Merge lp:~chris-gondolin/charms/precise/nrpe-external-master/trunk into lp:charms/nrpe-external-master

[Chris Stratford]

Added a conntrack usage check to warn when nf_conntrack gets too near its limits (always returns OK if conntrack isn't in use on the unit)

[Charles Butler (lazypower)]

Thank you for the submission Chris,

I took some time to review your changes and I have the following notes:

The charm as it stands today does not pass charm proof. there is a a missing category in the metadata.yaml, and no icon.svg - while this is not related to your change, could you possibly add those to the update to assist in future reviews?

I see that you added 2 configuration options: conntrack, and conntrackservice group - however there were no documentation updates to cover these configuration options. We request that any updates to charm configuration be accompanied by documentation edits to cover new items being added to the charm, and if this effects deployment, a deployment example (or scenario) where these configuration options would be used. I'm going to superimpose a temporarly blocker pending the README update.

The code quality is good, and I appreciate the cheetah templates being used. This is a great feature merge for the nrpe-external-master charm.

Thanks again for the submission. I'm going to change status of this MP to "needs work" and when you're ready for another review please click the "Request another review" button in the upper right hand corner of the commit message.

If you have any questions/comments/concerns about the review contact us in #juju on irc.freenode.net or email the mailing list <email address hidden>, or ask a question tagged with "juju" on http://askubuntu.com.

review: Needs Fixing

Unmerged revisions

32. By Chris Stratford on 2014-07-21

[chriss] Add conntrack usage check for RT#70035

=== modified file 'config.yaml'
--- config.yaml	2014-05-27 10:36:57 +0000
+++ config.yaml	2014-07-21 11:22:52 +0000
@@ -60,6 +60,14 @@
     default: "-w 90% -c 75%"
     type: string
     description: Swap check
+  conntrack:
+    default:  "-w 80 -c 90"
+    type: string
+    description: Conntrack check
+  conntrack_servicegroup:
+    default: "prompt-critical"
+    type: string
+    description: Conntrack nagios servicegroup
   hostgroups:
     default: ""
     type: string

=== added file 'files/check_conntrack.sh'
--- files/check_conntrack.sh	1970-01-01 00:00:00 +0000
+++ files/check_conntrack.sh	2014-07-21 11:22:52 +0000
@@ -0,0 +1,48 @@
+#!/bin/bash
+#
+#-------------------------------------------------------#
+# This file is Juju managed - do not make local changes #
+#-------------------------------------------------------#
+#
+# Check we're not close to conntrack_max limit
+
+set -u
+
+if [ ! -e /proc/sys/net/netfilter/nf_conntrack_max ]; then
+    echo "OK conntrack not in use"
+    exit 0
+fi
+
+max=$(cat /proc/sys/net/netfilter/nf_conntrack_max)
+current=$(cat /proc/sys/net/netfilter/nf_conntrack_count)
+warn=""
+crit=""
+
+while getopts "w:c:" opts; do
+    case "${opts}" in
+        w)
+            warn=${OPTARG}
+            ;;
+        c)
+            crit=${OPTARG}
+            ;;
+    esac
+done
+
+if [ -z "${warn}" -o -z "${crit}" ]; then
+    echo "Usage: check_conntrack -w warn -c crit"
+    exit 1
+fi
+
+usage=$(((100 * ${current}) / ${max}))
+
+if [ ${usage} -gt ${crit} ]; then
+    echo "CRITICAL conntrack at ${usage}%"
+    exit 2
+elif [ ${usage} -gt ${warn} ]; then
+    echo "WARNING conntrack at ${usage}%"
+    exit 1
+fi
+
+echo "OK conntrack at ${usage}%"
+exit 0

=== modified file 'hooks/config-changed'
--- hooks/config-changed	2014-05-27 10:36:57 +0000
+++ hooks/config-changed	2014-07-21 11:22:52 +0000
@@ -48,6 +48,7 @@
 LOAD=$(config-get load)
 USERS=$(config-get users)
 SWAP=$(config-get swap)
+CONNTRACK=$(config-get conntrack)
 
 if [[ "$PROCS" == "auto" ]]; then
     PROC_COUNT=$(nproc)
@@ -76,6 +77,9 @@
 echo "# Swap" > /etc/nagios/nrpe.d/check_swap.cfg
 echo "command[check_swap]=/usr/lib/nagios/plugins/check_swap ${SWAP}" >> /etc/nagios/nrpe.d/check_swap.cfg
 
+echo "# Conntrack usage" > /etc/nagios/nrpe.d/check_conntrack.cfg
+echo "command[check_conntrack]=/usr/local/lib/nagios/plugins/check_conntrack.sh ${CONNTRACK}" >> /etc/nagios/nrpe.d/check_conntrack.cfg
+
 #------------------------------------------------------
 # Host check
 #------------------------------------------------------
@@ -115,4 +119,11 @@
 # Service specific checks
 #------------------------------------------------------
 
+export DESCRIPTION="Check conntrack usage"
+export COMMAND="check_nrpe!check_conntrack"
+export SERVICEGROUPS=$(config-get conntrack_servicegroup)
+cheetah fill --env --oext compiled templates/service.cfg.tmpl
+rm -f /var/lib/nagios/export/service__conntrack.cfg
+cp templates/service.cfg.compiled /var/lib/nagios/export/service__${NAGIOS_HOSTNAME}_check_conntrack.cfg
+
 service nagios-nrpe-server reload

=== modified file 'hooks/install'
--- hooks/install	2014-03-11 13:57:58 +0000
+++ hooks/install	2014-07-21 11:22:52 +0000
@@ -14,6 +14,9 @@
 cp files/nagios_plugin.py /usr/lib/nagios/plugins/nagios_plugin.py
 ln -fs /usr/lib/nagios/plugins/nagios_plugin.py /usr/local/lib/nagios/plugins/nagios_plugin.py
 
+cp files/check_conntrack.sh /usr/lib/nagios/plugins/check_conntrack.sh
+ln -fs /usr/lib/nagios/plugins/check_conntrack.sh /usr/local/lib/nagios/plugins/check_conntrack.sh
+
 cp files/default_rsync /etc/default/rsync
 if [[ -d /etc/rsyncd.d ]]; then
     export RSYNC_MODULE_ONLY=1

=== added file 'templates/service.cfg.tmpl'
--- templates/service.cfg.tmpl	1970-01-01 00:00:00 +0000
+++ templates/service.cfg.tmpl	2014-07-21 11:22:52 +0000
@@ -0,0 +1,12 @@
+#---------------------------------------------------
+# This file is Juju managed
+#--------------------------------------------------
+
+define service {
+    use                     ${HOSTCHECK_INHERIT}
+    host_name               ${NAGIOS_HOSTNAME}
+    service_description     ${DESCRIPTION}
+    check_command           ${COMMAND}
+    servicegroups           ${SERVICEGROUPS}
+}
+