Closed zxhdaze closed 4 months ago
dashmage
commented
4 months ago This looks good but is there a way to hide a panel in case there's no data there (eg: SMART error log count)? And have we checked with the Managed Solutions team about whether they have any requirements for the dashboard?
jneo8
commented
4 months ago Here is the output by cat src/grafana_dashboards/smart.json | grep expr
"expr": "sum(smartctl_devices{instance=~\"$instance\", job=\"$job\"})",
"expr": "sum(smartctl_devices{instance=~\"$instance\", job=\"$job\"}) - sum(smartctl_device{instance=~\"$instance\", job=\"$job\"})",
"expr": "max(smartctl_device_smartctl_exit_status{instance=~\"$instance\", job=\"$job\"})",
"expr": "sum(smartctl_device_num_err_log_entries{instance=~\"$instance\", job=\"$job\"})",
"expr": "sum(smartctl_device_media_errors{instance=~\"$instance\", job=\"$job\"})",
"expr": "sum(smartctl_device_error_log_count{instance=~\"$instance\", job=\"$job\"})",
"expr": "sum(smartctl_device_critical_warning{instance=~\"$instance\", job=\"$job\"})",
"expr": "sum(1 - smartctl_device_smart_status{instance=~\"$instance\", job=\"$job\"})",
"expr": "smartctl_device{instance=~\"$instance\", job=\"$job\"}",
"expr": "sum(smartctl_device{instance=~\"$instance\", job=\"$job\"}) by (instance, device, model_name)",
"expr": "sum(smartctl_device_temperature{instance=~\"$instance\", job=\"$job\", temperature_type=\"current\"}) by (instance, device)",
"expr": "sum(smartctl_device_smartctl_exit_status{instance=~\"$instance\", job=\"$job\"}) by (instance, device)",
"expr": "sum(smartctl_device_smart_status{instance=~\"$instance\", job=\"$job\"}) by (instance, device)",
"expr": "sum(smartctl_device_power_on_seconds{instance=~\"$instance\", job=\"$job\"}) by (instance, device)",
"expr": "sum(smartctl_device_power_cycle_count{instance=~\"$instance\", job=\"$job\"}) by (instance, device)",
"expr": "sum(smartctl_device_interface_speed{instance=~\"$instance\", job=\"$job\", speed_type=\"current\"}) by (instance, device)",
"expr": "sum(smartctl_device_capacity_bytes{instance=~\"$instance\", job=\"$job\"}) by (instance, device)",
"expr": "sum(smartctl_device_capacity_blocks{instance=~\"$instance\", job=\"$job\"}) by (instance, device)",
"expr": "sum(smartctl_device_block_size{blocks_type=\"logical\", instance=~\"$instance\", job=\"$job\"}) by (instance, device)",
"expr": "sum(smartctl_device_block_size{blocks_type=\"physical\", instance=~\"$instance\", job=\"$job\"}) by (instance, device)",
"expr": "sum(smartctl_device_error_log_count{instance=~\"$instance\", job=\"$job\"}) by (instance, device)",
"expr": "smartctl_device_attribute{instance=~\"$instance\", job=\"$job\"} * on(device, instance) group_left(model_name) smartctl_device{instance=~\"$instance\", job=\"$job\"}",
"expr": "sum by (model_name) (smartctl_device_temperature{instance=~\"$instance\", job=\"$job\",temperature_type=\"current\"} * on(device, instance) group_left(model_name) smartctl_device{instance=~\"$instance\"})",
"expr": "sum by (model_name) (smartctl_device_smartctl_exit_status{instance=~\"$instance\", job=\"$job\"} * on(device, instance) group_left(model_name) smartctl_device{instance=~\"$instance\"})",
"expr": "sum by (model_name) (smartctl_device_smart_status{instance=~\"$instance\", job=\"$job\"} * on(device, instance) group_left(model_name) smartctl_device{instance=~\"$instance\"})",
"expr": "sum by (model_name) (increase(smartctl_device_power_on_seconds{instance=~\"$instance\", job=\"$job\"}[1h]) * on(device, instance) group_left(model_name) smartctl_device{instance=~\"$instance\"})",
"expr": "sum by (model_name) (smartctl_device_power_cycle_count{instance=~\"$instance\", job=\"$job\"} * on(device, instance) group_left(model_name) smartctl_device{instance=~\"$instance\"})",
"expr": "sum by (model_name) (smartctl_device_percentage_used{instance=~\"$instance\", job=\"$job\"} * on(device, instance) group_left(model_name) smartctl_device{instance=~\"$instance\"})",
"expr": "sum by (model_name) (smartctl_device_nvme_capacity_bytes{instance=~\"$instance\", job=\"$job\"} * on(device, instance) group_left(model_name) smartctl_device{instance=~\"$instance\"})",
"expr": "sum by (model_name) (smartctl_device_num_err_log_entries{instance=~\"$instance\", job=\"$job\"} * on(device, instance) group_left(model_name) smartctl_device{instance=~\"$instance\"})",
"expr": "sum by (model_name) (smartctl_device_media_errors{instance=~\"$instance\", job=\"$job\"} * on(device, instance) group_left(model_name) smartctl_device{instance=~\"$instance\"})",
"expr": "sum by (model_name) (smartctl_device_error_log_count{instance=~\"$instance\", job=\"$job\",error_log_type=\"summary\"} * on(device, instance) group_left(model_name) smartctl_device{instance=~\"$instance\"})",
"expr": "sum by (model_name) (smartctl_device_critical_warning{instance=~\"$instance\", job=\"$job\"} * on(device, instance) group_left(model_name) smartctl_device{instance=~\"$instance\"})",
"expr": "sum by (model_name) (smartctl_device_interface_speed{instance=~\"$instance\", job=\"$job\",speed_type=\"current\"} * on(device, instance) group_left(model_name) smartctl_device{instance=~\"$instance\"})",
"expr": "sum by (model_name) (smartctl_device_interface_speed{instance=~\"$instance\", job=\"$job\",speed_type=\"max\"} * on(device, instance) group_left(model_name) smartctl_device{instance=~\"$instance\"})",
"expr": "sum by (model_name) (smartctl_device_capacity_bytes{instance=~\"$instance\", job=\"$job\"} * on(device, instance) group_left(model_name) smartctl_device{instance=~\"$instance\"})",
"expr": "sum by (model_name) (smartctl_device_capacity_blocks{instance=~\"$instance\", job=\"$job\"} * on(device, instance) group_left(model_name) smartctl_device{instance=~\"$instance\"})",
"expr": "sum by (model_name) (smartctl_device_bytes_written{instance=~\"$instance\", job=\"$job\"} * on(device, instance) group_left(model_name) smartctl_device{instance=~\"$instance\"})",
"expr": "sum by (model_name) (smartctl_device_bytes_read{instance=~\"$instance\", job=\"$job\"} * on(device, instance) group_left(model_name) smartctl_device{instance=~\"$instance\"})",
"expr": "sum by (model_name) (smartctl_device_block_size{instance=~\"$instance\", job=\"$job\",blocks_type=\"logical\"} * on(device, instance) group_left(model_name) smartctl_device{instance=~\"$instance\"})",
"expr": "sum by (model_name) (smartctl_device_block_size{instance=~\"$instance\", job=\"$job\",blocks_type=\"physical\"} * on(device, instance) group_left(model_name) smartctl_device{instance=~\"$instance\"})",
"expr": "sum by (model_name) (smartctl_device_available_spare_threshold{instance=~\"$instance\", job=\"$job\"} * on(device, instance) group_left(model_name) smartctl_device{instance=~\"$instance\"})",
"expr": "sum by (model_name) (smartctl_device_available_spare{instance=~\"$instance\", job=\"$job\"} * on(device, instance) group_left(model_name) smartctl_device{instance=~\"$instance\"})",
"expr": "sum by (pod) (process_open_fds{instance=~\"$instance\", job=\"$job\"})",
"expr": "sum by (pod) (rate(process_cpu_seconds_total{instance=~\"$instance\", job=\"$job\"}[$__rate_interval]))",
"expr": "sum by (pod) (increase(process_resident_memory_bytes{instance=~\"$instance\", job=\"$job\"}[1h]))",
"expr": "avg(go_memstats_stack_sys_bytes{instance=~\"$instance\", job=\"$job\"})",
"expr": "avg(go_memstats_stack_inuse_bytes{instance=~\"$instance\", job=\"$job\"})",
"expr": "avg(go_memstats_heap_sys_bytes{instance=~\"$instance\", job=\"$job\"})",
"expr": "avg(go_memstats_heap_idle_bytes{instance=~\"$instance\", job=\"$job\"})",
"expr": "avg(go_memstats_heap_released_bytes{instance=~\"$instance\", job=\"$job\"})",
"expr": "avg(go_memstats_next_gc_bytes{instance=~\"$instance\", job=\"$job\"})",
"expr": "avg(go_memstats_heap_inuse_bytes{instance=~\"$instance\", job=\"$job\"})",
"expr": "avg(go_memstats_heap_alloc_bytes{instance=~\"$instance\", job=\"$job\"})",
"expr": "avg by (pod) (go_memstats_heap_objects{instance=~\"$instance\", job=\"$job\"})",
"expr": "sum by (pod) (go_threads{instance=~\"$instance\", job=\"$job\"})",
"expr": "sum by (pod) (go_goroutines{instance=~\"$instance\", job=\"$job\"})",
"expr": "sum by (pod) (rate(go_memstats_alloc_bytes_total{instance=~\"$instance\", job=\"$job\"}[$__rate_interval]))",
"expr": "sum by (pod) (go_memstats_alloc_bytes{instance=~\"$instance\", job=\"$job\"})",
"expr": "sum by (pod) (rate(go_gc_duration_seconds_count{instance=~\"$instance\", job=\"$job\"}[$__rate_interval]))",
"expr": "sum by (pod) (rate(go_gc_duration_seconds_sum{instance=~\"$instance\", job=\"$job\"}[$__rate_interval]))",
Add grafana dashboard, it will be automatically pushed to Grafana upon deployment. The dashboard has been tested with mocked data. Screenshots:
I've noticed that certain metrics are not being collected on my actual machine. Specifically, the smartctl_device_media_errors, smartctl_device_interface_speed, and smartctl_device_attribute metrics are missing.