feat - add ceph check

README.md

@@ -117,6 +117,7 @@ The following checks are deployed to `/usr/lib/nagios/plugins/` (or configured p
 | `nrpe_redis_fragments_critical` | `2.0` | `check_redis_health` | Critical threshold for fragmentation ratio. |
 | `nrpe_redis_replication_lag_warning` | `10` | `check_redis_health` | Warning threshold for replication lag (seconds). |
 | `nrpe_redis_replication_lag_critical` | `60` | `check_redis_health` | Critical threshold for replication lag (seconds). |
+| `nrpe_ceph` | false | `check_ceph` | Ceph server monitoring | 
 
 ## Example Playbooks
 

files/nrpe/check_ceph

@@ -0,0 +1,104 @@
+#!/bin/bash
+
+# Nagios Exit Codes
+OK=0
+WARNING=1
+CRITICAL=2
+UNKNOWN=3
+
+if [ -z "$1" ]; then
+    echo "Usage: $0 [health|osd|mon|cap|pg]"
+    exit $UNKNOWN
+fi
+
+COMMAND=$1
+
+case $COMMAND in
+    health)
+        HEALTH_DATA=$(sudo ceph health)
+        if [[ $HEALTH_DATA == *"HEALTH_OK"* ]]; then
+            echo "OK: Ceph is healthy"
+            exit $OK
+        elif [[ $HEALTH_DATA == *"HEALTH_WARN"* ]]; then
+            echo "WARNING: $HEALTH_DATA"
+            exit $WARNING
+        else
+            echo "CRITICAL: $HEALTH_DATA"
+            exit $CRITICAL
+        fi
+        ;;
+
+    osd)
+        OSD_JSON=$(sudo ceph osd stat --format json)
+        TOTAL=$(echo "$OSD_JSON" | python3 -c "import sys, json; print(json.load(sys.stdin).get('num_osds', 0))")
+        UP=$(echo "$OSD_JSON" | python3 -c "import sys, json; print(json.load(sys.stdin).get('num_up_osds', 0))")
+        IN=$(echo "$OSD_JSON" | python3 -c "import sys, json; print(json.load(sys.stdin).get('num_in_osds', 0))")
+
+        if [ "$UP" -lt "$TOTAL" ]; then
+            echo "CRITICAL: OSD down! ($UP/$TOTAL up) | osds=$TOTAL up=$UP in=$IN"
+            exit $CRITICAL
+        elif [ "$IN" -lt "$TOTAL" ]; then
+            echo "WARNING: OSD out! ($IN/$TOTAL in) | osds=$TOTAL up=$UP in=$IN"
+            exit $WARNING
+        else
+            echo "OK: All OSDs are up and in ($TOTAL/$TOTAL) | osds=$TOTAL up=$UP in=$IN"
+            exit $OK
+        fi
+        ;;
+
+    mon)
+        MON_JSON=$(sudo ceph mon stat --format json)
+        QUORUM_COUNT=$(echo "$MON_JSON" | python3 -c "import sys, json; print(len(json.load(sys.stdin).get('quorum', [])))")
+
+        if [ "$QUORUM_COUNT" -lt 3 ]; then
+            echo "CRITICAL: Monitor Quorum degraded! ($QUORUM_COUNT/3) | quorum=$QUORUM_COUNT"
+            exit $CRITICAL
+        else
+            echo "OK: Monitor Quorum is 3/3 | quorum=$QUORUM_COUNT"
+            exit $OK
+        fi
+        ;;
+
+    cap)
+        CAP_JSON=$(sudo ceph df --format json)
+        USAGE=$(echo "$CAP_JSON" | python3 -c "
+import sys, json
+try:
+    stats = json.load(sys.stdin)['stats']
+    ratio = stats.get('total_used_raw_ratio', stats.get('total_used_ratio', 0))
+    print(int(float(ratio) * 100))
+except Exception:
+    print(-1)
+")
+
+        if [ "$USAGE" -eq -1 ]; then
+            echo "UNKNOWN: Unable to parse Ceph capacity"
+            exit $UNKNOWN
+        elif [ "$USAGE" -ge 85 ]; then
+            echo "CRITICAL: Ceph capacity at $USAGE% | usage=$USAGE%"
+            exit $CRITICAL
+        elif [ "$USAGE" -ge 75 ]; then
+            echo "WARNING: Ceph capacity at $USAGE% | usage=$USAGE%"
+            exit $WARNING
+        else
+            echo "OK: Ceph capacity at $USAGE% | usage=$USAGE%"
+            exit $OK
+        fi
+        ;;
+
+    pg)
+        PG_STAT=$(sudo ceph pg stat)
+        if [[ $PG_STAT == *"active+clean"* ]] && [[ $PG_STAT != *"degraded"* ]] && [[ $PG_STAT != *"undersized"* ]]; then
+            echo "OK: PGs are active+clean"
+            exit $OK
+        else
+            echo "WARNING: PGs status: $PG_STAT"
+            exit $WARNING
+        fi
+        ;;
+
+    *)
+        echo "Unknown argument: $COMMAND"
+        exit $UNKNOWN
+        ;;
+esac

templates/nrpe.j2

@@ -145,3 +145,12 @@ command[check_nvme_temperature]=/usr/bin/sudo /usr/lib/nagios/plugins/check_nvme
 # zpool
 command[check_zpool_health]=/usr/bin/sudo /usr/lib/nagios/plugins/check_zpool_health {{ nrpe_zpool_name }}
 {% endif %}
+
+{% if nrpe_ceph is defined %}
+# ceph
+command[check_ceph_health]=/usr/lib/nagios/plugins/check_ceph health
+command[check_ceph_osd]=/usr/lib/nagios/plugins/check_ceph osd
+command[check_ceph_mon]=/usr/lib/nagios/plugins/check_ceph mon
+command[check_ceph_cap]=/usr/lib/nagios/plugins/check_ceph cap
+command[check_ceph_pg]=/usr/lib/nagios/plugins/check_ceph pg
+{% endif %}

templates/nrpe.sudoers.j2

@@ -36,3 +36,11 @@ nagios ALL=(ALL) NOPASSWD: /usr/lib/nagios/plugins/check_pvesr
 nagios ALL=(ALL) NOPASSWD: /usr/lib/nagios/plugins/check_nvme_smart -d {{ nrpe_nvme_device }} -w {{ nrpe_nvme_smart_warning }} -c {{ nrpe_nvme_smart_critical }}
 nagios ALL=(ALL) NOPASSWD: /usr/lib/nagios/plugins/check_nvme_temperature -d {{ nrpe_nvme_device }} -w {{ nrpe_nvme_temperature_warning }} -c {{ nrpe_nvme_temperature_critical }}
 {% endif %}
+
+{% if nrpe_ceph is defined %}
+nagios ALL=(ALL) NOPASSWD: /usr/bin/ceph health
+nagios ALL=(ALL) NOPASSWD: /usr/bin/ceph osd stat --format json
+nagios ALL=(ALL) NOPASSWD: /usr/bin/ceph mon stat --format json
+nagios ALL=(ALL) NOPASSWD: /usr/bin/ceph df --format json
+nagios ALL=(ALL) NOPASSWD: /usr/bin/ceph pg stat --format json
+{% endif %}