You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
I have an issue where I am trying to manually use kafka-mesos on a DCOS cluster. I am "Trying" to avoid installing anything on the node, thus I am using the Java Mesosphere bundles in /opt/mesosphere.
I can get the scheduler to run by using PATH:$PATH $JAVA_HOME/bin ./kafka-mesos.sh scheduler this works fine.
However, when my brokers try to run, I get
sh: 1: java: not found
So, ideally, if I could pass an ENV or update the PATH, or do something to allow me to pass environment information to the executors (like Spark does) that would be awesome.
The text was updated successfully, but these errors were encountered:
I have an issue where I am trying to manually use kafka-mesos on a DCOS cluster. I am "Trying" to avoid installing anything on the node, thus I am using the Java Mesosphere bundles in /opt/mesosphere.
I can get the scheduler to run by using PATH:$PATH $JAVA_HOME/bin ./kafka-mesos.sh scheduler this works fine.
However, when my brokers try to run, I get
sh: 1: java: not found
So, ideally, if I could pass an ENV or update the PATH, or do something to allow me to pass environment information to the executors (like Spark does) that would be awesome.
The text was updated successfully, but these errors were encountered: