat
    It worked really well actually.
    Timothy Perrett
    @timperrett
    yeah its a nice hack
    Timothy Perrett
    @timperrett
    @tpolecat alright all the build changes are in play - whenever you can just merge from master, either myself or @stew can merge it. I’ll be AFK for a few hours today, so whomever gets to it first. travis will automatically spin a release build.
    Rob Norris
    @tpolecat
    @timperrett @stew ok it's re-merged but the test failed because of network issues looks like .. pls restart the failed one here https://travis-ci.org/oncue/funnel/builds/92603316
    Mike (stew) O'Connor
    @stew
    restarted
    Rob Norris
    @tpolecat
    ta
    Rob Norris
    @tpolecat
    sigh. now it failed with exit code 137
    Timothy Perrett
    @timperrett
    @tpolecat that probally means our build process is using too much ram - lemme run your branch locally
    im tempted to just merge it and see what we can do about the build
    @tpolecat Error code 137 usually comes up when a script gets killed due to exhaustion of available system resources, in this case it's very likely memory. The infrastructure this build is running on has some limitations due to the underlying virtualization that can cause these errors.
    Rob Norris
    @tpolecat
    curious. i'll give it a go locally
    Timothy Perrett
    @timperrett
    yeah - also note that your PR did actually pass for scala 2.10
    it only failed for scala 2.11
    https://travis-ci.org/oncue/funnel/builds/92520044
    Rob Norris
    @tpolecat
    right
    Timothy Perrett
    @timperrett
    so, thats a marked improvement from where we were :D
    haha
    Rob Norris
    @tpolecat
    well they both passed for a while, then i rebased again and it failed. i'm guessing it's nondeterministic
    Timothy Perrett
    @timperrett
    in the case of 137 exit code, yeah, not a lot we can do about that on the shared oss infrastructure
    other than re-run it
    running out of system resources would be non-deterministic
    naturally
    Rob Norris
    @tpolecat
    seems like it would be equally likely to fail in 2.10
    Timothy Perrett
    @timperrett
    im curious though, with this, one gets two builds: https://github.com/oncue/funnel/pull/18/files#diff-354f30a63fb0907d4ad57269548329e3R4
    however
    with the way that crossVersions is setup
    https://travis-ci.org/oncue/funnel/jobs/92520046
    i think you’re doing both 2.10 and 2.11 in each version
    that job i restarted was the 2.11 job
    Rob Norris
    @tpolecat
    right
    Timothy Perrett
    @timperrett
    but its downloading 2.10 dependencies
    which seems… pointless
    we should fix that :-D
    Rob Norris
    @tpolecat
    hm that's all just for sbt itself
    at least up at the top .. but yeah now i see scalaz_stream etc