Update JNA to latest version

Updating to this version allows to configure a special JNA directory,
in case the /tmp directory is mounted with the noexec option, as JNA
extracts some data and tries to execute parts of it.

Also updated documentation to clarify mlockall and memory settings as well
as pointing to the new jna.tmpdir system property.

Closes #5493

docs/reference/setup/configuration.asciidoc

@@ -52,21 +52,36 @@ curl localhost:9200/_nodes/process?pretty
 [[setup-configuration-memory]]
 ==== Memory Settings
 
-There is an option to use
+The Linux kernel tries to use as much memory as possible for file system
+caches, and eagerly swaps out unused application memory, possibly resulting
+in the elasticsearch process being swapped out and suffer from low
+performance then.
+
+The first option is to ensure that the sysctl value `vm.swappiness` is set
+to `0`. This reduces the kernel's tendency to swap and should not lead to
+swapping under normal circumstances, while still allowing the whole system
+to swap in emergency conditions.
+
+In the above case there is an option to use
 http://opengroup.org/onlinepubs/007908799/xsh/mlockall.html[mlockall] to
-try to lock the process address space so it won't be swapped. For this
-to work, the `bootstrap.mlockall` should be set to `true` and it is
-recommended to set both the min and max memory allocation to be the
-same. Note: This option is only available on Linux/Unix operating
-systems.
+try to lock the process address space so it won't be swapped by configuring
+the `bootstrap.mlockall` setting to `true`.
+Note: This option is only available on Linux/Unix operating systems.
 
-In order to see if this works or not, set the `common.jna` logging to
-DEBUG level. A solution to "Unknown mlockall error 0" can be to set
-`ulimit -l unlimited`.
+After starting Elasticsearch can check the process part of the nodes
+info API to find out if the `mlockall` setting has been applied successfully.
 
-Note, `mlockall` might cause the JVM or shell
-session to exit if it fails to allocate the memory (because not enough
-memory is available on the machine).
+If the setting has not been applied, you can try to execute
+`ulimit -l unlimited` to allow memory locking.
+
+Another possible reason why `mlockall` can fail is when the directory pointed
+to by the `java.io.tmpdir` JVM system property - typically `/tmp` - is mounted
+with the `noexec` option. In this case you can specify an additional directory
+using the `jna.tmpdir` system property to use for loading the native library.
+
+Note, that `mlockall` might cause the JVM or shell session to exit if it fails
+to allocate the memory, possibly due to not enough memory being available on
+the machine.
 
 [float]
 [[settings]]

pom.xml

@@ -282,7 +282,7 @@
         <dependency>
             <groupId>net.java.dev.jna</groupId>
             <artifactId>jna</artifactId>
-            <version>3.3.0</version>
+            <version>4.1.0</version>
             <scope>compile</scope>
             <optional>true</optional>
         </dependency>

src/main/java/org/elasticsearch/common/jna/CLibrary.java

@@ -40,9 +40,9 @@ public class CLibrary {
         try {
             Native.register("c");
         } catch (NoClassDefFoundError e) {
-            logger.warn("jna not found. native methods (mlockall) will be disabled.");
+            logger.warn("JNA not found. native methods (mlockall) will be disabled.");
         } catch (UnsatisfiedLinkError e) {
-            logger.debug("unable to link C library. native methods (mlockall) will be disabled.");
+            logger.warn("unable to link C library. native methods (mlockall) will be disabled.");
         }
     }