| 1 | # you can disable an alarm notification by setting the 'to' line to: silent |
| 2 | |
| 3 | # --- Origin Health --- |
| 4 | |
| 5 | template: am_front_door_origin_health |
| 6 | on: azure_monitor.front_door.origin_health |
| 7 | class: Availability |
| 8 | type: Web Server |
| 9 | component: Azure Front Door |
| 10 | lookup: average -5m unaligned of health |
| 11 | units: percentage |
| 12 | every: 1m |
| 13 | warn: $this < (($status >= $WARNING) ? (99) : (95)) |
| 14 | crit: $this < (($status == $CRITICAL) ? (95) : (90)) |
| 15 | delay: down 5m multiplier 1.5 max 1h |
| 16 | summary: Front Door origin health on ${label:resource_name} |
| 17 | info: Origin health percentage for Azure Front Door ${label:resource_name} \ |
| 18 | in ${label:resource_group} (${label:region}). \ |
| 19 | Low origin health indicates backends are failing health probes |
| 20 | to: sysadmin |
| 21 | |
| 22 | # --- Latency --- |
| 23 | |
| 24 | template: am_front_door_total_latency |
| 25 | on: azure_monitor.front_door.latency |
| 26 | class: Latency |
| 27 | type: Web Server |
| 28 | component: Azure Front Door |
| 29 | lookup: average -5m unaligned of total |
| 30 | units: milliseconds |
| 31 | every: 1m |
| 32 | warn: $this > (($status >= $WARNING) ? (3000) : (4000)) |
| 33 | crit: $this > (($status == $CRITICAL) ? (5000) : (8000)) |
| 34 | delay: down 5m multiplier 1.5 max 1h |
| 35 | summary: Front Door total latency on ${label:resource_name} |
| 36 | info: Average total request latency (client to Front Door to origin and back) \ |
| 37 | for Azure Front Door ${label:resource_name} \ |
| 38 | in ${label:resource_group} (${label:region}) |
| 39 | to: sysadmin |
| 40 | |
| 41 | template: am_front_door_origin_latency |
| 42 | on: azure_monitor.front_door.latency |
| 43 | class: Latency |
| 44 | type: Web Server |
| 45 | component: Azure Front Door |
| 46 | lookup: average -5m unaligned of origin |
| 47 | units: milliseconds |
| 48 | every: 1m |
| 49 | warn: $this > (($status >= $WARNING) ? (3000) : (5000)) |
| 50 | crit: $this > (($status == $CRITICAL) ? (5000) : (10000)) |
| 51 | delay: down 5m multiplier 1.5 max 1h |
| 52 | summary: Front Door origin latency on ${label:resource_name} |
| 53 | info: Average origin response latency for Azure Front Door ${label:resource_name} \ |
| 54 | in ${label:resource_group} (${label:region}). \ |
| 55 | High origin latency indicates slow backends |
| 56 | to: sysadmin |
| 57 | |
| 58 | # --- Error Rates --- |
| 59 | |
| 60 | template: am_front_door_5xx_error_rate |
| 61 | on: azure_monitor.front_door.error_rate |
| 62 | class: Errors |
| 63 | type: Web Server |
| 64 | component: Azure Front Door |
| 65 | lookup: average -5m unaligned of 5xx |
| 66 | units: percentage |
| 67 | every: 1m |
| 68 | warn: $this > (($status >= $WARNING) ? (3) : (5)) |
| 69 | crit: $this > (($status == $CRITICAL) ? (5) : (10)) |
| 70 | delay: down 5m multiplier 1.5 max 1h |
| 71 | summary: Front Door 5xx error rate on ${label:resource_name} |
| 72 | info: Percentage of requests resulting in 5xx server errors \ |
| 73 | for Azure Front Door ${label:resource_name} \ |
| 74 | in ${label:resource_group} (${label:region}) |
| 75 | to: sysadmin |
| 76 | |
| 77 | template: am_front_door_4xx_error_rate |
| 78 | on: azure_monitor.front_door.error_rate |
| 79 | class: Errors |
| 80 | type: Web Server |
| 81 | component: Azure Front Door |
| 82 | lookup: average -5m unaligned of 4xx |
| 83 | units: percentage |
| 84 | every: 1m |
| 85 | warn: $this > (($status >= $WARNING) ? (15) : (25)) |
| 86 | delay: down 5m multiplier 1.5 max 1h |
| 87 | summary: Front Door 4xx error rate on ${label:resource_name} |
| 88 | info: Percentage of requests resulting in 4xx client errors \ |
| 89 | for Azure Front Door ${label:resource_name} \ |
| 90 | in ${label:resource_group} (${label:region}). \ |
| 91 | Sustained high 4xx rates may indicate misconfigured routing or abusive clients |
| 92 | to: sysadmin |
| 93 | |
| 94 | # --- Cache Performance --- |
| 95 | |
| 96 | template: am_front_door_byte_hit_ratio |
| 97 | on: azure_monitor.front_door.byte_hit_ratio |
| 98 | class: Utilization |
| 99 | type: Web Server |
| 100 | component: Azure Front Door |
| 101 | lookup: average -10m unaligned of hit_ratio |
| 102 | units: percentage |
| 103 | every: 1m |
| 104 | warn: $this != nan AND $this < (($status >= $WARNING) ? (50) : (40)) |
| 105 | delay: down 15m multiplier 1.5 max 1h |
| 106 | summary: Front Door cache hit ratio on ${label:resource_name} |
| 107 | info: Byte hit ratio (percentage of bytes served from cache) \ |
| 108 | for Azure Front Door ${label:resource_name} \ |
| 109 | in ${label:resource_group} (${label:region}). \ |
| 110 | Low cache hit ratio means most traffic goes to origin, increasing latency and origin load |
| 111 | to: sysadmin |
| 112 | |
| 113 | # --- WAF --- |
| 114 | |
| 115 | template: am_front_door_waf_rate_limited |
| 116 | on: azure_monitor.front_door.origin_shield_requests |
| 117 | class: Errors |
| 118 | type: Web Server |
| 119 | component: Azure Front Door |
| 120 | lookup: average -5m unaligned of rate_limited |
| 121 | units: requests/s |
| 122 | every: 1m |
| 123 | warn: $this != nan AND $this > (($status >= $WARNING) ? (50) : (100)) |
| 124 | delay: down 5m multiplier 1.5 max 1h |
| 125 | summary: Front Door origin shield rate limiting on ${label:resource_name} |
| 126 | info: Rate of origin shield requests being rate limited \ |
| 127 | for Azure Front Door ${label:resource_name} \ |
| 128 | in ${label:resource_group} (${label:region}) |
| 129 | to: sysadmin |