add redis

defaults/main.yml

@@ -45,3 +45,14 @@ nrpe_mysql_longqueries_critical: 1200
 
 nrpe_proc_age_warning: 400
 nrpe_proc_age_critical: 600
+
+nrpe_redis_memory_warning: 80
+nrpe_redis_memory_critical: 90
+nrpe_redis_connected_clients_warning: 200
+nrpe_redis_connected_clients_critical: 500
+nrpe_redis_hitrate_warning: 80
+nrpe_redis_hitrate_critical: 50
+nrpe_redis_fragments_warning: '1.5'
+nrpe_redis_fragments_critical: '2.0'
+nrpe_redis_replication_lag_warning: 10
+nrpe_redis_replication_lag_critical: 60

files/nrpe/check_redis_health

@@ -0,0 +1,211 @@
+#!/bin/bash
+#
+# Nagios plugin to check for redis health metrics.
+#
+# Copyright (c) 2026, GitHub Copilot
+#
+# This script is free software: you can redistribute it and/or modify
+# it under the terms of the GNU General Public License as published by
+# the Free Software Foundation, either version 3 of the License, or
+# (at your option) any later version.
+#
+# This script is distributed in the hope that it will be useful,
+# but WITHOUT ANY WARRANTY; without even the implied warranty of
+# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+# GNU General Public License for more details.
+#
+# You should have received a copy of the GNU General Public License
+# along with this program. If not, see <https://www.gnu.org/licenses/>.
+
+# Nagios exit codes
+STATE_OK=0
+STATE_WARNING=1
+STATE_CRITICAL=2
+STATE_UNKNOWN=3
+
+# --- Valeurs par défaut ---
+CHECK_TYPE=""
+WARN=0
+CRIT=0
+AUTH=""
+
+# --- Aide ---
+usage() {
+    cat << EOF
+Usage: $0 -x [type] -w <warning> -c <critical> [-a <password>]
+
+Types de check (-x):
+  ping        : Test de connexion simple (PONG)
+  memory      : Utilisation mémoire en % (Ex: -w 80 -c 90)
+  frag        : Ratio de fragmentation (Ex: -w 1.5 -c 2.0)
+  hitrate     : Taux d'efficacité du cache en % (Ex: -w 80 -c 50)
+  persistence : État des sauvegardes RDB/AOF
+  replication : Statut Master/Slave et lag (Ex: -w 10 -c 60)
+  clients     : Nombre de clients connectés (Ex: -w 200 -c 500)
+
+Optionnelle:
+  -a : Mot de passe Redis (sera passé via REDISCLI_AUTH)
+EOF
+    exit $STATE_UNKNOWN
+}
+
+# --- Parsing des arguments ---
+while getopts "x:w:c:a:?" opt; do
+    case "$opt" in
+        x) CHECK_TYPE="$OPTARG" ;;
+        w) WARN="$OPTARG" ;;
+        c) CRIT="$OPTARG" ;;
+        a) AUTH="$OPTARG" ;;
+        *) usage ;;
+    esac
+done
+
+if [ -z "$CHECK_TYPE" ]; then usage; fi
+
+# Export du mot de passe pour redis-cli
+if [ -n "$AUTH" ]; then
+    export REDISCLI_AUTH="$AUTH"
+fi
+
+EXIT_CODE=$STATE_OK
+INSTANCES_FOUND=0
+
+# Listes pour stocker les résultats
+CRIT_LIST=""
+WARN_LIST=""
+OK_LIST=""
+
+# 1. Détection des ports
+if command -v ss &> /dev/null; then
+    PORTS=$(ss -tlnp | grep 'redis-server' | awk '{print $4}' | awk -F: '{print $NF}' | sort -u)
+elif command -v netstat &> /dev/null; then
+    PORTS=$(netstat -tlnp | grep 'redis-server' | awk '{print $4}' | awk -F: '{print $NF}' | sort -u)
+else
+    PORTS=$(ps -ef | grep 'redis-server' | grep -v grep | grep -oP '(?<=:)\d{4,5}' | sort -u)
+fi
+
+if [ -z "$PORTS" ]; then
+    echo "OK: Aucun Redis détecté"
+    exit $STATE_OK
+fi
+
+# 2. Exécution du check
+for PORT in $PORTS; do
+    # On ignore les ports < 1024 sauf le 6379 standard
+    if [ "$PORT" -lt 1024 ] && [ "$PORT" -ne 6379 ]; then continue; fi
+
+    # Vérification si l'instance répond
+    if ! redis-cli -p "$PORT" PING 2>/dev/null | grep -q "PONG"; then
+        MSG="Port $PORT (Erreur Connexion/Auth)"
+        if [ -z "$CRIT_LIST" ]; then CRIT_LIST="$MSG"; else CRIT_LIST="$CRIT_LIST, $MSG"; fi
+        [ "$EXIT_CODE" -lt $STATE_CRITICAL ] && EXIT_CODE=$STATE_CRITICAL
+        ((INSTANCES_FOUND++))
+        continue
+    fi
+
+    ((INSTANCES_FOUND++))
+    CURR_STATUS=$STATE_OK
+    VAL_STR=""
+
+    case "$CHECK_TYPE" in
+        ping)
+            VAL_STR="PONG"
+            ;;
+        memory)
+            MEM_INFO=$(redis-cli -p "$PORT" info memory 2>/dev/null)
+            USED=$(echo "$MEM_INFO" | grep "used_memory:" | cut -d: -f2 | tr -d '\r')
+            MAX=$(echo "$MEM_INFO" | grep "maxmemory:" | cut -d: -f2 | tr -d '\r')
+            USED_H=$(echo "$MEM_INFO" | grep "used_memory_human:" | cut -d: -f2 | tr -d '\r')
+            MAX_H=$(echo "$MEM_INFO" | grep "maxmemory_human:" | cut -d: -f2 | tr -d '\r')
+            if [ "$MAX" -gt 0 ]; then
+                VALUE=$(( USED * 100 / MAX ))
+                VAL_STR="${VALUE}% (Used:${USED_H}, Max:${MAX_H})"
+                if [ "$VALUE" -ge "$CRIT" ]; then CURR_STATUS=$STATE_CRITICAL; elif [ "$VALUE" -ge "$WARN" ]; then CURR_STATUS=$STATE_WARNING; fi
+            else
+                VAL_STR="NoLimit (Used:${USED_H})"
+            fi
+            ;;
+        frag)
+            FRAG=$(redis-cli -p "$PORT" info memory 2>/dev/null | grep "mem_fragmentation_ratio:" | cut -d: -f2 | tr -d '\r')
+            IS_CRIT=$(echo "$FRAG >= $CRIT" | bc -l 2>/dev/null)
+            IS_WARN=$(echo "$FRAG >= $WARN" | bc -l 2>/dev/null)
+            VAL_STR="Ratio:$FRAG"
+            if [ "$IS_CRIT" == "1" ]; then CURR_STATUS=$STATE_CRITICAL; elif [ "$IS_WARN" == "1" ]; then CURR_STATUS=$STATE_WARNING; fi
+            ;;
+        hitrate)
+            STATS=$(redis-cli -p "$PORT" info stats 2>/dev/null)
+            HITS=$(echo "$STATS" | grep "keyspace_hits:" | cut -d: -f2 | tr -d '\r')
+            MISSES=$(echo "$STATS" | grep "keyspace_misses:" | cut -d: -f2 | tr -d '\r')
+            TOTAL=$((HITS + MISSES))
+            VALUE=100
+            [ "$TOTAL" -gt 0 ] && VALUE=$(( HITS * 100 / TOTAL ))
+            VAL_STR="${VALUE}% (Hits:$HITS, Misses:$MISSES)"
+            if [ "$VALUE" -le "$CRIT" ]; then CURR_STATUS=$STATE_CRITICAL; elif [ "$VALUE" -le "$WARN" ]; then CURR_STATUS=$STATE_WARNING; fi
+            ;;
+        persistence)
+            PERS=$(redis-cli -p "$PORT" info persistence 2>/dev/null)
+            RDB=$(echo "$PERS" | grep "rdb_last_bgsave_status:" | cut -d: -f2 | tr -d '\r')
+            AOF=$(echo "$PERS" | grep "aof_last_write_status:" | cut -d: -f2 | tr -d '\r')
+            VAL_STR="RDB:$RDB, AOF:$AOF"
+            if [ "$RDB" != "ok" ] || [ "$AOF" != "ok" ]; then CURR_STATUS=$STATE_CRITICAL; fi
+            ;;
+        replication)
+            REPL=$(redis-cli -p "$PORT" info replication 2>/dev/null)
+            ROLE=$(echo "$REPL" | grep "role:" | cut -d: -f2 | tr -d '\r')
+            if [ "$ROLE" == "slave" ]; then
+                LINK=$(echo "$REPL" | grep "master_link_status:" | cut -d: -f2 | tr -d '\r')
+                LAG=$(echo "$REPL" | grep "master_last_io_seconds_ago:" | cut -d: -f2 | tr -d '\r')
+                VAL_STR="Slave, Link:$LINK, Lag:${LAG}s"
+                if [ "$LINK" != "up" ] || [ "$LAG" -ge "$CRIT" ]; then CURR_STATUS=$STATE_CRITICAL; elif [ "$LAG" -ge "$WARN" ]; then CURR_STATUS=$STATE_WARNING; fi
+            else
+                SLAVE_COUNT=$(echo "$REPL" | grep "connected_slaves:" | cut -d: -f2 | tr -d '\r')
+                VAL_STR="Master, Slaves:$SLAVE_COUNT"
+            fi
+            ;;
+        clients)
+            CLIENTS=$(redis-cli -p "$PORT" info clients 2>/dev/null | grep "connected_clients:" | cut -d: -f2 | tr -d '\r')
+            MAX_CLIENTS=$(redis-cli -p "$PORT" config get maxclients 2>/dev/null | tail -n1)
+            VAL_STR="$CLIENTS clients"
+            if [ -n "$MAX_CLIENTS" ]; then VAL_STR="$VAL_STR (Max:$MAX_CLIENTS)"; fi
+            if [ "$CLIENTS" -ge "$CRIT" ]; then CURR_STATUS=$STATE_CRITICAL; elif [ "$CLIENTS" -ge "$WARN" ]; then CURR_STATUS=$STATE_WARNING; fi
+            ;;
+        *) usage ;;
+    esac
+
+    # Stockage du résultat
+    MSG="Port $PORT ($VAL_STR)"
+    case "$CURR_STATUS" in
+        $STATE_CRITICAL) 
+            if [ -z "$CRIT_LIST" ]; then CRIT_LIST="$MSG"; else CRIT_LIST="$CRIT_LIST, $MSG"; fi
+            [ "$EXIT_CODE" -lt $STATE_CRITICAL ] && EXIT_CODE=$STATE_CRITICAL 
+            ;;
+        $STATE_WARNING) 
+            if [ -z "$WARN_LIST" ]; then WARN_LIST="$MSG"; else WARN_LIST="$WARN_LIST, $MSG"; fi
+            [ "$EXIT_CODE" -lt $STATE_WARNING ] && EXIT_CODE=$STATE_WARNING 
+            ;;
+        $STATE_OK)
+            if [ -z "$OK_LIST" ]; then OK_LIST="$MSG"; else OK_LIST="$OK_LIST, $MSG"; fi
+            ;;
+    esac
+done
+
+# Construction du message de sortie
+if [ "$EXIT_CODE" -eq $STATE_OK ]; then
+    echo "OK: All $INSTANCES_FOUND instance(s) are healthy ($CHECK_TYPE). $OK_LIST"
+    exit $STATE_OK
+fi
+
+FINAL_MSG=""
+if [ "$EXIT_CODE" -eq $STATE_CRITICAL ]; then
+    FINAL_MSG="CRITICAL: "
+    if [ -n "$CRIT_LIST" ]; then FINAL_MSG="${FINAL_MSG}${CRIT_LIST}"; fi
+    if [ -n "$WARN_LIST" ]; then 
+        if [ -n "$CRIT_LIST" ]; then FINAL_MSG="${FINAL_MSG} - "; fi
+        FINAL_MSG="${FINAL_MSG}WARNING: ${WARN_LIST}"
+    fi
+elif [ "$EXIT_CODE" -eq $STATE_WARNING ]; then
+    FINAL_MSG="WARNING: $WARN_LIST"
+fi
+
+echo "$FINAL_MSG"
+exit "$EXIT_CODE"

templates/nrpe.j2

@@ -103,4 +103,23 @@ command[check_k8s_pv_pvc]=/usr/bin/sudo /usr/lib/nagios/plugins/check_k8s_pv_pvc
 command[check_k8s_replicasets]=/usr/bin/sudo /usr/lib/nagios/plugins/check_k8s_replicasets
 command[check_k8s_pod_restarts]=/usr/bin/sudo /usr/lib/nagios/plugins/check_k8s_pod_restarts
 {% endif %}
+{% endif %}
+
+{% if nrpe_redis_memory_warning is defined or nrpe_redis_memory_critical is defined or nrpe_redis_persistence is defined %}
+# redis
+command[check_redis_health]=/usr/bin/sudo /usr/lib/nagios/plugins/check_redis_health -x ping
+command[check_redis_memory]=/usr/bin/sudo /usr/lib/nagios/plugins/check_redis_health -x memory -w {{ nrpe_redis_memory_warning }} -c {{ nrpe_redis_memory_critical }}
+command[check_redis_persistence]=/usr/bin/sudo /usr/lib/nagios/plugins/check_redis_health -x persistence
+{% if nrpe_redis_connected_clients_warning is defined or nrpe_redis_connected_clients_critical is defined %}
+command[check_redis_health_clients]=/usr/bin/sudo /usr/lib/nagios/plugins/check_redis_health -x clients -w {{ nrpe_redis_connected_clients_warning }} -c {{ nrpe_redis_connected_clients_critical }}
+{% endif %}
+{% if nrpe_redis_hitrate_warning is defined or nrpe_redis_hitrate_critical is defined %}
+command[check_redis_health_hitrate]=/usr/bin/sudo /usr/lib/nagios/plugins/check_redis_health -x hitrate -w {{ nrpe_redis_hitrate_warning }} -c {{ nrpe_redis_hitrate_critical }}
+{% endif %}
+{% if nrpe_redis_fragments_warning is defined or nrpe_redis_fragments_critical is defined %}
+command[check_redis_health_frag]=/usr/bin/sudo /usr/lib/nagios/plugins/check_redis_health -x frag -w {{ nrpe_redis_fragments_warning }} -c {{ nrpe_redis_fragments_critical }}
+{% endif %}
+{% if nrpe_redis_replication_lag_warning is defined or nrpe_redis_replication_lag_critical is defined %}
+command[check_redis_health_replication]=/usr/bin/sudo /usr/lib/nagios/plugins/check_redis_health -x replication -w {{ nrpe_redis_replication_lag_warning }} -c {{ nrpe_redis_replication_lag_critical }}
+{% endif %}
 {% endif %}

templates/nrpe.sudoers.j2

@@ -11,4 +11,5 @@ nagios ALL=(ALL) NOPASSWD: /usr/lib/nagios/plugins/check_k8s_jobs_cronjobs
 nagios ALL=(ALL) NOPASSWD: /usr/lib/nagios/plugins/check_k8s_pki_certs
 nagios ALL=(ALL) NOPASSWD: /usr/lib/nagios/plugins/check_k8s_pv_pvc
 nagios ALL=(ALL) NOPASSWD: /usr/lib/nagios/plugins/check_k8s_replicasets
-nagios ALL=(ALL) NOPASSWD: /usr/lib/nagios/plugins/check_k8s_pod_restarts
+nagios ALL=(ALL) NOPASSWD: /usr/lib/nagios/plugins/check_k8s_pod_restarts
+nagios ALL=(ALL) NOPASSWD: /usr/lib/nagios/plugins/check_redis_health