Nothing so complicated... we are seeing mesos kill off our executors immediately. When I reroute logging to an NFS directory we have available, the executors survive fine. As such I am wondering if the spark workers are getting killed by mesos for exceeding their disk quota (which atm is 0). This could be a red herring, however.
2015-04-13 15:41 GMT-04:00 Patrick Wendell <pwend...@gmail.com>: > Hey Jonathan, > > Are you referring to disk space used for storing persisted RDD's? For > that, Spark does not bound the amount of data persisted to disk. It's > a similar story to how Spark's shuffle disk output works (and also > Hadoop and other frameworks make this assumption as well for their > shuffle data, AFAIK). > > We could (in theory) add a storage level that bounds the amount of > data persisted to disk and forces re-computation if the partition did > not fit. I'd be interested to hear more about a workload where that's > relevant though, before going that route. Maybe if people are using > SSD's that would make sense. > > - Patrick > > On Mon, Apr 13, 2015 at 8:19 AM, Jonathan Coveney <jcove...@gmail.com> > wrote: > > I'm surprised that I haven't been able to find this via google, but I > > haven't... > > > > What is the setting that requests some amount of disk space for the > > executors? Maybe I'm misunderstanding how this is configured... > > > > Thanks for any help! >
