Re: spark shuffle service failing to start on slaves

Does change spark.shuffle.service.port to a unused port works for you? On Wed, Nov 4, 2015 at 2:56 AM, Rodrick Brown wrote: > My slave instance runs as user “mesos” and I have the following directive > set > > $ cat /etc/mesos-slave/switch_user > false > > When trying to start start-mesos-shuffl

Re: Cluster Maintanence

I thought there was a ticket around allowing operators to induce a kill task, but I can't seem to find it. There are some subtleties, for example kill tasks currently only manifest from a framework's request (so it's not clear what assumptions folks have made). Feel free to create a ticket for this

Re: Marathon 0.11.1 - Mesos 0.25 - Mesos-DNS 0.4.0

No, it wasn't specified at all. I was using an old config.json, thus I had to add the setting with the host listed first for it to work. Not sure why docker ended up being first in line there. On Tue, Nov 3, 2015 at 2:02 PM, James DeFelice wrote: > The default value of IPSources doesn't have `do

Re: mesos fetch uri when behind a squid proxy

It's possible that Spark sets the executor environment explicitly, which would lead to the http_proxy and https_proxy environment variables not being passed along to the executor. You could try using the `--executor_environment_variables` command-line flag when running the agent to specify these en

Re: Marathon 0.11.1 - Mesos 0.25 - Mesos-DNS 0.4.0

The default value of IPSources doesn't have `docker` listed. As long as that's not in the list you shouldn't have had a problem, unless some bad actor was writing the wrong labels into the task. I don't see support for NetworkInfos (`netinfos`) in marathon yet. Which means that `host` should have b

spark shuffle service failing to start on slaves

My slave instance runs as user “mesos” and I have the following directive set $ cat /etc/mesos-slave/switch_user false When trying to start start-mesos-shuffle-service.sh process I I’m getting the following execution Spark Command: /usr/java/jdk1.8.0_60/jre/bin/java -cp /opt/spark-1.5.0-bin-h

Remove top level CommandInfo::ContainerInfo protobuf message

Hello, Is anybody using CommandInfo::ContainerInfo protobuf message from Mesos 0.19? It was deprecated in Mesos 0.20 in favor of the top-level ContainerInfo message used as a field in TaskInfo/ExecutorInfo. As per MESOS-2315 we’d like to remove

Re: [Breaking Change 0.26] Credential secret change

Hello everyone, Small reminder that this change was committed and will be part of 0.26. A pending patch to add a section to upgrades.md is here: https://reviews.apache.org/r/39908/ @Niklas: There will not be an intermediate phase where both formats will work. On Fri, Oct 9, 2015 at 10:05 AM, Ni

Re: Marathon 0.11.1 - Mesos 0.25 - Mesos-DNS 0.4.0

I used "IPSources": ["host", "netinfo", "mesos"] With the thought that I would preference for the host at this point. When network isolation works in Marathon, then I will likely switch to netinfo. On Mon, Nov 2, 2015 at 7:28 PM, James DeFelice wrote: > What settings worked for you? We did ai