up{instance=~".*:9100"}

up{instance=~".*:9101"}

up{instance=~".*:9650"}

up{instance=~".*:9323"}

up{instance=~".*:9598"}

up{job="etcd"}

up{job="minio"}

up{instance=~".*:9630"}

up{instance=~".*:9631"}

up{instance=~".*:9854"}

up{instance=~".*:8008"}

up{job="mysql"}

up{job="kafka",role=~".+"}

up{job="kafka",role=""}

etcd_up < 1

min(etcd_server_has_leader) by (cls) < 1

etcd:cls:quota_usage > 0.90

etcd:ins:network_peer_rt_p95_5m > 0.200

etcd:ins:wal_fsync_rt_p95_5m > 0.050

sum without (type, client_api_version) (increase(etcd_server_client_requests_total{job="etcd"}[1m]))

increase(etcd_server_proposals_committed_total{job="etcd"}[1m])

etcd_mvcc_db_total_size_in_bytes{job="etcd"} / etcd_server_quota_backend_bytes{job="etcd"}

max by (job,cls) (etcd_mvcc_db_total_size_in_bytes{job="etcd"} / etcd_server_quota_backend_bytes{job="etcd"})

etcd_debugging_mvcc_current_revision{job="etcd"} - etcd_debugging_mvcc_compact_revision{job="etcd"}

max by (job,cls) (etcd_debugging_mvcc_current_revision{job="etcd"} - etcd_debugging_mvcc_compact_revision{job="etcd"})

histogram_quantile(0.95, rate(etcd_disk_wal_fsync_duration_seconds_bucket{job="etcd"}[5m]))

histogram_quantile(0.95, rate(etcd_disk_backend_commit_duration_seconds_bucket{job="etcd"}[5m]))

histogram_quantile(0.95, rate(etcd_network_peer_round_trip_time_seconds_bucket{job="etcd"}[5m]))

infra_up < 1

agent_up < 1

up{job="infra"}

up{job!~"infra|node|etcd|minio|pgsql|redis|mysql"}

up{job="kafka",role=~".+"} < 1

up{job="kafka",role=""} < 1

jmx_scrape_error{job="kafka"} > 0

kafka:ins:jvm_heap_used_ratio > 0.90

jvm_threads_deadlocked{job="kafka"} > 0

kafka_server_request_handler_idle_ratio{job="kafka"} < 0.10

kafka_server_replica_manager_under_replicated_partitions{job="kafka"} > 0

kafka_server_replica_manager_under_min_isr_partitions{job="kafka"} > 0

kafka_log_manager_offline_log_directories{job="kafka"} > 0

max by (cls) (kafka_controller_offline_partition_count{job="kafka"}) > 0

(sum by (cls) (kafka_controller_active_controller_count{job="kafka"}) or 0 * count by (cls) (up{job="kafka",role=~"combined|controller"})) != 1

max by (cls) (kafka_controller_fenced_broker_count{job="kafka"}) > 0

increase(kafka_controller_unclean_leader_elections_total{job="kafka"}[5m]) > 0

kafka_network_processor_idle_ratio{job="kafka"} < 0.10

kafka:csg:lag > 100000 and on (job, cls, consumergroup) delta(kafka:csg:lag[30m]) > 0

sum by (job, cls, ins, ip, instance, role, node_id) (jvm_memory_used_bytes{job="kafka",area="heap"}) / clamp_min(sum by (job, cls, ins, ip, instance, role, node_id) (jvm_memory_max_bytes{job="kafka",area="heap"}), 1)

rate(process_cpu_seconds_total{job="kafka",role=~".+"}[5m])

1 - clamp( min by (job, cls, ins, ip, instance, role, node_id) ( kafka_server_request_handler_idle_ratio{job="kafka"} or kafka_network_processor_idle_ratio{job="kafka"} ), 0, 1)

avg by (job, cls) (kafka:ins:load)

sum without (gc) (rate(jvm_gc_collection_seconds_sum{job="kafka"}[5m]))

rate(kafka_server_broker_messages_in_total{job="kafka"}[5m])

rate(kafka_server_broker_bytes_in_total{job="kafka"}[5m])

rate(kafka_server_broker_bytes_out_total{job="kafka"}[5m])

sum without (request, error) (rate(kafka_network_request_errors_total{job="kafka",error!="NONE"}[5m]))

sum by (job, cls) (kafka_server_replica_manager_under_replicated_partitions{job="kafka"})

max by (job, cls) (kafka_controller_offline_partition_count{job="kafka"})

sum by (job, cls, topic) (max without (ins, ip, instance) (clamp_min(rate(kafka_topic_partition_current_offset[1m]), 0)))

sum by (job, cls, topic) (max without (ins, ip, instance) (clamp_min(rate(kafka_topic_partition_current_offset[5m]), 0)))

sum without (topic) (kafka:topic:msg_rate1m)

sum without (topic) (kafka:topic:msg_rate5m)

sum by (job, cls, consumergroup, topic) (max without (ins, ip, instance) (clamp_min(rate(kafka_consumergroup_current_offset[5m]), 0)))

sum by (job, cls, consumergroup, topic) (max without (ins, ip, instance) (clamp_min(kafka_consumergroup_lag, 0)))

sum without (topic) (kafka:csg_topic:lag)

sum without (consumergroup) (kafka:csg:lag)

minio_up < 1

max by (cls) (avg_over_time(minio_cluster_health_nodes_offline_count{job="minio"}[5m])) > 0

max by (cls) (avg_over_time(minio_cluster_health_drives_offline_count{job="minio"}[5m])) > 0

min by (cls) (
  minio_cluster_erasure_set_overall_health{job="minio"}
  or (minio_cluster_erasure_set_overall_write_quorum{job="minio"} * 0)
) < 1

max by (cls) (
  1 - (
    (
      minio_cluster_health_capacity_usable_free_bytes{job="minio"}
      or (minio_cluster_health_capacity_usable_total_bytes{job="minio"} * 0)
    )
    / minio_cluster_health_capacity_usable_total_bytes{job="minio"}
  )
) > 0.90

mysql:ins:exporter_up < 1

mysql:ins:exporter_up == 1 and on (job, cls, ins, ip, instance, topology) mysql:ins:up < 1

mysql_exporter_collector_success{job="mysql"} < 1

mysql:ins:uptime < 600

mysql:ins:connection_usage > 0.80 and mysql:ins:connection_usage <= 0.95

mysql:ins:connection_usage > 0.95

mysql:ins:connection_errors > 0.1

mysql:ins:aborted_connects > 1 or mysql:ins:aborted_clients > 1

mysql:ins:slow_query_ratio > 0.01 and mysql:ins:qps > 10

mysql:ins:tmp_disk_ratio > 0.25 and mysql:ins:tmp_tables > 1

mysql:ins:full_joins > 1

mysql:ins:buffer_pool_hit_ratio < 0.95 and mysql:ins:qps > 100

mysql:ins:buffer_pool_waits > 0

mysql:ins:log_waits > 0

mysql:ins:redo_utilization > 0.80

mysql:ins:deadlocks > 0

mysql:ins:history_list_length > 1000000

mysql:cls:gr_primary_members{topology="innodb_cluster"} < 1

mysql:cls:gr_primary_members{topology="innodb_cluster"} > 1

mysql:cls:gr_quorum{topology="innodb_cluster"} < 1

mysql:ins:up{topology="innodb_cluster"} == 1 unless on (job, cls, ins, ip, instance, topology) mysql:ins:gr_online == 1

mysql:ins:gr_certifier_queue > 1000 or mysql:ins:gr_applier_queue > 1000

mysql:ins:gr_certifier_queue > 10000 or mysql:ins:gr_applier_queue > 10000

mysql:ins:gr_certifier_queue_ratio > 0.80 or mysql:ins:gr_applier_queue_ratio > 0.80

mysql_global_variables_read_only{job="mysql"} > 0 and on (job, cls, ins, ip, instance, topology) mysql:ins:gr_primary == 1

mysql_global_variables_super_read_only{job="mysql"} < 1 and on (job, cls, ins, ip, instance, topology) mysql:ins:gr_secondary == 1

max without (collector) (mysql_exporter_collector_duration_seconds{job="mysql"}) > 5

up{job="mysql"}

mysql_up{job="mysql"}

mysql_global_status_uptime{job="mysql"}

count by (cls, topology) (up{job="mysql"})

sum by (cls, topology) (mysql:ins:up) or (0 * count by (cls, topology) (up{job="mysql"}))

rate(mysql_global_status_questions{job="mysql"}[1m])

sum without (command) (rate(mysql_global_status_commands_total{job="mysql",command=~"commit|rollback"}[1m]))

sum without (command) (rate(mysql_global_status_commands_total{job="mysql",command=~"select|show.*"}[1m]))

sum without (command) (rate(mysql_global_status_commands_total{job="mysql",command=~"insert|update|delete|replace|load"}[1m]))

rate(mysql_global_status_innodb_row_ops_total{job="mysql"}[1m])

sum without (digest, schema) (rate(mysql_perf_schema_events_statements_sum_total{job="mysql"}[5m]))

sum without (digest, schema) (rate(mysql_perf_schema_events_statements_sum_timer_wait{job="mysql"}[5m])) / clamp_min(mysql:ins:statement_rate, 0.001)

sum without (digest, schema) (rate(mysql_perf_schema_events_statements_sum_rows_examined{job="mysql"}[5m])) / clamp_min(mysql:ins:statement_rate, 0.001)

mysql_global_status_threads_connected{job="mysql"}

mysql_global_status_threads_connected{job="mysql"} / clamp_min(mysql_global_variables_max_connections{job="mysql"}, 1)

rate(mysql_global_status_connections{job="mysql"}[1m])

mysql_global_status_threads_running{job="mysql"}

mysql_global_status_threads_cached{job="mysql"}

rate(mysql_global_status_slow_queries{job="mysql"}[5m])

mysql:ins:slow_queries / clamp_min(mysql:ins:qps, 0.001)

rate(mysql_global_status_aborted_connects{job="mysql"}[5m])

rate(mysql_global_status_aborted_clients{job="mysql"}[5m])

sum without (error) (rate(mysql_global_status_connection_errors_total{job="mysql"}[5m]))

rate(mysql_global_status_bytes_received{job="mysql"}[1m])

rate(mysql_global_status_bytes_sent{job="mysql"}[1m])

rate(mysql_global_status_created_tmp_tables{job="mysql"}[5m])

rate(mysql_global_status_created_tmp_disk_tables{job="mysql"}[5m])

mysql:ins:tmp_disk_tables / clamp_min(mysql:ins:tmp_tables, 0.001)

rate(mysql_global_status_select_full_join{job="mysql"}[5m])

rate(mysql_global_status_select_scan{job="mysql"}[5m])

rate(mysql_global_status_sort_merge_passes{job="mysql"}[5m])

rate(mysql_global_status_table_open_cache_hits{job="mysql"}[5m]) / clamp_min(rate(mysql_global_status_table_open_cache_hits{job="mysql"}[5m]) + rate(mysql_global_status_table_open_cache_misses{job="mysql"}[5m]), 0.001)

mysql_global_status_open_files{job="mysql"} / clamp_min(mysql_global_variables_open_files_limit{job="mysql"}, 1)

sum without (digest, schema) (rate(mysql_perf_schema_events_statements_sum_errors{job="mysql"}[5m]))

sum without (digest, schema) (rate(mysql_perf_schema_events_statements_sum_no_index_used{job="mysql"}[5m]))

clamp_min(1 - rate(mysql_global_status_innodb_buffer_pool_reads{job="mysql"}[5m]) / clamp_min(rate(mysql_global_status_innodb_buffer_pool_read_requests{job="mysql"}[5m]), 1), 0)

mysql_global_status_innodb_buffer_pool_bytes_data{job="mysql"} / clamp_min(mysql_global_variables_innodb_buffer_pool_size{job="mysql"}, 1)

mysql_global_status_innodb_buffer_pool_bytes_dirty{job="mysql"} / clamp_min(mysql_global_status_innodb_buffer_pool_bytes_data{job="mysql"}, 1)

rate(mysql_global_status_innodb_buffer_pool_wait_free{job="mysql"}[5m])

rate(mysql_global_status_innodb_data_read{job="mysql"}[1m])

rate(mysql_global_status_innodb_data_written{job="mysql"}[1m])

rate(mysql_global_status_innodb_data_reads{job="mysql"}[1m])

rate(mysql_global_status_innodb_data_writes{job="mysql"}[1m])

rate(mysql_global_status_innodb_data_fsyncs{job="mysql"}[1m])

rate(mysql_global_status_innodb_os_log_written{job="mysql"}[1m])

clamp_min(mysql_global_status_innodb_redo_log_current_lsn{job="mysql"} - mysql_global_status_innodb_redo_log_checkpoint_lsn{job="mysql"}, 0) / clamp_min(mysql_global_status_innodb_redo_log_capacity_resized{job="mysql"}, 1)

rate(mysql_global_status_innodb_log_waits{job="mysql"}[5m])

rate(mysql_global_status_innodb_row_lock_waits{job="mysql"}[5m])

rate(mysql_global_status_innodb_row_lock_time{job="mysql"}[5m]) / 1000

rate(mysql_info_schema_innodb_metrics_lock_lock_deadlocks_total{job="mysql"}[5m])

mysql_info_schema_innodb_metrics_transaction_trx_rseg_history_len{job="mysql"}

sum without (log_file) (mysql_binlog_size_bytes{job="mysql"})

mysql_perf_schema_replication_group_member_info{job="mysql"} and on (job, cls, ins, member_host) label_replace(mysql_up{job="mysql"} == 1, "member_host", "$1", "ip", "(.*)")

max by (job, cls, ins, ip, instance, topology) (mysql:ins:gr_member{member_state="ONLINE"})

max by (job, cls, ins, ip, instance, topology) (mysql:ins:gr_member{member_state="ONLINE",member_role="PRIMARY"})

max by (job, cls, ins, ip, instance, topology) (mysql:ins:gr_member{member_state="ONLINE",member_role="SECONDARY"})

sum by (cls, topology) (mysql:ins:gr_online) or (0 * count by (cls, topology) (up{job="mysql",topology="innodb_cluster"}))

sum by (cls, topology) (mysql:ins:gr_primary) or (0 * count by (cls, topology) (up{job="mysql",topology="innodb_cluster"}))

mysql:cls:gr_online_members > bool mysql:cls:instances / 2

mysql:cls:gr_primary_members == bool 1

mysql:cls:gr_quorum * mysql:cls:gr_single_primary * (1 + (mysql:cls:gr_online_members == bool mysql:cls:instances)) or 2 * mysql:cls:up{topology="standalone"}

mysql_perf_schema_transactions_in_queue{job="mysql"}

mysql_perf_schema_transactions_remote_in_applier_queue{job="mysql"}

rate(mysql_perf_schema_conflicts_detected_total{job="mysql"}[5m])

rate(mysql_perf_schema_transactions_checked_total{job="mysql"}[5m])

rate(mysql_perf_schema_transactions_remote_applied_total{job="mysql"}[5m])

mysql:ins:gr_certifier_queue / clamp_min(mysql_global_variables_group_replication_flow_control_certifier_threshold{job="mysql"}, 1)

mysql:ins:gr_applier_queue / clamp_min(mysql_global_variables_group_replication_flow_control_applier_threshold{job="mysql"}, 1)

node_up < 1

haproxy_up < 1

vector_up < 1

docker_up < 1

keepalived_up < 1

node:ins:cpu_usage_1m > 0.70

node:ins:stdload1 > 1

node:ins:mem_avail < 0.10

node:ins:swap_usage > 0.01

node:fs:space_usage > 0.90

node:fs:inode_usage > 0.90

node:ins:fd_usage > 0.90

node:dev:disk_read_rt_1m{device="dfa"} > 0.032 or node:dev:disk_write_rt_1m{device="dfa"} > 0.032

rate(node:ins:tcp_error[1m]) > 1

node:ins:tcp_retrans_ratio1m > 1e-2

node_timex_sync_status != 1

time() - node_boot_time_seconds

sum without (machine, release, version, sysname, domainname) (
  label_replace(node_uname_info, "id", "$1", "nodename", "(.+)") OR
  label_replace(node_uname_info, "id", "$1", "instance", "(.+)\\:\\d+") OR
  label_replace(node_uname_info, "id", "$1", "ins", "(.+)")
)

irate(node_cpu_seconds_total[1m])

sum without (mode) (node:cpu:time_irate1m)

sum without (mode) (node:cpu:time_irate1m{mode="idle"})

1 - node:cpu:idle_time_irate1m / node:cpu:total_time_irate1m

avg_over_time(node:cpu:usage[1m])

avg_over_time(node:cpu:usage[5m])

avg_over_time(node:cpu:usage[15m])

count without (cpu) (node:cpu:usage)

sum by (cls, job) (node:ins:cpu_count)

sum by (job) (node:ins:cpu_count)

avg without (cpu) (node:cpu:usage)

avg by (cls, job) (node:cpu:usage)

sum by (job) (node:cls:cpu_usage * node:cls:cpu_count) / sum by (job) (node:cls:cpu_count)

avg_over_time(node:ins:cpu_usage[1m])

avg_over_time(node:ins:cpu_usage[5m])

avg_over_time(node:ins:cpu_usage[15m])

avg_over_time(node:cls:cpu_usage[1m])

avg_over_time(node:cls:cpu_usage[5m])

avg_over_time(node:cls:cpu_usage[15m])

avg_over_time(node:env:cpu_usage[1m])

avg_over_time(node:env:cpu_usage[5m])

avg_over_time(node:env:cpu_usage[15m])

rate(node_schedstat_timeslices_total[1m])

sum without (cpu) (node:cpu:sched_timeslices)

rate(node_schedstat_waiting_seconds_total[1m])

avg without (cpu) (node:cpu:sched_wait_rate1m)

rate(node_forks_total[1m])

rate(node_intr_total[1m])

rate(node_context_switches_total{}[1m])

node_load1 / node:ins:cpu_count

node_load5 / node:ins:cpu_count

node_load15 / node:ins:cpu_count

sum by (cls, job) (node_load1) / node:cls:cpu_count

sum by (cls, job) (node_load5) / node:cls:cpu_count

sum by (cls, job) (node_load15) / node:cls:cpu_count

sum by (job) (node:cls:stdload1) / node:env:cpu_count

sum by (job) (node:cls:stdload5) / node:env:cpu_count

sum by (job) (node:cls:stdload15) / node:env:cpu_count

node_memory_KernelStack_bytes +
node_memory_VmallocUsed_bytes +
node_memory_HardwareCorrupted_bytes +
node_memory_PageTables_bytes +
node_memory_Slab_bytes +
node_memory_PageTables_bytes

node_memory_MemTotal_bytes -
(node_memory_KernelStack_bytes +
node_memory_VmallocUsed_bytes +
node_memory_HardwareCorrupted_bytes +
node_memory_PageTables_bytes +
node_memory_Slab_bytes +
node_memory_PageTables_bytes) - 
node_memory_Buffers_bytes -
(node_memory_Hugepagesize_bytes * node_memory_HugePages_Total) -
(node_memory_Cached_bytes - node_memory_Mapped_bytes ) -
node_memory_MemFree_bytes

node_memory_MemAvailable_bytes / node_memory_MemTotal_bytes

1 - node_memory_MemAvailable_bytes / node_memory_MemTotal_bytes

1 - node_memory_SwapFree_bytes / node_memory_SwapTotal_bytes

node_memory_Committed_AS_bytes / node_memory_CommitLimit_bytes

1 - sum by (cls, job) (node_memory_MemAvailable_bytes) / sum by (cls, job) (node_memory_MemTotal_bytes)

sum by (job) (node_memory_MemTotal_bytes)

sum by (job) (node_memory_MemAvailable_bytes)

1 - node:env:mem_avail / node:env:mem_total

rate(node_vmstat_pgfault[1m])

rate(node_vmstat_pgmajfault[1m])

rate(node_vmstat_pgpgin[1m])

rate(node_vmstat_pgpgout[1m])

rate(node_vmstat_pswpin[1m])

rate(node_vmstat_pswpout[1m])

rate(node_disk_io_time_seconds_total{device=~"[a-z0-9A-Z-_]+"}[1m])

rate(node_disk_io_time_weighted_seconds_total{device=~"[a-z0-9A-Z-_]+"}[1m])

rate(node_disk_reads_completed_total{device=~"[a-z0-9A-Z-_]+"}[1m])

sum without (device) (node:dev:disk_reads_rate1m)

sum by (cls, job) (node:ins:disk_reads_rate1m)

rate(node_disk_writes_completed_total{device=~"[a-z0-9A-Z-_]+"}[1m])

sum without (device) (node:dev:disk_writes_rate1m)

sum by (cls, job) (node:ins:disk_writes_rate1m)

rate(node_disk_reads_merged_total{device=~"[a-z0-9A-Z-_]+"}[1m])

sum without (device) (node:dev:disk_mreads_rate1m)

sum by (cls, job) (node:ins:disk_mreads_rate1m)

rate(node_disk_writes_merged_total{device=~"[a-z0-9A-Z-_]+"}[1m])

sum without (device) (node:dev:disk_mwrites_rate1m)

sum by (cls, job) (node:ins:disk_mwrites_rate1m)

node:dev:disk_reads_rate1m + node:dev:disk_writes_rate1m

node:ins:disk_reads_rate1m + node:ins:disk_writes_rate1m

node:cls:disk_reads_rate1m + node:cls:disk_writes_rate1m

node:dev:disk_mreads_rate1m / (node:dev:disk_reads_rate1m + node:dev:disk_mreads_rate1m)

node:ins:disk_mreads_rate1m / (node:ins:disk_reads_rate1m + node:ins:disk_mreads_rate1m)

node:cls:disk_mreads_rate1m / (node:cls:disk_reads_rate1m + node:cls:disk_mreads_rate1m)

node:dev:disk_mwrites_rate1m / (node:dev:disk_writes_rate1m + node:dev:disk_mwrites_rate1m)

node:ins:disk_mwrites_rate1m / (node:ins:disk_writes_rate1m + node:ins:disk_mwrites_rate1m)

node:cls:disk_mwrites_rate1m / (node:cls:disk_writes_rate1m + node:cls:disk_mwrites_rate1m)

rate(node_disk_read_bytes_total{device=~"[a-z0-9A-Z-_]+"}[1m])

sum without (device) (node:dev:disk_read_bytes_rate1m)

sum by (cls, job) (node:ins:disk_read_bytes_rate1m)

rate(node_disk_written_bytes_total{device=~"[a-z0-9A-Z-_]+"}[1m])

sum without (device) (node:dev:disk_write_bytes_rate1m)

sum by (cls, job) (node:ins:disk_write_bytes_rate1m)

node:dev:disk_read_bytes_rate1m + node:dev:disk_write_bytes_rate1m

node:ins:disk_read_bytes_rate1m + node:ins:disk_write_bytes_rate1m

node:cls:disk_read_bytes_rate1m + node:cls:disk_write_bytes_rate1m

rate(node_disk_read_time_seconds_total{device=~"[a-z0-9A-Z-_]+"}[1m])

rate(node_disk_write_time_seconds_total{device=~"[a-z0-9A-Z-_]+"}[1m])

node:dev:disk_read_time_rate1m + node:dev:disk_write_time_rate1m

node:dev:disk_read_time_rate1m / node:dev:disk_reads_rate1m

node:dev:disk_write_time_rate1m / node:dev:disk_writes_rate1m

node:dev:disk_io_time_rate1m / node:dev:disk_iops_1m

node:dev:disk_read_bytes_rate1m / node:dev:disk_reads_rate1m

node:dev:disk_write_bytes_rate1m / node:dev:disk_writes_rate1m

node:dev:disk_io_bytes_rate1m / node:dev:disk_iops_1m

node_filesystem_free_bytes{fstype!~"(n|root|tmp)fs.*"}

node_filesystem_avail_bytes{fstype!~"(n|root|tmp)fs.*"}

node_filesystem_size_bytes{fstype!~"(n|root|tmp)fs.*"}

sum without(device,mountpoint,fstype) (node:fs:free_bytes{fstype!~"fuse.juicefs"})

sum without(device,mountpoint,fstype) (node:fs:avail_bytes{fstype!~"fuse.juicefs"})

sum without(device,mountpoint,fstype) (node:fs:size_bytes{fstype!~"fuse.juicefs"})

sum by (job,cls) (node:ins:free_bytes)

sum by (job,cls) (node:ins:avail_bytes)

sum by (job,cls) (node:ins:size_bytes)

sum by(job) (node:ins:free_bytes)

sum by(job) (node:ins:avail_bytes)

sum by(job) (node:ins:size_bytes)

1 - node:fs:avail_bytes / node:fs:size_bytes

1 - node:cls:avail_bytes / node:cls:size_bytes

1 - node:env:avail_bytes / node:env:size_bytes

max without (device, fstype, mountpoint) (node:fs:space_usage)

max by (job, cls) (node:ins:space_usage_max)

max by (job, device, fstype, mountpoint) (node:fs:space_usage)

deriv(node:fs:avail_bytes[1h])

clamp(node:fs:avail_bytes / - node:fs:space_deriv1h, -1, 126230400)

predict_linear(node:fs:avail_bytes[1h], 86400)

node_filesystem_files_free{fstype!~"(n|root|tmp)fs.*"}

node_filesystem_files{fstype!~"(n|root|tmp)fs.*"}

node:fs:inode_total - node:fs:inode_free

1 - (node:fs:inode_free / node:fs:inode_total)

sum without (fstype, device, mountpoint) (node:fs:inode_used{fstype!~"fuse.juicefs"}) /
sum without (fstype, device, mountpoint) (node:fs:inode_total{fstype!~"fuse.juicefs"})

node_filefd_allocated / node_filefd_maximum

rate(node_filefd_allocated[1m])

rate(node_network_transmit_packets_total[1m])

sum without (device) (node:dev:network_tx_pps1m{device!~"lo|bond.*"})

sum by (cls, job) (node:ins:network_tx_pps1m)

rate(node_network_receive_packets_total[1m])

sum without (device) (node:dev:network_rx_pps1m{device!~"lo|bond.*"})

sum by (cls, job) (node:ins:network_rx_pps1m)

rate(node_network_transmit_bytes_total[1m])

sum without (device) (node:dev:network_tx_bytes_rate1m{device!~"lo|bond.*"})

sum by (cls, job) (node:ins:network_tx_bytes_rate1m)

rate(node_network_receive_bytes_total[1m])

sum without (device) (node:dev:network_rx_bytes_rate1m{device!~"lo|bond.*"})

sum by (cls, job) (node:ins:network_rx_bytes_rate1m)

node:dev:network_tx_bytes_rate1m + node:dev:network_rx_bytes_rate1m

node:ins:network_tx_bytes_rate1m + node:ins:network_rx_bytes_rate1m

node:cls:network_tx_bytes_rate1m + node:cls:network_rx_bytes_rate1m

rate(node_netstat_Tcp_InSegs[1m])

rate(node_netstat_Tcp_OutSegs[1m])

rate(node_netstat_Tcp_RetransSegs[1m])

node:ins:tcp_insegs_rate1m + node:ins:tcp_outsegs_rate1m

node:ins:tcp_retranssegs_rate1m / node:ins:tcp_outsegs_rate1m

node_netstat_TcpExt_ListenOverflows +
node_netstat_TcpExt_ListenDrops +
node_netstat_Tcp_InErrs

rate(node:ins:tcp_error[1m])

rate(node_netstat_Tcp_PassiveOpens[1m])

rate(node_netstat_Tcp_ActiveOpens[1m])

rate(node_netstat_Tcp_AttemptFails[1m])

rate(node_netstat_Tcp_EstabResets[1m])

rate(node_netstat_TcpExt_ListenOverflows[1m])

rate(node_netstat_TcpExt_ListenDrops[1m])

rate(node_netstat_Udp_InDatagrams{}[1m])

rate(node_netstat_Udp_OutDatagrams[1m])

abs(node_timex_offset_seconds)

max by (cls, job) (node:ins:time_drift)

max by (cls, job) (node_ntp_offset_seconds) - min by (cls, job) (node_ntp_offset_seconds)

(100 - haproxy_process_idle_time_percent) / 100

avg by (cls, job) (haproxy:ins:usage)

time() - haproxy_process_start_time_seconds

sum by (ins,proxy) (haproxy_server_status{state="UP", proxy!='stats'}) > bool 0

sum by (cls,proxy) (haproxy_server_status{state="UP", proxy!='stats'}) > bool 0

sum without (list) (pgbouncer_list_items{list="pools"})

sum without (list) (pgbouncer_list_items{list="databases"})

sum without (list) (pgbouncer_list_items{list="users"})

sum without (list) (pgbouncer_list_items{list="login_clients"})

sum without (list) (pgbouncer_list_items{list="free_clients"})

sum without (list) (pgbouncer_list_items{list="used_clients"})

sum without (list) (pgbouncer_list_items{list="free_servers"})

pgbouncer_database_pool_size{datname!="pgbouncer"} + pgbouncer_database_reserve_pool{datname!="pgbouncer"}

pgbouncer_database_pool_size{datname!="pgbouncer"}

pgbouncer_database_reserve_pool{datname!="pgbouncer"}

pgbouncer_database_current_connections{datname!="pgbouncer"}

pgbouncer_database_paused{datname!="pgbouncer"}

pgbouncer_database_disabled{datname!="pgbouncer"}

pgbouncer:db:pool_current_conn / pgbouncer:db:pool_size

pgbouncer:db:pool_current_conn / pgbouncer:db:pool_max_conn

pgbouncer_pool_maxwait{datname!="pgbouncer"} + pgbouncer_pool_maxwait_us{datname!="pgbouncer"} / 1000000

max without(user, pool_mode) (pgbouncer:pool:maxwait)

max without(datname) (pgbouncer:db:maxwait)

max by (job, cls) (pgbouncer:ins:maxwait)

max by (job) (pgbouncer:cls:maxwait)

pgbouncer_pool_active_servers{datname!="pgbouncer"}

sum without(user, pool_mode) (pgbouncer:pool:active_servers)

sum without(datname) (pgbouncer:db:active_servers)

sum by (job, cls) (pgbouncer:ins:active_servers)

sum by (job) (pgbouncer:cls:active_servers)

pgbouncer_pool_idle_servers{datname!="pgbouncer"}

max without(user, pool_mode) (pgbouncer:pool:idle_servers)

sum without(datname) (pgbouncer:db:idle_servers)

sum by (job, cls) (pgbouncer:ins:idle_servers)

sum by (job) (pgbouncer:cls:idle_servers)

pgbouncer_pool_used_servers{datname!="pgbouncer"}

max without(user, pool_mode) (pgbouncer:pool:used_servers)

sum without(datname) (pgbouncer:db:used_servers)

sum by (job, cls) (pgbouncer:ins:used_servers)

sum by (job) (pgbouncer:cls:used_servers)

pgbouncer_pool_tested_servers{datname!="pgbouncer"}

max without(user, pool_mode) (pgbouncer:pool:tested_servers)

sum without(datname) (pgbouncer:db:tested_servers)

sum by (job, cls) (pgbouncer:ins:tested_servers)

sum by (job) (pgbouncer:cls:tested_servers)

pgbouncer_pool_login_servers{datname!="pgbouncer"}

max without(user, pool_mode) (pgbouncer:pool:login_servers)

sum without(datname) (pgbouncer:db:login_servers)

sum by (job, cls) (pgbouncer:ins:login_servers)

sum by (job) (pgbouncer:cls:login_servers)

pgbouncer_pool_active_clients{datname!="pgbouncer"}

max without(user, pool_mode) (pgbouncer:pool:active_clients)

sum without(datname) (pgbouncer:db:active_clients)

sum by (job, cls) (pgbouncer:ins:active_clients)

sum by (job) (pgbouncer:cls:active_clients)

pgbouncer_pool_waiting_clients{datname!="pgbouncer"}

max without(user, pool_mode) (pgbouncer:pool:waiting_clients)

sum without(datname) (pgbouncer:db:waiting_clients)

sum by (job, cls) (pgbouncer:ins:waiting_clients)

sum by (job) (pgbouncer:cls:waiting_clients)

rate(pgbouncer_stat_total_sent{datname!="pgbouncer"}[1m])

sum without (datname) (pgbouncer:db:tx_rate1m)

sum by (job, cls) (pgbouncer:ins:tx_rate1m)

sum by (job) (pgbouncer:cls:tx_rate1m)

rate(pgbouncer_stat_total_received{datname!="pgbouncer"}[1m])

sum without (datname) (pgbouncer:db:rx_rate1m)

sum by (job, cls) (pgbouncer:ins:rx_rate1m)

sum by (job) (pgbouncer:cls:rx_rate1m)

rate(pgbouncer_stat_total_xact_count{datname!="pgbouncer"}[1m])

sum without(datname) (pgbouncer:db:xact_rate1m)

sum by (job, cls) (pgbouncer:ins:xact_rate1m)

sum by (job) (pgbouncer:cls:xact_rate1m)

rate(pgbouncer_stat_total_xact_count{datname!="pgbouncer"}[5m])

sum without(datname) (pgbouncer:db:xact_rate5m)

sum by (job, cls) (pgbouncer:ins:xact_rate5m)

sum by (job) (pgbouncer:cls:xact_rate5m)

rate(pgbouncer_stat_total_xact_count{datname!="pgbouncer"}[15m])

sum without(datname) (pgbouncer:db:xact_rate15m)

sum by (job, cls) (pgbouncer:ins:xact_rate15m)

sum by (job) (pgbouncer:cls:xact_rate15m)

rate(pgbouncer_stat_total_xact_time{datname!="pgbouncer"}[1m])

sum without(datname) (pgbouncer:db:xact_time_rate1m)

sum by (job, cls) (pgbouncer:ins:xact_time_rate1m)

sum by (job) (pgbouncer:cls:xact_time_rate1m)

rate(pgbouncer_stat_total_xact_time{datname!="pgbouncer"}[5m])

sum without(datname) (pgbouncer:db:xact_time_rate5m)

sum by (job, cls) (pgbouncer:ins:xact_time_rate5m)

sum by (job) (pgbouncer:cls:xact_time_rate5m)

rate(pgbouncer_stat_total_xact_time{datname!="pgbouncer"}[15m])

sum without(datname) (pgbouncer:db:xact_time_rate15m)

sum by (job, cls) (pgbouncer:ins:xact_time_rate15m)

sum by (job) (pgbouncer:cls:xact_time_rate15m)

pgbouncer:db:xact_time_rate1m / pgbouncer:db:xact_rate1m

pgbouncer:ins:xact_time_rate1m / pgbouncer:ins:xact_rate1m

pgbouncer:cls:xact_time_rate1m / pgbouncer:cls:xact_rate1m

pgbouncer:env:xact_time_rate1m / pgbouncer:env:xact_rate1m

pgbouncer:db:xact_time_rate5m / pgbouncer:db:xact_rate5m

pgbouncer:ins:xact_time_rate5m / pgbouncer:ins:xact_rate5m

pgbouncer:cls:xact_time_rate5m / pgbouncer:cls:xact_rate5m

pgbouncer:env:xact_time_rate5m / pgbouncer:env:xact_rate5m

pgbouncer:db:xact_time_rate15m / pgbouncer:db:xact_rate15m

pgbouncer:ins:xact_time_rate15m / pgbouncer:ins:xact_rate15m

pgbouncer:cls:xact_time_rate15m / pgbouncer:cls:xact_rate15m

pgbouncer:env:xact_time_rate15m / pgbouncer:env:xact_rate15m

rate(pgbouncer_stat_total_query_count{datname!="pgbouncer"}[1m])

sum without(datname) (pgbouncer:db:query_rate1m)

sum by (job, cls) (pgbouncer:ins:query_rate1m)

sum by (job) (pgbouncer:cls:query_rate1m)

rate(pgbouncer_stat_total_query_count{datname!="pgbouncer"}[5m])

sum without(datname) (pgbouncer:db:query_rate5m)

sum by (job, cls) (pgbouncer:ins:query_rate5m)

sum by (job) (pgbouncer:cls:query_rate5m)

rate(pgbouncer_stat_total_query_count{datname!="pgbouncer"}[15m])

sum without(datname) (pgbouncer:db:query_rate15m)

sum by (job, cls) (pgbouncer:ins:query_rate15m)

sum by (job) (pgbouncer:cls:query_rate15m)

rate(pgbouncer_stat_total_query_time{datname!="pgbouncer"}[1m])

sum without(datname) (pgbouncer:db:query_time_rate1m)

sum by (job, cls) (pgbouncer:ins:query_time_rate1m)

sum by (job) (pgbouncer:cls:query_time_rate1m)

rate(pgbouncer_stat_total_query_time{datname!="pgbouncer"}[5m])

sum without(datname) (pgbouncer:db:query_time_rate5m)

sum by (job, cls) (pgbouncer:ins:query_time_rate5m)

sum by (job) (pgbouncer:cls:query_time_rate5m)