| 1 | # you can disable an alarm notification by setting the 'to' line to: silent |
| 2 | |
| 3 | # --- Run Failure Rate --- |
| 4 | |
| 5 | # Run failure percentage (0-100%). High is bad. |
| 6 | # AMBA: RunFailurePercentage > 0 is Sev1 |
| 7 | |
| 8 | template: am_logic_apps_run_failure_rate |
| 9 | on: azure_monitor.logic_apps.run_failure_rate |
| 10 | class: Errors |
| 11 | type: Other |
| 12 | component: Logic Apps |
| 13 | lookup: average -5m unaligned of failure_rate |
| 14 | units: percentage |
| 15 | every: 1m |
| 16 | warn: $this > (($status >= $WARNING) ? (1) : (5)) |
| 17 | crit: $this > (($status == $CRITICAL) ? (5) : (10)) |
| 18 | delay: down 5m multiplier 1.5 max 1h |
| 19 | summary: Logic Apps run failure rate on ${label:resource_name} |
| 20 | info: Percentage of workflow runs failing on Logic App ${label:resource_name} \ |
| 21 | in ${label:resource_group} (${label:region}) |
| 22 | to: sysadmin |
| 23 | |
| 24 | # --- Run Failures (absolute count) --- |
| 25 | |
| 26 | # Failed runs rate. Any sustained failures need attention. |
| 27 | |
| 28 | template: am_logic_apps_runs_failed |
| 29 | on: azure_monitor.logic_apps.run_lifecycle |
| 30 | class: Errors |
| 31 | type: Other |
| 32 | component: Logic Apps |
| 33 | lookup: average -5m unaligned of failed |
| 34 | units: runs/s |
| 35 | every: 1m |
| 36 | warn: $this > (($status >= $WARNING) ? (0) : (1)) |
| 37 | crit: $this > (($status == $CRITICAL) ? (1) : (5)) |
| 38 | delay: down 5m multiplier 1.5 max 1h |
| 39 | summary: Logic Apps failed runs on ${label:resource_name} |
| 40 | info: Rate of failed workflow runs on Logic App ${label:resource_name} \ |
| 41 | in ${label:resource_group} (${label:region}) |
| 42 | to: sysadmin |
| 43 | |
| 44 | # --- Run Latency --- |
| 45 | |
| 46 | # Overall run latency (seconds). High latency indicates slow workflows. |
| 47 | |
| 48 | template: am_logic_apps_run_latency |
| 49 | on: azure_monitor.logic_apps.run_latency |
| 50 | class: Latency |
| 51 | type: Other |
| 52 | component: Logic Apps |
| 53 | lookup: average -5m unaligned of all |
| 54 | units: seconds |
| 55 | every: 1m |
| 56 | warn: $this > (($status >= $WARNING) ? (30) : (60)) |
| 57 | crit: $this > (($status == $CRITICAL) ? (60) : (120)) |
| 58 | delay: down 5m multiplier 1.5 max 1h |
| 59 | summary: Logic Apps run latency on ${label:resource_name} |
| 60 | info: Average workflow run latency on Logic App ${label:resource_name} \ |
| 61 | in ${label:resource_group} (${label:region}) |
| 62 | to: sysadmin |
| 63 | |
| 64 | # --- Run Throttling --- |
| 65 | |
| 66 | # Throttled run events indicate the workflow is hitting |
| 67 | # Azure rate limits. |
| 68 | |
| 69 | template: am_logic_apps_run_throttled |
| 70 | on: azure_monitor.logic_apps.run_throttling |
| 71 | class: Errors |
| 72 | type: Other |
| 73 | component: Logic Apps |
| 74 | lookup: average -5m unaligned |
| 75 | units: events/s |
| 76 | every: 1m |
| 77 | warn: $this > (($status >= $WARNING) ? (0) : (1)) |
| 78 | delay: down 5m multiplier 1.5 max 1h |
| 79 | summary: Logic Apps run throttling on ${label:resource_name} |
| 80 | info: Rate of throttled run events on Logic App ${label:resource_name} \ |
| 81 | in ${label:resource_group} (${label:region}). \ |
| 82 | Throttling indicates Azure rate limits are being hit |
| 83 | to: sysadmin |
| 84 | |
| 85 | # --- Action Failures --- |
| 86 | |
| 87 | # Failed actions rate. Failing actions cause workflow failures. |
| 88 | |
| 89 | template: am_logic_apps_actions_failed |
| 90 | on: azure_monitor.logic_apps.action_lifecycle |
| 91 | class: Errors |
| 92 | type: Other |
| 93 | component: Logic Apps |
| 94 | lookup: average -5m unaligned of failed |
| 95 | units: actions/s |
| 96 | every: 1m |
| 97 | warn: $this > (($status >= $WARNING) ? (0) : (1)) |
| 98 | crit: $this > (($status == $CRITICAL) ? (1) : (5)) |
| 99 | delay: down 5m multiplier 1.5 max 1h |
| 100 | summary: Logic Apps failed actions on ${label:resource_name} |
| 101 | info: Rate of failed actions on Logic App ${label:resource_name} \ |
| 102 | in ${label:resource_group} (${label:region}) |
| 103 | to: sysadmin |
| 104 | |
| 105 | # --- Action Latency --- |
| 106 | |
| 107 | # Average action latency. High values indicate slow connectors |
| 108 | # or backend dependencies. |
| 109 | |
| 110 | template: am_logic_apps_action_latency |
| 111 | on: azure_monitor.logic_apps.action_latency |
| 112 | class: Latency |
| 113 | type: Other |
| 114 | component: Logic Apps |
| 115 | lookup: average -5m unaligned of all |
| 116 | units: seconds |
| 117 | every: 1m |
| 118 | warn: $this > (($status >= $WARNING) ? (15) : (30)) |
| 119 | crit: $this > (($status == $CRITICAL) ? (30) : (60)) |
| 120 | delay: down 5m multiplier 1.5 max 1h |
| 121 | summary: Logic Apps action latency on ${label:resource_name} |
| 122 | info: Average action latency on Logic App ${label:resource_name} \ |
| 123 | in ${label:resource_group} (${label:region}) |
| 124 | to: sysadmin |
| 125 | |
| 126 | # --- Action Throttling --- |
| 127 | |
| 128 | # Throttled action events indicate connectors or actions |
| 129 | # are hitting Azure rate limits. |
| 130 | |
| 131 | template: am_logic_apps_action_throttled |
| 132 | on: azure_monitor.logic_apps.action_throttling |
| 133 | class: Errors |
| 134 | type: Other |
| 135 | component: Logic Apps |
| 136 | lookup: average -5m unaligned of total |
| 137 | units: events/s |
| 138 | every: 1m |
| 139 | warn: $this > (($status >= $WARNING) ? (0) : (1)) |
| 140 | delay: down 5m multiplier 1.5 max 1h |
| 141 | summary: Logic Apps action throttling on ${label:resource_name} |
| 142 | info: Rate of throttled action events on Logic App ${label:resource_name} \ |
| 143 | in ${label:resource_group} (${label:region}). \ |
| 144 | Throttling indicates connector rate limits are being hit |
| 145 | to: sysadmin |
| 146 | |
| 147 | # --- Trigger Failures --- |
| 148 | |
| 149 | # Failed triggers prevent workflows from starting. |
| 150 | |
| 151 | template: am_logic_apps_triggers_failed |
| 152 | on: azure_monitor.logic_apps.trigger_lifecycle |
| 153 | class: Errors |
| 154 | type: Other |
| 155 | component: Logic Apps |
| 156 | lookup: average -5m unaligned of failed |
| 157 | units: triggers/s |
| 158 | every: 1m |
| 159 | warn: $this > (($status >= $WARNING) ? (0) : (1)) |
| 160 | crit: $this > (($status == $CRITICAL) ? (1) : (5)) |
| 161 | delay: down 5m multiplier 1.5 max 1h |
| 162 | summary: Logic Apps failed triggers on ${label:resource_name} |
| 163 | info: Rate of failed triggers on Logic App ${label:resource_name} \ |
| 164 | in ${label:resource_group} (${label:region}). \ |
| 165 | Failed triggers prevent workflows from executing |
| 166 | to: sysadmin |
| 167 | |
| 168 | # --- Trigger Latency --- |
| 169 | |
| 170 | # Overall trigger latency. High values indicate slow polling |
| 171 | # or webhook delays. |
| 172 | |
| 173 | template: am_logic_apps_trigger_latency |
| 174 | on: azure_monitor.logic_apps.trigger_latency |
| 175 | class: Latency |
| 176 | type: Other |
| 177 | component: Logic Apps |
| 178 | lookup: average -5m unaligned of all |
| 179 | units: seconds |
| 180 | every: 1m |
| 181 | warn: $this > (($status >= $WARNING) ? (15) : (30)) |
| 182 | crit: $this > (($status == $CRITICAL) ? (30) : (60)) |
| 183 | delay: down 5m multiplier 1.5 max 1h |
| 184 | summary: Logic Apps trigger latency on ${label:resource_name} |
| 185 | info: Average trigger latency on Logic App ${label:resource_name} \ |
| 186 | in ${label:resource_group} (${label:region}) |
| 187 | to: sysadmin |
| 188 | |
| 189 | # --- Trigger Throttling --- |
| 190 | |
| 191 | # Throttled trigger events indicate polling or webhook triggers |
| 192 | # are hitting Azure rate limits. |
| 193 | |
| 194 | template: am_logic_apps_trigger_throttled |
| 195 | on: azure_monitor.logic_apps.trigger_throttling |
| 196 | class: Errors |
| 197 | type: Other |
| 198 | component: Logic Apps |
| 199 | lookup: average -5m unaligned of total |
| 200 | units: events/s |
| 201 | every: 1m |
| 202 | warn: $this > (($status >= $WARNING) ? (0) : (1)) |
| 203 | delay: down 5m multiplier 1.5 max 1h |
| 204 | summary: Logic Apps trigger throttling on ${label:resource_name} |
| 205 | info: Rate of throttled trigger events on Logic App ${label:resource_name} \ |
| 206 | in ${label:resource_group} (${label:region}). \ |
| 207 | Throttling indicates trigger rate limits are being hit |
| 208 | to: sysadmin |
| 209 | |
| 210 | # --- AI Agent Token Overflow --- |
| 211 | |
| 212 | # Completion token overflow events indicate AI agent responses |
| 213 | # are exceeding token limits (AI-enabled Logic Apps only). |
| 214 | |
| 215 | template: am_logic_apps_completion_token_overflow |
| 216 | on: azure_monitor.logic_apps.agent |
| 217 | class: Errors |
| 218 | type: Other |
| 219 | component: Logic Apps |
| 220 | lookup: average -5m unaligned of completion_overflow |
| 221 | units: events/s |
| 222 | every: 1m |
| 223 | warn: $this != nan AND $this > (($status >= $WARNING) ? (0) : (1)) |
| 224 | delay: down 5m multiplier 1.5 max 1h |
| 225 | summary: Logic Apps completion token overflow on ${label:resource_name} |
| 226 | info: Rate of completion token overflow events on Logic App ${label:resource_name} \ |
| 227 | in ${label:resource_group} (${label:region}). \ |
| 228 | Indicates AI agent responses are exceeding token limits |
| 229 | to: sysadmin |
| 230 | |
| 231 | # Prompt token overflow events indicate AI agent prompts |
| 232 | # are exceeding token limits (AI-enabled Logic Apps only). |
| 233 | |
| 234 | template: am_logic_apps_prompt_token_overflow |
| 235 | on: azure_monitor.logic_apps.agent |
| 236 | class: Errors |
| 237 | type: Other |
| 238 | component: Logic Apps |
| 239 | lookup: average -5m unaligned of prompt_overflow |
| 240 | units: events/s |
| 241 | every: 1m |
| 242 | warn: $this != nan AND $this > (($status >= $WARNING) ? (0) : (1)) |
| 243 | delay: down 5m multiplier 1.5 max 1h |
| 244 | summary: Logic Apps prompt token overflow on ${label:resource_name} |
| 245 | info: Rate of prompt token overflow events on Logic App ${label:resource_name} \ |
| 246 | in ${label:resource_group} (${label:region}). \ |
| 247 | Indicates AI agent prompts are exceeding token limits |
| 248 | to: sysadmin |