{ "leansignal_demand_export": "1.0", "exported_at": "2026-07-17T00:00:00Z", "source_tenant": "leansignal-catalog", "catalog": { "slug": "azure-monitor-otel-demand-extended", "integration": "azure-monitor", "variant": "extended", "title": "Azure Monitor — Extended", "category": "Cloud", "dashboards": 1, "panels": 25, "alerts": 12, "docs_url": "https://docs.leansignal.io/integrations/azure-monitor/" }, "demand": { "ref": "demand", "name": "Azure Monitor — Extended", "description": "Deep per-service diagnostics: B-series CPU credits and disk IOPS saturation, App Service CPU time, Azure SQL log-write and worker saturation, storage latency and ingress/egress, and Service Bus errors and throttling — plus the full alert set.", "dashboards": [ { "ref": "dash-azure-monitor", "name": "Azure Monitor — Extended", "description": "Azure Monitor metrics for Virtual Machines, App Service, Azure SQL, Storage accounts, and Service Bus — CPU, memory, 5xx rates, storage headroom, availability, latency, and message backlogs.", "data": { "display": { "name": "Azure Monitor — Extended" }, "panels": { "vm_cpu": { "kind": "Panel", "spec": { "display": { "name": "CPU utilization", "description": "Per-VM CPU as the hypervisor sees it — the first thing to check on a slow VM." }, "plugin": { "kind": "TimeSeriesChart", "spec": { "legend": { "position": "bottom", "mode": "table", "values": [ "last-number", "mean", "max" ] }, "yAxis": { "format": { "unit": "percent" }, "min": 0, "max": 100 } } }, "queries": [ { "kind": "TimeSeriesQuery", "spec": { "plugin": { "kind": "PrometheusTimeSeriesQuery", "spec": { "query": "avg by (resource_group, name) (azure_percentage_cpu_average{resource_group=~\"$instance\",type=\"Microsoft.Compute/virtualMachines\"})", "seriesNameFormat": "{{name}}" } } } } ] } }, "vm_available_memory": { "kind": "Panel", "spec": { "display": { "name": "Available memory", "description": "Memory still free on the VM. A steady decline ends in paging, then in the OOM killer." }, "plugin": { "kind": "TimeSeriesChart", "spec": { "legend": { "position": "bottom", "mode": "table", "values": [ "last-number", "mean", "max" ] }, "yAxis": { "format": { "unit": "bytes" }, "min": 0 } } }, "queries": [ { "kind": "TimeSeriesQuery", "spec": { "plugin": { "kind": "PrometheusTimeSeriesQuery", "spec": { "query": "min by (resource_group, name) (azure_available_memory_bytes_average{resource_group=~\"$instance\",type=\"Microsoft.Compute/virtualMachines\"})", "seriesNameFormat": "{{name}}" } } } } ] } }, "vm_network_io": { "kind": "Panel", "spec": { "display": { "name": "Network in / out (per period)", "description": "Bytes across the VM network interfaces per Azure Monitor period." }, "plugin": { "kind": "TimeSeriesChart", "spec": { "legend": { "position": "bottom", "mode": "table", "values": [ "last-number", "mean", "max" ] }, "yAxis": { "format": { "unit": "bytes" }, "min": 0 } } }, "queries": [ { "kind": "TimeSeriesQuery", "spec": { "plugin": { "kind": "PrometheusTimeSeriesQuery", "spec": { "query": "sum by (resource_group, name) (azure_network_in_total_total{resource_group=~\"$instance\",type=\"Microsoft.Compute/virtualMachines\"})", "seriesNameFormat": "{{name}} · in" } } } }, { "kind": "TimeSeriesQuery", "spec": { "plugin": { "kind": "PrometheusTimeSeriesQuery", "spec": { "query": "sum by (resource_group, name) (azure_network_out_total_total{resource_group=~\"$instance\",type=\"Microsoft.Compute/virtualMachines\"})", "seriesNameFormat": "{{name}} · out" } } } } ] } }, "vm_disk_io": { "kind": "Panel", "spec": { "display": { "name": "Disk read / write (per period)", "description": "Bytes read from and written to the VM disks per period." }, "plugin": { "kind": "TimeSeriesChart", "spec": { "legend": { "position": "bottom", "mode": "table", "values": [ "last-number", "mean", "max" ] }, "yAxis": { "format": { "unit": "bytes" }, "min": 0 } } }, "queries": [ { "kind": "TimeSeriesQuery", "spec": { "plugin": { "kind": "PrometheusTimeSeriesQuery", "spec": { "query": "sum by (resource_group, name) (azure_disk_read_bytes_total{resource_group=~\"$instance\",type=\"Microsoft.Compute/virtualMachines\"})", "seriesNameFormat": "{{name}} · read" } } } }, { "kind": "TimeSeriesQuery", "spec": { "plugin": { "kind": "PrometheusTimeSeriesQuery", "spec": { "query": "sum by (resource_group, name) (azure_disk_write_bytes_total{resource_group=~\"$instance\",type=\"Microsoft.Compute/virtualMachines\"})", "seriesNameFormat": "{{name}} · write" } } } } ] } }, "vm_cpu_credits": { "kind": "Panel", "spec": { "display": { "name": "CPU credits remaining (B-series)", "description": "Burstable VMs hard-throttle when this reaches zero — the classic \"VM suddenly slow\" cause. Non-B sizes emit nothing." }, "plugin": { "kind": "TimeSeriesChart", "spec": { "legend": { "position": "bottom", "mode": "table", "values": [ "last-number", "mean", "max" ] }, "yAxis": { "format": { "unit": "decimal" }, "min": 0 } } }, "queries": [ { "kind": "TimeSeriesQuery", "spec": { "plugin": { "kind": "PrometheusTimeSeriesQuery", "spec": { "query": "min by (resource_group, name) (azure_cpu_credits_remaining_average{resource_group=~\"$instance\",type=\"Microsoft.Compute/virtualMachines\"})", "seriesNameFormat": "{{name}}" } } } } ] } }, "vm_disk_iops_consumed": { "kind": "Panel", "spec": { "display": { "name": "Disk IOPS consumed", "description": "How much of the disk IOPS budget is in use. Sustained 100% means the disk, not the workload, sets the pace." }, "plugin": { "kind": "TimeSeriesChart", "spec": { "legend": { "position": "bottom", "mode": "table", "values": [ "last-number", "mean", "max" ] }, "yAxis": { "format": { "unit": "percent" }, "min": 0, "max": 100 } } }, "queries": [ { "kind": "TimeSeriesQuery", "spec": { "plugin": { "kind": "PrometheusTimeSeriesQuery", "spec": { "query": "max by (resource_group, name) (azure_os_disk_iops_consumed_percentage_average{resource_group=~\"$instance\",type=\"Microsoft.Compute/virtualMachines\"})", "seriesNameFormat": "{{name}} · os disk" } } } }, { "kind": "TimeSeriesQuery", "spec": { "plugin": { "kind": "PrometheusTimeSeriesQuery", "spec": { "query": "max by (resource_group, name) (azure_data_disk_iops_consumed_percentage_average{resource_group=~\"$instance\",type=\"Microsoft.Compute/virtualMachines\"})", "seriesNameFormat": "{{name}} · data disk" } } } } ] } }, "app_requests": { "kind": "Panel", "spec": { "display": { "name": "Requests (per period)", "description": "Requests served per app per Azure Monitor period." }, "plugin": { "kind": "TimeSeriesChart", "spec": { "legend": { "position": "bottom", "mode": "table", "values": [ "last-number", "mean", "max" ] }, "yAxis": { "format": { "unit": "decimal" }, "min": 0 } } }, "queries": [ { "kind": "TimeSeriesQuery", "spec": { "plugin": { "kind": "PrometheusTimeSeriesQuery", "spec": { "query": "sum by (resource_group, name) (azure_requests_total{resource_group=~\"$instance\",type=\"Microsoft.Web/sites\"})", "seriesNameFormat": "{{name}}" } } } } ] } }, "app_errors": { "kind": "Panel", "spec": { "display": { "name": "4xx / 5xx responses (per period)", "description": "Client errors beside server errors — 5xx is yours to fix, a 4xx spike usually is not." }, "plugin": { "kind": "TimeSeriesChart", "spec": { "legend": { "position": "bottom", "mode": "table", "values": [ "last-number", "mean", "max" ] }, "yAxis": { "format": { "unit": "decimal" }, "min": 0 } } }, "queries": [ { "kind": "TimeSeriesQuery", "spec": { "plugin": { "kind": "PrometheusTimeSeriesQuery", "spec": { "query": "sum by (resource_group, name) (azure_http5xx_total{resource_group=~\"$instance\",type=\"Microsoft.Web/sites\"})", "seriesNameFormat": "{{name}} · 5xx" } } } }, { "kind": "TimeSeriesQuery", "spec": { "plugin": { "kind": "PrometheusTimeSeriesQuery", "spec": { "query": "sum by (resource_group, name) (azure_http4xx_total{resource_group=~\"$instance\",type=\"Microsoft.Web/sites\"})", "seriesNameFormat": "{{name}} · 4xx" } } } } ] } }, "app_response_time": { "kind": "Panel", "spec": { "display": { "name": "Response time (avg)", "description": "Time to serve a request, averaged over the period." }, "plugin": { "kind": "TimeSeriesChart", "spec": { "legend": { "position": "bottom", "mode": "table", "values": [ "last-number", "mean", "max" ] }, "yAxis": { "format": { "unit": "seconds" }, "min": 0 } } }, "queries": [ { "kind": "TimeSeriesQuery", "spec": { "plugin": { "kind": "PrometheusTimeSeriesQuery", "spec": { "query": "max by (resource_group, name) (azure_httpresponsetime_average{resource_group=~\"$instance\",type=\"Microsoft.Web/sites\"})", "seriesNameFormat": "{{name}}" } } } } ] } }, "app_memory": { "kind": "Panel", "spec": { "display": { "name": "Memory working set", "description": "Memory the app holds — read against your App Service plan size; apps are recycled when the plan runs out." }, "plugin": { "kind": "TimeSeriesChart", "spec": { "legend": { "position": "bottom", "mode": "table", "values": [ "last-number", "mean", "max" ] }, "yAxis": { "format": { "unit": "bytes" }, "min": 0 } } }, "queries": [ { "kind": "TimeSeriesQuery", "spec": { "plugin": { "kind": "PrometheusTimeSeriesQuery", "spec": { "query": "max by (resource_group, name) (azure_memoryworkingset_average{resource_group=~\"$instance\",type=\"Microsoft.Web/sites\"})", "seriesNameFormat": "{{name}}" } } } } ] } }, "app_cpu_time": { "kind": "Panel", "spec": { "display": { "name": "CPU time (per period)", "description": "CPU seconds consumed per period — the metric the Free and Shared plan quotas are enforced on." }, "plugin": { "kind": "TimeSeriesChart", "spec": { "legend": { "position": "bottom", "mode": "table", "values": [ "last-number", "mean", "max" ] }, "yAxis": { "format": { "unit": "seconds" }, "min": 0 } } }, "queries": [ { "kind": "TimeSeriesQuery", "spec": { "plugin": { "kind": "PrometheusTimeSeriesQuery", "spec": { "query": "sum by (resource_group, name) (azure_cputime_total{resource_group=~\"$instance\",type=\"Microsoft.Web/sites\"})", "seriesNameFormat": "{{name}}" } } } } ] } }, "sql_cpu": { "kind": "Panel", "spec": { "display": { "name": "CPU utilization", "description": "Per-database CPU against its service tier — this is a fraction of the tier, not of a machine." }, "plugin": { "kind": "TimeSeriesChart", "spec": { "legend": { "position": "bottom", "mode": "table", "values": [ "last-number", "mean", "max" ] }, "yAxis": { "format": { "unit": "percent" }, "min": 0, "max": 100 } } }, "queries": [ { "kind": "TimeSeriesQuery", "spec": { "plugin": { "kind": "PrometheusTimeSeriesQuery", "spec": { "query": "avg by (resource_group, name) (azure_cpu_percent_average{resource_group=~\"$instance\",type=\"Microsoft.Sql/servers/databases\"})", "seriesNameFormat": "{{name}}" } } } } ] } }, "sql_storage": { "kind": "Panel", "spec": { "display": { "name": "Storage used", "description": "A database that fills its allocated storage stops accepting writes — watch the trend, not just the level." }, "plugin": { "kind": "TimeSeriesChart", "spec": { "legend": { "position": "bottom", "mode": "table", "values": [ "last-number", "mean", "max" ] }, "yAxis": { "format": { "unit": "percent" }, "min": 0, "max": 100 } } }, "queries": [ { "kind": "TimeSeriesQuery", "spec": { "plugin": { "kind": "PrometheusTimeSeriesQuery", "spec": { "query": "max by (resource_group, name) (azure_storage_percent_average{resource_group=~\"$instance\",type=\"Microsoft.Sql/servers/databases\"})", "seriesNameFormat": "{{name}}" } } } } ] } }, "sql_connections": { "kind": "Panel", "spec": { "display": { "name": "Connections (per period)", "description": "Successful and failed logins side by side — a failed-login spike is a credential, firewall, or limit problem." }, "plugin": { "kind": "TimeSeriesChart", "spec": { "legend": { "position": "bottom", "mode": "table", "values": [ "last-number", "mean", "max" ] }, "yAxis": { "format": { "unit": "decimal" }, "min": 0 } } }, "queries": [ { "kind": "TimeSeriesQuery", "spec": { "plugin": { "kind": "PrometheusTimeSeriesQuery", "spec": { "query": "sum by (resource_group, name) (azure_connection_successful_total{resource_group=~\"$instance\",type=\"Microsoft.Sql/servers/databases\"})", "seriesNameFormat": "{{name}} · successful" } } } }, { "kind": "TimeSeriesQuery", "spec": { "plugin": { "kind": "PrometheusTimeSeriesQuery", "spec": { "query": "sum by (resource_group, name) (azure_connection_failed_total{resource_group=~\"$instance\",type=\"Microsoft.Sql/servers/databases\"})", "seriesNameFormat": "{{name}} · failed" } } } } ] } }, "sql_deadlocks": { "kind": "Panel", "spec": { "display": { "name": "Deadlocks (per period)", "description": "Deadlocked transactions the engine had to kill — nonzero means application queries are fighting each other." }, "plugin": { "kind": "TimeSeriesChart", "spec": { "legend": { "position": "bottom", "mode": "table", "values": [ "last-number", "mean", "max" ] }, "yAxis": { "format": { "unit": "decimal" }, "min": 0 } } }, "queries": [ { "kind": "TimeSeriesQuery", "spec": { "plugin": { "kind": "PrometheusTimeSeriesQuery", "spec": { "query": "sum by (resource_group, name) (azure_deadlock_total{resource_group=~\"$instance\",type=\"Microsoft.Sql/servers/databases\"})", "seriesNameFormat": "{{name}}" } } } } ] } }, "sql_io": { "kind": "Panel", "spec": { "display": { "name": "Data I/O and log write", "description": "The two I/O budgets of the service tier. Log write pinned at 100% throttles every writer." }, "plugin": { "kind": "TimeSeriesChart", "spec": { "legend": { "position": "bottom", "mode": "table", "values": [ "last-number", "mean", "max" ] }, "yAxis": { "format": { "unit": "percent" }, "min": 0, "max": 100 } } }, "queries": [ { "kind": "TimeSeriesQuery", "spec": { "plugin": { "kind": "PrometheusTimeSeriesQuery", "spec": { "query": "max by (resource_group, name) (azure_physical_data_read_percent_average{resource_group=~\"$instance\",type=\"Microsoft.Sql/servers/databases\"})", "seriesNameFormat": "{{name}} · data I/O" } } } }, { "kind": "TimeSeriesQuery", "spec": { "plugin": { "kind": "PrometheusTimeSeriesQuery", "spec": { "query": "max by (resource_group, name) (azure_log_write_percent_average{resource_group=~\"$instance\",type=\"Microsoft.Sql/servers/databases\"})", "seriesNameFormat": "{{name}} · log write" } } } } ] } }, "sql_workers": { "kind": "Panel", "spec": { "display": { "name": "Workers and sessions", "description": "Worker-thread and session limits of the tier — hitting either rejects new work while CPU still looks fine." }, "plugin": { "kind": "TimeSeriesChart", "spec": { "legend": { "position": "bottom", "mode": "table", "values": [ "last-number", "mean", "max" ] }, "yAxis": { "format": { "unit": "percent" }, "min": 0, "max": 100 } } }, "queries": [ { "kind": "TimeSeriesQuery", "spec": { "plugin": { "kind": "PrometheusTimeSeriesQuery", "spec": { "query": "max by (resource_group, name) (azure_workers_percent_average{resource_group=~\"$instance\",type=\"Microsoft.Sql/servers/databases\"})", "seriesNameFormat": "{{name}} · workers" } } } }, { "kind": "TimeSeriesQuery", "spec": { "plugin": { "kind": "PrometheusTimeSeriesQuery", "spec": { "query": "max by (resource_group, name) (azure_sessions_percent_average{resource_group=~\"$instance\",type=\"Microsoft.Sql/servers/databases\"})", "seriesNameFormat": "{{name}} · sessions" } } } } ] } }, "storage_availability": { "kind": "Panel", "spec": { "display": { "name": "Availability", "description": "Share of requests the storage account served without a service-side error — the single best storage health signal." }, "plugin": { "kind": "TimeSeriesChart", "spec": { "legend": { "position": "bottom", "mode": "table", "values": [ "last-number", "mean", "max" ] }, "yAxis": { "format": { "unit": "percent" }, "min": 0, "max": 100 } } }, "queries": [ { "kind": "TimeSeriesQuery", "spec": { "plugin": { "kind": "PrometheusTimeSeriesQuery", "spec": { "query": "min by (resource_group, name) (azure_availability_average{resource_group=~\"$instance\",type=\"Microsoft.Storage/storageAccounts\"})", "seriesNameFormat": "{{name}}" } } } } ] } }, "storage_transactions": { "kind": "Panel", "spec": { "display": { "name": "Transactions (per period)", "description": "Requests against the account per period — the number the transaction bill is computed from." }, "plugin": { "kind": "TimeSeriesChart", "spec": { "legend": { "position": "bottom", "mode": "table", "values": [ "last-number", "mean", "max" ] }, "yAxis": { "format": { "unit": "decimal" }, "min": 0 } } }, "queries": [ { "kind": "TimeSeriesQuery", "spec": { "plugin": { "kind": "PrometheusTimeSeriesQuery", "spec": { "query": "sum by (resource_group, name) (azure_transactions_total{resource_group=~\"$instance\",type=\"Microsoft.Storage/storageAccounts\"})", "seriesNameFormat": "{{name}}" } } } } ] } }, "storage_capacity": { "kind": "Panel", "spec": { "display": { "name": "Used capacity", "description": "Bytes stored in the account. Reported once or twice a day, so the line is deliberately coarse." }, "plugin": { "kind": "TimeSeriesChart", "spec": { "legend": { "position": "bottom", "mode": "table", "values": [ "last-number", "mean", "max" ] }, "yAxis": { "format": { "unit": "bytes" }, "min": 0 } } }, "queries": [ { "kind": "TimeSeriesQuery", "spec": { "plugin": { "kind": "PrometheusTimeSeriesQuery", "spec": { "query": "max by (resource_group, name) (azure_usedcapacity_average{resource_group=~\"$instance\",type=\"Microsoft.Storage/storageAccounts\"})", "seriesNameFormat": "{{name}}" } } } } ] } }, "storage_latency": { "kind": "Panel", "spec": { "display": { "name": "End-to-end / server latency", "description": "Server latency is Azure’s time to process; end-to-end adds network and client time. A gap between them is a client-side problem." }, "plugin": { "kind": "TimeSeriesChart", "spec": { "legend": { "position": "bottom", "mode": "table", "values": [ "last-number", "mean", "max" ] }, "yAxis": { "format": { "unit": "milliseconds" }, "min": 0 } } }, "queries": [ { "kind": "TimeSeriesQuery", "spec": { "plugin": { "kind": "PrometheusTimeSeriesQuery", "spec": { "query": "max by (resource_group, name) (azure_successe2elatency_average{resource_group=~\"$instance\",type=\"Microsoft.Storage/storageAccounts\"})", "seriesNameFormat": "{{name}} · end-to-end" } } } }, { "kind": "TimeSeriesQuery", "spec": { "plugin": { "kind": "PrometheusTimeSeriesQuery", "spec": { "query": "max by (resource_group, name) (azure_successserverlatency_average{resource_group=~\"$instance\",type=\"Microsoft.Storage/storageAccounts\"})", "seriesNameFormat": "{{name}} · server" } } } } ] } }, "storage_traffic": { "kind": "Panel", "spec": { "display": { "name": "Ingress / egress (per period)", "description": "Bytes in and out of the account — egress is the line that shows up on the bandwidth bill." }, "plugin": { "kind": "TimeSeriesChart", "spec": { "legend": { "position": "bottom", "mode": "table", "values": [ "last-number", "mean", "max" ] }, "yAxis": { "format": { "unit": "bytes" }, "min": 0 } } }, "queries": [ { "kind": "TimeSeriesQuery", "spec": { "plugin": { "kind": "PrometheusTimeSeriesQuery", "spec": { "query": "sum by (resource_group, name) (azure_ingress_total{resource_group=~\"$instance\",type=\"Microsoft.Storage/storageAccounts\"})", "seriesNameFormat": "{{name}} · ingress" } } } }, { "kind": "TimeSeriesQuery", "spec": { "plugin": { "kind": "PrometheusTimeSeriesQuery", "spec": { "query": "sum by (resource_group, name) (azure_egress_total{resource_group=~\"$instance\",type=\"Microsoft.Storage/storageAccounts\"})", "seriesNameFormat": "{{name}} · egress" } } } } ] } }, "sb_backlog": { "kind": "Panel", "spec": { "display": { "name": "Active and dead-lettered messages", "description": "Messages waiting to be received, and messages that gave up trying — a rising dead-letter count is a stuck consumer." }, "plugin": { "kind": "TimeSeriesChart", "spec": { "legend": { "position": "bottom", "mode": "table", "values": [ "last-number", "mean", "max" ] }, "yAxis": { "format": { "unit": "decimal" }, "min": 0 } } }, "queries": [ { "kind": "TimeSeriesQuery", "spec": { "plugin": { "kind": "PrometheusTimeSeriesQuery", "spec": { "query": "sum by (resource_group, name) (azure_activemessages_average{resource_group=~\"$instance\",type=\"Microsoft.ServiceBus/namespaces\"})", "seriesNameFormat": "{{name}} · active" } } } }, { "kind": "TimeSeriesQuery", "spec": { "plugin": { "kind": "PrometheusTimeSeriesQuery", "spec": { "query": "sum by (resource_group, name) (azure_deadletteredmessages_average{resource_group=~\"$instance\",type=\"Microsoft.ServiceBus/namespaces\"})", "seriesNameFormat": "{{name}} · dead-lettered" } } } } ] } }, "sb_throughput": { "kind": "Panel", "spec": { "display": { "name": "Incoming / outgoing messages (per period)", "description": "Producer and consumer throughput side by side — a widening gap is a growing backlog." }, "plugin": { "kind": "TimeSeriesChart", "spec": { "legend": { "position": "bottom", "mode": "table", "values": [ "last-number", "mean", "max" ] }, "yAxis": { "format": { "unit": "decimal" }, "min": 0 } } }, "queries": [ { "kind": "TimeSeriesQuery", "spec": { "plugin": { "kind": "PrometheusTimeSeriesQuery", "spec": { "query": "sum by (resource_group, name) (azure_incomingmessages_total{resource_group=~\"$instance\",type=\"Microsoft.ServiceBus/namespaces\"})", "seriesNameFormat": "{{name}} · incoming" } } } }, { "kind": "TimeSeriesQuery", "spec": { "plugin": { "kind": "PrometheusTimeSeriesQuery", "spec": { "query": "sum by (resource_group, name) (azure_outgoingmessages_total{resource_group=~\"$instance\",type=\"Microsoft.ServiceBus/namespaces\"})", "seriesNameFormat": "{{name}} · outgoing" } } } } ] } }, "sb_errors": { "kind": "Panel", "spec": { "display": { "name": "Errors and throttling (per period)", "description": "Server errors are the namespace’s fault, user errors the client’s, throttled requests mean the tier’s quota is the limit." }, "plugin": { "kind": "TimeSeriesChart", "spec": { "legend": { "position": "bottom", "mode": "table", "values": [ "last-number", "mean", "max" ] }, "yAxis": { "format": { "unit": "decimal" }, "min": 0 } } }, "queries": [ { "kind": "TimeSeriesQuery", "spec": { "plugin": { "kind": "PrometheusTimeSeriesQuery", "spec": { "query": "sum by (resource_group, name) (azure_servererrors_total{resource_group=~\"$instance\",type=\"Microsoft.ServiceBus/namespaces\"})", "seriesNameFormat": "{{name}} · server" } } } }, { "kind": "TimeSeriesQuery", "spec": { "plugin": { "kind": "PrometheusTimeSeriesQuery", "spec": { "query": "sum by (resource_group, name) (azure_usererrors_total{resource_group=~\"$instance\",type=\"Microsoft.ServiceBus/namespaces\"})", "seriesNameFormat": "{{name}} · user" } } } }, { "kind": "TimeSeriesQuery", "spec": { "plugin": { "kind": "PrometheusTimeSeriesQuery", "spec": { "query": "sum by (resource_group, name) (azure_throttledrequests_total{resource_group=~\"$instance\",type=\"Microsoft.ServiceBus/namespaces\"})", "seriesNameFormat": "{{name}} · throttled" } } } } ] } } }, "layouts": [ { "kind": "Grid", "spec": { "display": { "title": "Virtual Machines", "collapse": { "open": true } }, "items": [ { "x": 0, "y": 0, "width": 12, "height": 8, "content": { "$ref": "#/spec/panels/vm_cpu" } }, { "x": 12, "y": 0, "width": 12, "height": 8, "content": { "$ref": "#/spec/panels/vm_available_memory" } }, { "x": 0, "y": 8, "width": 12, "height": 8, "content": { "$ref": "#/spec/panels/vm_network_io" } }, { "x": 12, "y": 8, "width": 12, "height": 8, "content": { "$ref": "#/spec/panels/vm_disk_io" } }, { "x": 0, "y": 16, "width": 12, "height": 8, "content": { "$ref": "#/spec/panels/vm_cpu_credits" } }, { "x": 12, "y": 16, "width": 12, "height": 8, "content": { "$ref": "#/spec/panels/vm_disk_iops_consumed" } } ] } }, { "kind": "Grid", "spec": { "display": { "title": "App Service", "collapse": { "open": true } }, "items": [ { "x": 0, "y": 0, "width": 12, "height": 8, "content": { "$ref": "#/spec/panels/app_requests" } }, { "x": 12, "y": 0, "width": 12, "height": 8, "content": { "$ref": "#/spec/panels/app_errors" } }, { "x": 0, "y": 8, "width": 12, "height": 8, "content": { "$ref": "#/spec/panels/app_response_time" } }, { "x": 12, "y": 8, "width": 12, "height": 8, "content": { "$ref": "#/spec/panels/app_memory" } }, { "x": 0, "y": 16, "width": 12, "height": 8, "content": { "$ref": "#/spec/panels/app_cpu_time" } } ] } }, { "kind": "Grid", "spec": { "display": { "title": "Azure SQL", "collapse": { "open": true } }, "items": [ { "x": 0, "y": 0, "width": 12, "height": 8, "content": { "$ref": "#/spec/panels/sql_cpu" } }, { "x": 12, "y": 0, "width": 12, "height": 8, "content": { "$ref": "#/spec/panels/sql_storage" } }, { "x": 0, "y": 8, "width": 12, "height": 8, "content": { "$ref": "#/spec/panels/sql_connections" } }, { "x": 12, "y": 8, "width": 12, "height": 8, "content": { "$ref": "#/spec/panels/sql_deadlocks" } }, { "x": 0, "y": 16, "width": 12, "height": 8, "content": { "$ref": "#/spec/panels/sql_io" } }, { "x": 12, "y": 16, "width": 12, "height": 8, "content": { "$ref": "#/spec/panels/sql_workers" } } ] } }, { "kind": "Grid", "spec": { "display": { "title": "Storage", "collapse": { "open": true } }, "items": [ { "x": 0, "y": 0, "width": 12, "height": 8, "content": { "$ref": "#/spec/panels/storage_availability" } }, { "x": 12, "y": 0, "width": 12, "height": 8, "content": { "$ref": "#/spec/panels/storage_transactions" } }, { "x": 0, "y": 8, "width": 12, "height": 8, "content": { "$ref": "#/spec/panels/storage_capacity" } }, { "x": 12, "y": 8, "width": 12, "height": 8, "content": { "$ref": "#/spec/panels/storage_latency" } }, { "x": 0, "y": 16, "width": 12, "height": 8, "content": { "$ref": "#/spec/panels/storage_traffic" } } ] } }, { "kind": "Grid", "spec": { "display": { "title": "Service Bus", "collapse": { "open": true } }, "items": [ { "x": 0, "y": 0, "width": 12, "height": 8, "content": { "$ref": "#/spec/panels/sb_backlog" } }, { "x": 12, "y": 0, "width": 12, "height": 8, "content": { "$ref": "#/spec/panels/sb_throughput" } }, { "x": 0, "y": 8, "width": 12, "height": 8, "content": { "$ref": "#/spec/panels/sb_errors" } } ] } } ], "variables": [ { "kind": "ListVariable", "spec": { "name": "instance", "display": { "name": "Instance" }, "allowAllValue": true, "allowMultiple": true, "customAllValue": ".*", "defaultValue": ".*", "plugin": { "kind": "PrometheusLabelValuesVariable", "spec": { "labelName": "resource_group" } } } } ], "duration": "1h", "refreshInterval": "30s", "datasources": {} } } ], "alert_rules": [ { "ref": "alert-azure-app-high-5xx-ratio", "name": "App Service high 5xx ratio", "description": "A large share of requests to an App Service app are getting 5xx responses, so users are seeing errors.", "query": "sum by (resource_group, name) (azure_http5xx_total{type=\"Microsoft.Web/sites\"}) / clamp_min(sum by (resource_group, name) (azure_requests_total{type=\"Microsoft.Web/sites\"}), 1)", "condition_op": "gt", "threshold": 0.05, "interval": "1m", "for": "10m", "severity": "warning", "message": "App Service {{labels.name}} ({{labels.resource_group}}) is serving {{value | percent:1}} 5xx responses.", "notify_on_resolve": true, "paused": false, "channels": [] }, { "ref": "alert-azure-sql-storage-high", "name": "Azure SQL storage nearly full", "description": "An Azure SQL database is close to its allocated storage; it stops accepting writes when full.", "query": "max by (resource_group, name) (azure_storage_percent_average{type=\"Microsoft.Sql/servers/databases\"})", "condition_op": "gt", "threshold": 90, "interval": "1m", "for": "15m", "severity": "critical", "message": "Database {{labels.name}} ({{labels.resource_group}}) is {{value}}% full — it stops accepting writes at 100%.", "notify_on_resolve": true, "paused": false, "channels": [] }, { "ref": "alert-azure-storage-availability-low", "name": "Storage account availability degraded", "description": "A storage account is failing requests service-side, so applications reading or writing it see errors.", "query": "min by (resource_group, name) (azure_availability_average{type=\"Microsoft.Storage/storageAccounts\"})", "condition_op": "lt", "threshold": 99, "interval": "1m", "for": "10m", "severity": "critical", "message": "Storage account {{labels.name}} ({{labels.resource_group}}) is at {{value}}% availability — requests are failing service-side.", "notify_on_resolve": true, "paused": false, "channels": [] }, { "ref": "alert-azure-servicebus-deadletter", "name": "Service Bus dead-lettered messages", "description": "Messages are landing in the dead-letter queue, meaning consumers are rejecting or timing out.", "query": "sum by (resource_group, name) (azure_deadletteredmessages_average{type=\"Microsoft.ServiceBus/namespaces\"})", "condition_op": "gt", "threshold": 0, "interval": "1m", "for": "15m", "severity": "warning", "message": "Service Bus namespace {{labels.name}} ({{labels.resource_group}}) holds {{value}} dead-lettered message(s) — consumers are rejecting or timing out.", "notify_on_resolve": true, "paused": false, "channels": [] }, { "ref": "alert-azure-vm-low-memory", "name": "VM low available memory", "description": "A VM is nearly out of available memory, so paging and the OOM killer come next.", "query": "min by (resource_group, name) (azure_available_memory_bytes_average{type=\"Microsoft.Compute/virtualMachines\"})", "condition_op": "lt", "threshold": 268435456, "interval": "1m", "for": "15m", "severity": "critical", "message": "VM {{labels.name}} ({{labels.resource_group}}) has {{value | bytes}} of memory left — paging and the OOM killer come next.", "notify_on_resolve": true, "paused": false, "channels": [] }, { "ref": "alert-azure-vm-high-cpu", "name": "VM high CPU", "description": "A VM has been running at sustained high CPU, leaving no headroom for load spikes.", "query": "avg by (resource_group, name) (azure_percentage_cpu_average{type=\"Microsoft.Compute/virtualMachines\"})", "condition_op": "gt", "threshold": 90, "interval": "1m", "for": "30m", "severity": "warning", "message": "VM {{labels.name}} ({{labels.resource_group}}) has averaged {{value}}% CPU for 30 minutes.", "notify_on_resolve": true, "paused": false, "channels": [] }, { "ref": "alert-azure-sql-high-cpu", "name": "Azure SQL high CPU", "description": "An Azure SQL database is running at sustained high CPU for its service tier, degrading query latency.", "query": "avg by (resource_group, name) (azure_cpu_percent_average{type=\"Microsoft.Sql/servers/databases\"})", "condition_op": "gt", "threshold": 90, "interval": "1m", "for": "15m", "severity": "warning", "message": "Database {{labels.name}} ({{labels.resource_group}}) is at {{value}}% CPU of its service tier.", "notify_on_resolve": true, "paused": false, "channels": [] }, { "ref": "alert-azure-sql-connection-failures", "name": "Azure SQL connection failures", "description": "Logins to an Azure SQL database keep failing, pointing at credentials, firewall rules, or limits.", "query": "sum by (resource_group, name) (azure_connection_failed_total{type=\"Microsoft.Sql/servers/databases\"})", "condition_op": "gt", "threshold": 10, "interval": "1m", "for": "10m", "severity": "warning", "message": "Database {{labels.name}} ({{labels.resource_group}}) is rejecting {{value}} logins per period — check credentials, firewall rules, and connection limits.", "notify_on_resolve": true, "paused": false, "channels": [] }, { "ref": "alert-azure-app-slow-responses", "name": "App Service slow responses", "description": "An App Service app is taking a long time to serve requests on average, so users are waiting.", "query": "max by (resource_group, name) (azure_httpresponsetime_average{type=\"Microsoft.Web/sites\"})", "condition_op": "gt", "threshold": 2, "interval": "1m", "for": "15m", "severity": "warning", "message": "App Service {{labels.name}} ({{labels.resource_group}}) is averaging {{value}}s per request.", "notify_on_resolve": true, "paused": false, "channels": [] }, { "ref": "alert-azure-vm-cpu-credits-low", "name": "VM CPU credits low", "description": "A burstable B-series VM is nearly out of CPU credits and will hard-throttle at zero.", "query": "min by (resource_group, name) (azure_cpu_credits_remaining_average{type=\"Microsoft.Compute/virtualMachines\"})", "condition_op": "lt", "threshold": 20, "interval": "1m", "for": "15m", "severity": "warning", "message": "VM {{labels.name}} ({{labels.resource_group}}) is down to {{value}} CPU credits — it will hard-throttle at zero.", "notify_on_resolve": true, "paused": false, "channels": [] }, { "ref": "alert-azure-sql-workers-high", "name": "Azure SQL worker saturation", "description": "An Azure SQL database is near the worker-thread limit of its tier, after which new queries are rejected.", "query": "max by (resource_group, name) (azure_workers_percent_average{type=\"Microsoft.Sql/servers/databases\"})", "condition_op": "gt", "threshold": 80, "interval": "1m", "for": "15m", "severity": "warning", "message": "Database {{labels.name}} ({{labels.resource_group}}) is using {{value}}% of its worker threads — new queries are rejected at 100%.", "notify_on_resolve": true, "paused": false, "channels": [] }, { "ref": "alert-azure-servicebus-throttled", "name": "Service Bus throttled requests", "description": "Service Bus is throttling requests because the namespace hit its tier quota, so senders and receivers slow down.", "query": "sum by (resource_group, name) (azure_throttledrequests_total{type=\"Microsoft.ServiceBus/namespaces\"})", "condition_op": "gt", "threshold": 0, "interval": "1m", "for": "10m", "severity": "warning", "message": "Service Bus namespace {{labels.name}} ({{labels.resource_group}}) is being throttled ({{value}} requests per period) — the tier quota is the limit.", "notify_on_resolve": true, "paused": false, "channels": [] } ] } }