I have different jvms installed on my servers, and as the services running on them have been 'validated' for production for certain versions I had to install the jre8 to get reaper to work. This is no issue but the problem is that the service runs with the default java binary, which in my case is the 7 and I intend to keep it that way.
I also have for security reasons /tmp which is not executable.
Thus the service did not run by default and I had to add 2 lines in the /usr/local/bin/cassandra-reaper file to specify JAVA_HOME (and change the call to java to $JAVA_HOME/bin/java) and also had to add the -Djava.io.tmpdir flag to specify where it should go
I would expect the init.d script to incorporate default mechanism like sourcing a /etc/default/cassandra-reaper file where I would put env variables (like the JVM_OPTS variable from your cassandra-reaper script) and the JAVA_HOME path for example
┆Issue is synchronized with this Jira Story by Unito
┆Issue Number: REAP-178
Project board link
I have different jvms installed on my servers, and as the services running on them have been 'validated' for production for certain versions I had to install the jre8 to get reaper to work. This is no issue but the problem is that the service runs with the default java binary, which in my case is the 7 and I intend to keep it that way.
I also have for security reasons /tmp which is not executable.
Thus the service did not run by default and I had to add 2 lines in the /usr/local/bin/cassandra-reaper file to specify JAVA_HOME (and change the call to java to $JAVA_HOME/bin/java) and also had to add the -Djava.io.tmpdir flag to specify where it should go
I would expect the init.d script to incorporate default mechanism like sourcing a /etc/default/cassandra-reaper file where I would put env variables (like the JVM_OPTS variable from your cassandra-reaper script) and the JAVA_HOME path for example
┆Issue is synchronized with this Jira Story by Unito ┆Issue Number: REAP-178