first commit

scripts/gluster_check.sh

@@ -0,0 +1,92 @@
+#!/bin/bash
+#Roy Cohen 					:roy@wondercohen.nl
+#objective					:Gluster checks for Nagios
+#First line of code			:15/01/2019
+#last update				:17/01/2019
+#version					:0.1
+#status						:Not PRD ready
+
+#Some general vars
+#Server IP adress in the storage network TODO: osg servers
+STORAGE_IP=$(ip -4 addr show dev storage | grep -oP '(?<=inet\s)\d+(\.\d+){3}')
+#number of active bricks on server acourding to gluster
+NUMBER_OF_ACTIVE_BRICKS=$(gluster volume info|grep $STORAGE_IP| wc -l)
+#name of active volumes on server acourding to gluster
+GLUSTER_ACTIVE_VOLUM_NAMES=$(gluster volume info|grep $STORAGE_IP | cut -d "/" -f3)
+#numer of glusterfsd (brick daemon) running processes
+GLUSTER_BRICK_NUMBER_PID=$(pidof glusterfsd |wc -w)
+#glusterd pid status
+GLUSTERD_STAT=$(pidof glusterd &>/dev/null)
+#glusterfsd (brick daemon) pid status
+GLUSTER_BRICK_STAT=$(pidof glusterfsd &>/dev/null)
+
+######START OF SCRIPT#######
+
+check_if_glusterd_is_running() {
+#check if glusterd is running
+if ! $GLUSTERD_STAT &>/dev/null; then
+                echo "CRITICAL: glusterd management daemon not running"
+        else
+                echo "OK: glusterd management daemon is running"
+fi
+}
+
+check_if_glusterfsd_is_running() {
+# check for glusterfsd (brick daemon)
+if ! $GLUSTER_BRICK_STAT; then
+                echo "CRITICAL: glusterfsd brick daemon not running"
+        else
+                echo "OK: glusterfsd brick daemon is running"
+fi
+}
+
+check_if_a_brick_is_down() {
+# check number of active bricks
+if  [ "$NUMBER_OF_ACTIVE_BRICKS" == "$GLUSTER_BRICK_NUMBER_PID" ] ; then
+                echo "OK: all bricks are active"
+        else
+                echo "CRITICAL: There is one or more bricks down"
+fi
+}
+
+check_heal_status_of_a_volume() {
+# get volume heal status and redirect it to a tmp file
+for volume in $(gluster volume info|grep $STORAGE_IP | cut -d "/" -f3)
+do
+	echo "volumes, $volume" 
+	gluster v heal $volume info |grep entries|cut -d ":" -f2
+done >  /tmp/gluster_monitoring 
+
+OUT=$(cat /tmp/gluster_monitoring|grep -v volumes |sed "s/^[ \t]*//" |grep -v ^0 > /dev/null 2>&1)
+if [ "$?" -eq 0 ]; then
+		echo "WARNING:  volumes healing at this moment $(cat /tmp/gluster_monitoring |sed "s/^[ \t]*//" | grep -v ^0)"
+	else
+		echo "OK: no volumes healing at this moment"
+fi
+echo "" > /tmp/gluster_monitoring
+}
+
+
+
+#### CALL THE FUNCTIONS ######
+check_if_glusterd_is_running
+check_if_glusterfsd_is_running
+check_if_a_brick_is_down
+check_heal_status_of_a_volume
+
+
+# some logs test
+#cat /var/log/glusterfs/glusterd.log|grep "E" |grep "failed"
+#  oc rsh -n glusterfs $(oc get pods -n glusterfs | grep Running | grep 'glusterfs-storage' | head -n 1 | cut -d' ' -f1) gluster volume status all
+
+if []
+
+
+
+OUT=$(cat /tmp/gluster_monitoring|grep -v volumes |sed "s/^[ \t]*//" |grep -v ^0 > /dev/null 2>&1)
+if [ "$?" -eq 0 ]; then
+                echo "WARNING:  volumes healing at this moment $(cat /tmp/gluster_monitoring |sed "s/^[ \t]*//" | grep -v ^0)"
+        else
+                echo "OK: no volumes healing at this moment"
+fi
+echo "" > /tmp/gluster_monitoring