check_jvm

I had to modify the script to add this line to make it work under ncpa_passive:

(( EUID != PUID )) && exec sudo -u "#$PUID" $0 -n $NAME -p $PROPERTY -w $WARNING -c $CRITICAL

You have to add it before this line:

[ "${PUID}" = "${EUID}" ] || p_unknown "JVM is running with different username, run this script with UID $PUID"

You still need to give nagios sudo access to run the plugin as per the README and add or update this line in the "[plugin directives]" section of ncpa.cfg:

run_with_sudo = check_jvm

Hello,
seems to not works with G1GC mode !? No error (exit code still 0, no output console), any idea to workaround or check ?

I've changed my process command line from :
-Xms24576m
-Xmx24576m
-XX:NewRatio=3
-XX:SurvivorRatio=4
-XX:TargetSurvivorRatio=90
-XX:MaxTenuringThreshold=8
-XX:+UseConcMarkSweepGC
-XX:+UseParNewGC
-XX:ConcGCThreads=4
-XX:ParallelGCThreads=4
-XX:+CMSScavengeBeforeRemark
-XX:PretenureSizeThreshold=64m
-XX:+UseCMSInitiatingOccupancyOnly
-XX:CMSInitiatingOccupancyFraction=50
-XX:CMSMaxAbortablePrecleanTime=6000
-XX:+CMSParallelRemarkEnabled
-XX:+ParallelRefProcEnabled
-verbose:gc
-XX:+PrintHeapAtGC
-XX:+PrintGCDetails
-XX:+PrintGCDateStamps
-XX:+PrintGCTimeStamps
-XX:+PrintTenuringDistribution
-XX:+PrintGCApplicationStoppedTime
-Xloggc:/var/log/cloudera-hdp-solr/solr_gc.log
-XX:+UseGCLogFileRotation
-XX:NumberOfGCLogFiles=9
-XX:GCLogFileSize=20M
-Dcom.sun.management.jmxremote
-Dcom.sun.management.jmxremote.local.only=false
-Dcom.sun.management.jmxremote.ssl=false
-Dcom.sun.management.jmxremote.authenticate=false
-Dcom.sun.management.jmxremote.port=18983
-Dcom.sun.management.jmxremote.rmi.port=18983
...

to
-server
-Xms20480m
-Xmx20480m
-XX:+UseG1GC
-XX:+PerfDisableSharedMem
-XX:+ParallelRefProcEnabled
-XX:G1HeapRegionSize=8m
-XX:MaxGCPauseMillis=250
-XX:InitiatingHeapOccupancyPercent=75
-XX:+UseLargePages
-XX:+AggressiveOpts
-verbose:gc
-XX:+PrintHeapAtGC
-XX:+PrintGCDetails
-XX:+PrintGCDateStamps
-XX:+PrintGCTimeStamps
-XX:+PrintTenuringDistribution
-XX:+PrintGCApplicationStoppedTime
-Xloggc:/var/log/cloudera-hdp-solr/solr_gc.log
-XX:+UseGCLogFileRotation
-XX:NumberOfGCLogFiles=9
-XX:GCLogFileSize=20M
-Dcom.sun.management.jmxremote
-Dcom.sun.management.jmxremote.local.only=false
-Dcom.sun.management.jmxremote.ssl=false
-Dcom.sun.management.jmxremote.authenticate=false
-Dcom.sun.management.jmxremote.port=18983
-Dcom.sun.management.jmxremote.rmi.port=18983

Thanks
Olivier

Hello, Nice work.
But it seems that since Java 9 this plugin is no longer working, because it uses a class that was in the tools.jar library (sun.tools.attach.LinuxAttachProvider) .
And tools.jar is no longer available since Java 9.
Am I wrong ?
And has anybody got a workaround for this problem ?
Have a nice day!
A.

Hi,

I am using the plugin to monitor threads and sessions as below but the Nagios core server is unable to read the output. Running these check manually (as commands) on the client host is working though.

command[check_tomcat_threads]=/usr/bin/sudo -u tomcat7 /usr/local/nagios/libexec/check_jvm -n org.apache.catalina.startup.Bootstrap -p threads -w 105 -c 135
command[check_tomcat_conn]=/usr/bin/sudo -u tomcat7 /usr/local/nagios/libexec/check_jvm -n org.apache.catalina.startup.Bootstrap -p sessions -w 280 -c 360

Locally running the command:
# /usr/bin/sudo -u tomcat7 /usr/local/nagios/libexec/check_jvm -n org.apache.catalina.startup.Bootstrap -p threads -w 105 -c 135
CRITICAL 153 |threads=153;;;

Any suggestions?

Thanks.

Hi,

So happy to find this check.
I add -k optional parameter to check multiple JVM on the same machine :

...
Usage: $SCRIPT_NAME -n|--name -k -p|--property -w|--warning -c|--critical

Where: is one of: "heap|non-heap|threads|classes"
Hint: You can use "jps -l" or "java -jar JvmInspector.jar all" to get the java name
Example: $SCRIPT_NAME -n org.apache.catalina.startup.Bootstrap -p heap -w 1073741824 -c 2147483648

EOF
exit 0
}

[ "$#" -eq 0 ] && usage

while [ ! -z "$1" ]; do
case $1 in
-n|--name) shift; NAME="$1";;
-k|--keyword) shift; KEYWORD="$1";;
-p|--property) shift; PROPERTY="$1";;
-w|--warning) shift; WARNING="$1";;
-c|--critical) shift; CRITICAL="$1";;
-h|--help) usage;;
esac
shift
done

[ -z "$NAME" ] && p_unknown "Missing JVM app class name, use -n "
[ -z "$PROPERTY" ] && p_unknown "Missing property, use -p "
[ -z "$WARNING" ] && p_unknown "Missing warning thresholds, use -w "
[ -z "$CRITICAL" ] && p_unknown "Missing critical thresholds, use -c "

expr ${WARNING} : '[0-9]+$' >/dev/null || p_unknown "Invalid warning threshold"
expr ${CRITICAL} : '[0-9]+$' >/dev/null || p_unknown "Invalid critical threshold"
[ -f "$JVMINSPECTOR" ] || p_unknown "Can't find JvmInspector.jar, please install it and set JVMINSPECTOR var in this script"

if [ -z "$KEYWORD" ]; then
PSLINE="$(ps axo pid,uid,command | grep [j]ava | grep $NAME | head -1)"
else
PSLINE="$(ps axo pid,uid,command | grep [j]ava | grep $NAME | grep $KEYWORD)"
fi
...

So, now, you can use the script like this :

./check_jvm --name org.apache.catalina.startup.Bootstrap -k confluence -p heap -w 100 -c 200

./check_jvm --name org.apache.catalina.startup.Bootstrap -k jira -p heap -w 100 -c 200

Thanks a lot

Bye

Hello.

When I've JVM running as "tomcat" user, and run the script with root, I get:

UNKNOWN JVM is running with different username, run this script with UID 111

That is a problem to run with Nagios.

Thanks!

Awesome concept. How does this work on servers that host multiple JVMs running under the same uid? Do you need to specify the PID?