Basic Metrics: Node Exporter Metrics

This section describes the types, names, and meanings of metrics reported by Node Exporter to AOM.

**Table 1** Metrics of containers running in CCE or on-premises Kubernetes clusters
Job	Metric	Description
node-exporter	node_filesystem_size_bytes	Consumed space of a file system
	node_filesystem_readonly	Read-only file system
	node_filesystem_free_bytes	Remaining space of a file system
	node_filesystem_avail_bytes	File system space that is available for use
	node_cpu_seconds_total	Seconds each CPU spent doing each type of work
	node_network_receive_bytes_total	Total amount of received data
	node_network_receive_errs_total	Cumulative number of errors encountered during reception
	node_network_transmit_bytes_total	Total amount of transmitted data
	node_network_receive_packets_total	Cumulative number of packets received
	node_network_transmit_drop_total	Cumulative number of dropped packets during transmission
	node_network_transmit_errs_total	Cumulative number of errors encountered during transmission
	node_network_up	NIC status
	node_network_transmit_packets_total	Cumulative number of packets transmitted
	node_network_receive_drop_total	Cumulative number of packets dropped during reception
	go_gc_duration_seconds	This value is obtained by calling the debug.ReadGCStats() function. When this function is called, the PauseQuantile field of the GCStats structure is set to 5. In this way, the function returns 5 GC pause time percentiles (the minimum percentile, 25%, 50%, 75%, and maximum percentile). Then, the Prometheus Go client creates a summary metric based on the returned GC pause time percentile, NumGC, and PauseTotal.
	node_load5	5-minute average CPU load
	node_filefd_allocated	Allocated file descriptors
	node_exporter_build_info	Node exporter build information
	node_disk_written_bytes_total	Number of bytes that are written
	node_disk_writes_completed_total	Number of writes completed
	node_disk_write_time_seconds_total	Number of seconds spent by all writes
	node_nf_conntrack_entries	Number of currently allocated flow entries for connection tracking
	node_nf_conntrack_entries_limit	Maximum size of a connection tracking table
	node_processes_max_processes	PID limit value
	node_processes_pids	Number of PIDs
	node_sockstat_TCP_alloc	Number of allocated TCP sockets
	node_sockstat_TCP_inuse	Number of TCP sockets in use
	node_sockstat_TCP_tw	Number of TCP sockets in the TIME_WAIT state
	node_timex_offset_seconds	Time offset
	node_timex_sync_status	Synchronization status of node clocks
	node_uname_info	Labeled system information as provided by the uname system call
	node_vmstat_pgfault	Number of page faults the system has made per second in /proc/vmstat
	node_vmstat_pgmajfault	Number of major faults per second in /proc/vmstat
	node_vmstat_pgpgin	Number of page in between main memory and block device in /proc/vmstat
	node_vmstat_pgpgout	Number of page out between main memory and block device in /proc/vmstat
	node_disk_reads_completed_total	Number of reads completed
	node_disk_read_time_seconds_total	Number of seconds spent by all reads
	process_cpu_seconds_total	The value is obtained based on the utime parameter (the number of ticks executed by the Go process in user mode) and the stime parameter (the number of ticks executed by the Go process in kernel mode, for example, during system invocation). Unit: jiffies, which measure the tick time between two system timer interruptions. process_cpu_seconds_total = (utime + stime)/USER_HZ Based on the preceding formula, you can obtain the total time (unit: seconds) for a process to run on the OS.
	node_disk_read_bytes_total	Number of bytes that are read
	node_disk_io_time_weighted_seconds_total	The weighted number of seconds spent doing I/Os
	node_disk_io_time_seconds_total	Total seconds spent doing I/Os
	node_disk_io_now	Number of I/Os in progress
	node_context_switches_total	Number of context switches
	node_boot_time_seconds	Node boot time
	process_resident_memory_bytes	Resident set size (RSS), which is the memory actually used by a process. It includes the shared memory, but does not include the allocated but unused memory or swapped-out memory.
	node_intr_total	Number of interruptions that occurred
	node_load1	1-minute average CPU load
	go_goroutines	This value is obtained by calling runtime.NumGoroutine() and calculated based on the sched scheduler structure and global allglen variable. Fields in the sched structure may change concurrently. Therefore, the system checks whether the calculated value is less than 1. If the value is less than 1, the system returns 1.
	scrape_duration_seconds	Time spent on collecting information about the scrape target
	node_load15	15-minute average CPU load
	scrape_samples_post_metric_relabeling	Number of remaining samples after metrics are relabeled
	node_netstat_Tcp_PassiveOpens	Number of TCP connections that directly change from the LISTEN state to the SYN-RCVD state
	scrape_samples_scraped	Number of samples scraped
	node_netstat_Tcp_CurrEstab	Number of TCP connections in the ESTABLISHED or CLOSE-WAIT state
	scrape_series_added	Number of series added to the scrape target
	node_netstat_Tcp_ActiveOpens	Number of TCP connections that directly change from the CLOSED state to the SYN-SENT state
	node_memory_MemTotal_bytes	Total memory of a node
	node_memory_MemFree_bytes	Free memory of a node
	node_memory_MemAvailable_bytes	Available memory of a node
	node_memory_Cached_bytes	Memory for the node page cache
	up	Scrape target status
	node_memory_Buffers_bytes	Memory of the node buffer