Yeah, in general, make sure you use exactly the same “cluster URL” string shown on the master’s web UI. There’s currently a limitation in Akka where different ways of specifying the hostname won’t work.
Matei On Dec 6, 2013, at 10:54 AM, Nathan Kronenfeld <[email protected]> wrote: > Never mind, I figured it out - apparently it was different DNS resolutions > locally and within the cluster; when I use the IP address instead of the > machine name in MASTER, it all seems to work. > > > On Fri, Dec 6, 2013 at 1:38 PM, Nathan Kronenfeld > <[email protected]> wrote: > Hi, all. > > I'm trying to connect to a remote cluster from my machine, using spark 0.7.3. > In conf/spark-env.sh, I've set MASTER, SCALA_HOME, SPARK_MASTER_IP, and > SPARK_MASTER_PORT. > > When I try to run a job, it starts, but never gets anywhere, and I keep > getting the following error message: > > 13/12/06 13:37:20 WARN cluster.ClusterScheduler: Initial job has not accepted > any resources; check your cluster UI to ensure that workers are registered > > I look at the cluster UI in a browser, and it says it has 8 workers > registered, all alive. > > What does this error mean? I assume I'm missing something in the setup - > does anyone know what? > > Thanks in advance, > -Nathan Kronenfeld > > > > > > -- > Nathan Kronenfeld > Senior Visualization Developer > Oculus Info Inc > 2 Berkeley Street, Suite 600, > Toronto, Ontario M5A 4J5 > Phone: +1-416-203-3003 x 238 > Email: [email protected]
