Prevent elasticsearch from using swap

[mrdavidlaing]

Although we set bootstrap.mlockall: true in the elasticsearch config, it doesn't appear to be working, since I get the following WARN in the elasticsearch node logs:

[2015-09-14 06:33:51,181][WARN ][bootstrap                ] Unable to lock JVM memory (ENOMEM). This can result in part of the JVM being swapped out. Increase RLIMIT_MEMLOCK (ulimit).

and curl http://127.0.0.1:9200/_nodes/process?pretty | grep mlockall gives:

        "mlockall" : false
        "mlockall" : false
        "mlockall" : false
        "mlockall" : false

https://www.elastic.co/guide/en/elasticsearch/reference/current/setup-configuration.html#setup-configuration-memory seems to contain some guidance as to what is going wrong

[dpb587]

I think the best first step would be adding ulimit -l unlimited to the control scripts, similar to how we raise the file limit.

It's not a good idea to disable swap unless you have a really, really good understanding of what's happening on the VM and I don't think that's something we should try to generalize in our release. That's more something to be handled by a colocated, generic utility release responsible for configuring sysctl, /proc/sys, and other low level settings.