| 1 | # you can disable an alarm notification by setting the 'to' line to: silent |
| 2 | |
| 3 | # --- Availability --- |
| 4 | |
| 5 | # AMBA Sev4: DatapathAvailability < 90% (30m window) |
| 6 | # NAT Gateway datapath availability below 100% means the service is degraded. |
| 7 | # Below 90% indicates significant packet loss through the gateway. |
| 8 | |
| 9 | template: am_nat_gateway_datapath_availability |
| 10 | on: azure_monitor.nat_gateway.datapath_availability |
| 11 | class: Availability |
| 12 | type: Other |
| 13 | component: Azure NAT Gateway |
| 14 | lookup: average -5m unaligned of average |
| 15 | units: percentage |
| 16 | every: 1m |
| 17 | warn: $this < (($status >= $WARNING) ? (99) : (95)) |
| 18 | crit: $this < (($status == $CRITICAL) ? (95) : (90)) |
| 19 | delay: down 5m multiplier 1.5 max 1h |
| 20 | summary: NAT Gateway availability on ${label:resource_name} |
| 21 | info: Datapath availability of NAT Gateway ${label:resource_name} \ |
| 22 | in ${label:resource_group} (${label:region}). \ |
| 23 | Values below 100% indicate packet loss through the gateway. |
| 24 | to: sysadmin |
| 25 | |
| 26 | # --- Errors --- |
| 27 | |
| 28 | # AMBA Sev1 (Critical): PacketDropCount > 5 packets/s (5m window) |
| 29 | # Dropped packets indicate SNAT port exhaustion, flow timeouts, or |
| 30 | # exceeding the connection limit of the NAT gateway. |
| 31 | |
| 32 | template: am_nat_gateway_dropped_packets |
| 33 | on: azure_monitor.nat_gateway.dropped_packets |
| 34 | class: Errors |
| 35 | type: Other |
| 36 | component: Azure NAT Gateway |
| 37 | lookup: average -5m unaligned of total |
| 38 | units: packets/s |
| 39 | every: 1m |
| 40 | warn: $this > (($status >= $WARNING) ? (0) : (5)) |
| 41 | crit: $this > (($status == $CRITICAL) ? (5) : (50)) |
| 42 | delay: down 5m multiplier 1.5 max 1h |
| 43 | summary: NAT Gateway packet drops on ${label:resource_name} |
| 44 | info: Packets being dropped by NAT Gateway ${label:resource_name} \ |
| 45 | in ${label:resource_group} (${label:region}). \ |
| 46 | Drops indicate SNAT port exhaustion or connection limit reached. |
| 47 | to: sysadmin |
| 48 | |
| 49 | # --- Connections --- |
| 50 | |
| 51 | # Azure NAT Gateway supports up to 2 million total SNAT connections. |
| 52 | # Azure recommends alerting at 80% (1.6M). We warn at 60% (1.2M). |
| 53 | |
| 54 | template: am_nat_gateway_snat_connections |
| 55 | on: azure_monitor.nat_gateway.snat_connections |
| 56 | class: Utilization |
| 57 | type: Other |
| 58 | component: Azure NAT Gateway |
| 59 | lookup: average -5m unaligned of total |
| 60 | units: connections |
| 61 | every: 1m |
| 62 | warn: $this > (($status >= $WARNING) ? (1000000) : (1200000)) |
| 63 | crit: $this > (($status == $CRITICAL) ? (1200000) : (1600000)) |
| 64 | delay: down 5m multiplier 1.5 max 1h |
| 65 | summary: NAT Gateway SNAT connections on ${label:resource_name} |
| 66 | info: SNAT connection count on NAT Gateway ${label:resource_name} \ |
| 67 | in ${label:resource_group} (${label:region}). \ |
| 68 | Azure limit is 2 million connections per gateway. |
| 69 | to: sysadmin |
| 70 | |
| 71 | template: am_nat_gateway_total_connections |
| 72 | on: azure_monitor.nat_gateway.total_connections |
| 73 | class: Utilization |
| 74 | type: Other |
| 75 | component: Azure NAT Gateway |
| 76 | lookup: average -5m unaligned of total |
| 77 | units: connections |
| 78 | every: 1m |
| 79 | warn: $this > (($status >= $WARNING) ? (1000000) : (1200000)) |
| 80 | crit: $this > (($status == $CRITICAL) ? (1200000) : (1600000)) |
| 81 | delay: down 5m multiplier 1.5 max 1h |
| 82 | summary: NAT Gateway total connections on ${label:resource_name} |
| 83 | info: Total SNAT connection count on NAT Gateway ${label:resource_name} \ |
| 84 | in ${label:resource_group} (${label:region}). \ |
| 85 | Azure limit is 2 million connections per gateway. |
| 86 | to: sysadmin |