微网站建设包含哪些内容,信息流优化师面试常见问题,进入百度app,网站主色调简介怎么说目录 指标详解常用指标 指标详解
指标描述node_arp_entriesARP#xff08;Address Resolution Protocol#xff09;表中的条目数量#xff0c;用于将IP地址映射到MAC地址。node_boot_time_seconds系统启动时间的Unix时间戳#xff0c;表示从1970年1月1日以来的秒数。node…目录 指标详解常用指标
指标详解
指标描述node_arp_entriesARPAddress Resolution Protocol表中的条目数量用于将IP地址映射到MAC地址。node_boot_time_seconds系统启动时间的Unix时间戳表示从1970年1月1日以来的秒数。node_context_switches_total发生的上下文切换的总数。node_cooling_device_cur_state冷却设备的当前状态。通常用于监视处理器散热器的状态。node_cooling_device_max_state冷却设备的最大状态。node_cpu_guest_seconds_total虚拟CPU的用户模式时间以秒为单位。node_cpu_seconds_totalCPU核心的用户、系统和其他模式的CPU时间以秒为单位。node_disk_device_mapper_info关于设备映射器device mapper的信息如名称、大小、状态等。node_disk_filesystem_info文件系统的信息包括名称、容量、类型和挂载点。node_disk_info磁盘的信息包括名称、容量、类型和状态。node_disk_io_now当前的磁盘I/O操作数。node_disk_io_time_seconds_total磁盘I/O操作的总时间以秒为单位。node_disk_io_time_weighted_seconds_total加权磁盘I/O操作的总时间以秒为单位。node_disk_read_bytes_total读取的总字节数。node_disk_read_time_seconds_total磁盘读取操作的总时间以秒为单位。node_disk_reads_completed_total已完成的磁盘读取操作的总数。node_disk_reads_merged_total合并的磁盘读取操作的总数。node_disk_write_time_seconds_total磁盘写入操作的总时间以秒为单位。node_disk_writes_completed_total已完成的磁盘写入操作的总数。node_disk_writes_merged_total合并的磁盘写入操作的总数。node_disk_written_bytes_total写入的总字节数。node_dmi_info有关系统DMIDesktop Management Interface的信息包括制造商、型号和序列号。node_entropy_available_bits系统熵池中可用的比特数。node_entropy_pool_size_bits系统熵池的大小以比特为单位。node_exporter_build_infoNode Exporter的版本和构建信息。node_filefd_allocated已分配的文件描述符数量。node_filefd_maximum系统允许的最大文件描述符数量。node_filesystem_avail_bytes文件系统可用空间的字节数。node_filesystem_device_error文件系统设备是否出错通常是0或1。node_filesystem_files文件系统中的文件总数。node_filesystem_files_free文件系统中的可用文件数。node_filesystem_free_bytes文件系统的空闲字节数。node_filesystem_readonly文件系统是否为只读通常是0或1。node_filesystem_size_bytes文件系统的总字节数。node_forks_total系统中的总进程分支数。node_intr_total中断的总数。node_load1系统的1分钟平均负载。node_load15系统的15分钟平均负载。node_load5系统的5分钟平均负载。node_memory_Active_anon_bytes活跃的匿名内存的字节数指的是进程正在使用的内存而不是文件系统缓存。node_memory_Active_bytes活跃内存的字节数包括匿名内存和文件系统缓存。node_memory_Active_file_bytes活跃的文件系统缓存内存的字节数。node_memory_AnonHugePages_bytes匿名巨大页面的字节数。node_memory_AnonPages_bytes匿名内存页面的字节数。node_memory_Bounce_bytes内存反弹缓冲区的字节数。node_memory_Buffers_bytes系统缓冲区占用的内存的字节数。node_memory_Cached_bytes文件系统缓存的内存的字节数。node_memory_CmaFree_bytesCMAContiguous Memory Allocator空闲内存的字节数。node_memory_CmaTotal_bytesCMA总内存的字节数。node_memory_CommitLimit_bytes系统内存提交限制的字节数。node_memory_Committed_AS_bytes已分配的内存的字节数。node_memory_DirectMap1G_bytes直接映射到1GB内存的字节数。node_memory_DirectMap2M_bytes直接映射到2MB内存的字节数。node_memory_DirectMap4k_bytes直接映射到4KB内存的字节数。node_memory_Dirty_bytes脏页面的字节数。node_memory_HardwareCorrupted_bytes硬件损坏内存的字节数。node_memory_HugePages_Free空闲的巨大页面数。node_memory_HugePages_Rsvd已保留的巨大页面数。node_memory_HugePages_Surp额外的巨大页面数。node_memory_HugePages_Total总的巨大页面数。node_memory_Hugepagesize_bytes巨大页面的页面大小以字节数为单位。node_memory_Inactive_anon_bytes非活跃的匿名内存的字节数。node_memory_Inactive_bytes非活跃内存的字节数。node_memory_Inactive_file_bytes非活跃的文件系统缓存内存的字节数。node_memory_KernelStack_bytes内核栈内存的字节数。node_memory_Mapped_bytes映射内存的字节数。node_memory_MemAvailable_bytes系统可用内存的字节数。node_memory_MemFree_bytes系统可用空闲内存的字节数。node_memory_MemTotal_bytes系统物理内存的总字节数。node_memory_Mlocked_bytes锁定的内存的字节数。node_memory_NFS_Unstable_bytesNFSNetwork File System不稳定内存的字节数。node_memory_PageTables_bytes页表内存的字节数。node_memory_Percpu_bytes每个CPU核心的分配内存的字节数。node_memory_SReclaimable_bytes可回收内存的字节数。node_memory_SUnreclaim_bytes不可回收内存的字节数。node_memory_Shmem_bytes共享内存的字节数。node_memory_Slab_bytes内核SLAB分配内存的字节数。node_memory_SwapCached_bytes交换缓存内存的字节数。node_memory_SwapFree_bytes交换空闲内存的字节数。node_memory_SwapTotal_bytes交换总内存的字节数。node_memory_Unevictable_bytes无法剥夺的内存的字节数。node_memory_VmallocChunk_bytes虚拟内存分配的块大小的字节数。node_memory_VmallocTotal_bytes总的虚拟内存的字节数。node_memory_VmallocUsed_bytes已使用的虚拟内存的字节数。node_memory_WritebackTmp_bytes写回临时内存的字节数。node_memory_Writeback_bytes写回内存的字节数。node_netstat_Icmp6_InErrorsIPv6 ICMP接收错误的次数。node_netstat_Icmp6_InMsgsIPv6 ICMP接收的消息数量。node_netstat_Icmp6_OutMsgsIPv6 ICMP传输的消息数量。node_netstat_Icmp_InErrorsIPv4 ICMP接收错误的次数。node_netstat_Icmp_InMsgsIPv4 ICMP接收的消息数量。node_netstat_Icmp_OutMsgsIPv4 ICMP传输的消息数量。node_netstat_Ip6_InOctetsIPv6接收的字节数。node_netstat_Ip6_OutOctetsIPv6传输的字节数。node_netstat_IpExt_InOctets扩展IP接收的字节数。node_netstat_IpExt_OutOctets扩展IP传输的字节数。node_netstat_Ip_ForwardingIP包转发状态通常为0或1。node_netstat_TcpExt_ListenDropsTCP监听队列中的连接被拒绝的次数。node_netstat_TcpExt_ListenOverflowsTCP监听队列溢出的次数。node_netstat_TcpExt_SyncookiesFailedTCP SYN cookie处理失败的次数。node_netstat_TcpExt_SyncookiesRecvTCP SYN cookie接收的次数。node_netstat_TcpExt_SyncookiesSentTCP SYN cookie发送的次数。node_netstat_TcpExt_TCPSynRetransTCP SYN重传的次数。node_netstat_TcpExt_TCPTimeoutsTCP连接超时的次数。node_netstat_Tcp_ActiveOpensTCP主动打开连接的次数。node_netstat_Tcp_CurrEstab当前TCP连接的数量。node_netstat_Tcp_InErrsTCP接收错误的次数。node_netstat_Tcp_InSegsTCP接收的数据包的数量。node_netstat_Tcp_OutRstsTCP传输的复位包的数量。node_netstat_Tcp_OutSegsTCP传输的数据包的数量。node_netstat_Tcp_PassiveOpensTCP被动打开连接的次数。node_netstat_Tcp_RetransSegsTCP重传的数据包的数量。node_netstat_Udp6_InDatagramsIPv6 UDP接收的数据包的数量。node_netstat_Udp6_InErrorsIPv6 UDP接收错误的次数。node_netstat_Udp6_NoPortsIPv6 UDP接收到的数据包但没有适当的端口可用。node_netstat_Udp6_OutDatagramsIPv6 UDP传输的数据包的数量。node_netstat_Udp6_RcvbufErrorsIPv6 UDP接收缓冲区错误的次数。node_netstat_Udp6_SndbufErrorsIPv6 UDP发送缓冲区错误的次数。node_netstat_UdpLite6_InErrorsIPv6 UDP Lite接收错误的次数。node_netstat_UdpLite_InErrorsUDP Lite接收错误的次数。node_netstat_Udp_InDatagramsUDP接收的数据包的数量。node_netstat_Udp_InErrorsUDP接收错误的次数。node_netstat_Udp_NoPortsUDP接收到的数据包但没有适当的端口可用。node_netstat_Udp_OutDatagramsUDP传输的数据包的数量。node_netstat_Udp_RcvbufErrorsUDP接收缓冲区错误的次数。node_netstat_Udp_SndbufErrorsUDP发送缓冲区错误的次数。node_network_address_assign_type网络地址分配类型。node_network_carrier网络接口是否连接到网络通常是0或1。node_network_carrier_changes_total网络接口连接状态更改的总数。node_network_device_id网络接口的唯一标识符。node_network_dormant网络接口是否处于休眠状态通常是0或1。node_network_flags网络接口的标志和属性。node_network_iface_id网络接口的ID。node_network_iface_link网络接口是否链接通常是0或1。node_network_iface_link_mode网络接口的链接模式。node_network_info有关网络接口的信息如名称、MAC地址等。node_network_mtu_bytes网络接口的最大传输单元MTU大小以字节为单位。node_network_net_dev_group网络接口的网络设备组。node_network_protocol_type网络接口的协议类型。node_network_receive_bytes_total接收的总字节数。node_network_receive_compressed_total接收的压缩字节数的总数。node_network_receive_drop_total接收的丢弃数据包的总数。node_network_receive_errs_total接收的错误数据包的总数。node_network_receive_fifo_total接收的FIFO数据包的总数。node_network_receive_frame_total接收的帧数据包的总数。node_network_receive_multicast_total接收的多播数据包的总数。node_network_receive_nohandler_total没有处理的接收数据包的总数。node_network_receive_packets_total接收的数据包的总数。node_network_speed_bytes网络接口的速度以字节为单位。node_network_transmit_bytes_total传输的总字节数。node_network_transmit_carrier_total传输的载波数据包的总数。node_network_transmit_colls_total传输的碰撞数据包的总数。node_network_transmit_compressed_total传输的压缩数据包的总数。node_network_transmit_drop_total传输的丢弃数据包的总数。node_network_transmit_errs_total传输的错误数据包的总数。node_network_transmit_fifo_total传输的FIFO数据包的总数。node_network_transmit_packets_total传输的数据包的总数。node_network_transmit_queue_length传输队列的长度。node_network_up网络接口是否处于活动状态通常是0或1。node_os_info操作系统的信息。node_os_version操作系统的版本信息。node_power_supply_info电源供应的信息。node_power_supply_online电源是否在线通常是0或1。node_procs_blocked被阻塞的进程数量。node_procs_running正在运行的进程数量。node_schedstat_running_seconds_totalCPU运行时间的总秒数。node_schedstat_timeslices_totalCPU时间片的总数。node_schedstat_waiting_seconds_totalCPU等待时间的总秒数。node_scrape_collector_duration_seconds采集器运行的时间秒。node_scrape_collector_success采集器是否成功运行通常是0或1。node_selinux_enabledSELinux是否启用通常是0或1。node_sockstat_FRAG6_inuseIPv6分片数据包的使用数量。node_sockstat_FRAG6_memoryIPv6分片数据包的内存占用。node_sockstat_FRAG_inuseIPv4分片数据包的使用数量。node_sockstat_FRAG_memoryIPv4分片数据包的内存占用。node_sockstat_RAW6_inuseIPv6原始套接字的使用数量。node_sockstat_RAW_inuseIPv4原始套接字的使用数量。node_sockstat_TCP6_inuseIPv6 TCP套接字的使用数量。node_sockstat_TCP_allocTCP套接字的分配数量。node_sockstat_TCP_inuseIPv4 TCP套接字的使用数量。node_sockstat_TCP_memTCP套接字内存的使用情况。node_sockstat_TCP_mem_bytesTCP套接字内存的使用情况以字节为单位。node_sockstat_TCP_orphanTCP套接字的孤立数量。node_sockstat_TCP_twTCP套接字的TIME_WAIT状态数量。node_sockstat_UDP6_inuseIPv6 UDP套接字的使用数量。node_sockstat_UDPLITE6_inuseIPv6 UDP Lite套接字的使用数量。node_sockstat_UDPLITE_inuseUDP Lite套接字的使用数量。node_sockstat_UDP_inuseIPv4 UDP套接字的使用数量。node_sockstat_UDP_memUDP套接字内存的使用情况。node_sockstat_UDP_mem_bytesUDP套接字内存的使用情况以字节为单位。node_sockstat_sockets_used使用的套接字数量。node_softnet_backlog_len软网络设备队列的长度。node_softnet_cpu_collision_totalCPU冲突的次数。node_softnet_dropped_total丢弃的数据包的总数。node_softnet_flow_limit_count_total流量限制计数的总数。node_softnet_processed_total已处理的数据包的总数。node_softnet_received_rps_total接收到的RPSReceive Packet Steering数据包的总数。node_softnet_times_squeezed_total已挤压的时间片的总数。node_textfile_scrape_error文本文件采集器的错误信息。node_time_clocksource_available_info可用时钟源的信息。node_time_clocksource_current_info当前时钟源的信息。node_time_seconds系统时间的秒数。node_time_zone_offset_seconds系统时区的偏移秒数。node_timex_estimated_error_seconds估计的时钟误差秒。node_timex_frequency_adjustment_ratio频率调整比率。node_timex_loop_time_constant时钟环回时间常数。node_timex_maxerror_seconds最大时钟误差秒。node_timex_offset_seconds时钟偏移秒。node_timex_pps_calibration_totalPPSPulse-Per-Second校准次数。node_timex_pps_error_totalPPS错误次数。node_timex_pps_frequency_hertzPPS频率赫兹。node_timex_pps_jitter_secondsPPS抖动秒。node_timex_pps_jitter_totalPPS抖动总数。node_timex_pps_shift_secondsPPS偏移秒。node_timex_pps_stability_exceeded_totalPPS稳定性超过阈值的次数。node_timex_pps_stability_hertzPPS稳定性赫兹。node_timex_status时钟状态信息。node_timex_sync_status同步状态信息。node_timex_tai_offset_secondsTAIInternational Atomic Time时钟偏移秒。node_timex_tick_seconds时钟滴答秒。node_udp_queuesUDP队列信息。node_vmstat_pgfault虚拟内存统计中的页错误次数。node_vmstat_pgmajfault虚拟内存统计中的主要页错误次数。node_vmstat_pgpgin虚拟内存统计中的页入操作次数。node_vmstat_pgpgout虚拟内存统计中的页出操作次数。node_vmstat_pswpin虚拟内存统计中的页交换入次数。node_vmstat_pswpout虚拟内存统计中的页交换出次数。node_xfs_allocation_btree_compares_totalXFS分配B树比较次数的总数。node_xfs_allocation_btree_lookups_totalXFS分配B树查找次数的总数。node_xfs_allocation_btree_records_deleted_totalXFS分配B树记录删除次数的总数。node_xfs_allocation_btree_records_inserted_totalXFS分配B树记录插入次数的总数。node_xfs_block_map_btree_compares_totalXFS块映射B树比较次数的总数。node_xfs_block_map_btree_lookups_totalXFS块映射B树查找次数的总数。node_xfs_block_map_btree_records_deleted_totalXFS块映射B树记录删除次数的总数。node_xfs_block_map_btree_records_inserted_totalXFS块映射B树记录插入次数的总数。node_xfs_block_mapping_extent_list_compares_totalXFS块映射范围列表比较次数的总数。node_xfs_block_mapping_extent_list_deletions_totalXFS块映射范围列表删除次数的总数。node_xfs_block_mapping_extent_list_insertions_totalXFS块映射范围列表插入次数的总数。node_xfs_block_mapping_extent_list_lookups_totalXFS块映射范围列表查找次数的总数。node_xfs_block_mapping_reads_totalXFS块映射读取次数的总数。node_xfs_block_mapping_unmaps_totalXFS块映射取消映射次数的总数。node_xfs_block_mapping_writes_totalXFS块映射写入次数的总数。node_xfs_directory_operation_create_totalXFS目录操作创建次数的总数。node_xfs_directory_operation_getdents_totalXFS目录操作获取条目次数的总数。node_xfs_directory_operation_lookup_totalXFS目录操作查找次数的总数。node_xfs_directory_operation_remove_totalXFS目录操作删除次数的总数。node_xfs_extent_allocation_blocks_allocated_totalXFS范围分配块分配次数的总数。node_xfs_extent_allocation_blocks_freed_totalXFS范围分配块释放次数的总数。node_xfs_extent_allocation_extents_allocated_totalXFS范围分配范围分配次数的总数。node_xfs_extent_allocation_extents_freed_totalXFS范围分配范围释放次数的总数。node_xfs_inode_operation_attempts_totalXFS节点操作尝试次数的总数。node_xfs_inode_operation_attribute_changes_totalXFS节点操作属性更改次数的总数。node_xfs_inode_operation_duplicates_totalXFS节点操作重复次数的总数。node_xfs_inode_operation_found_totalXFS节点操作查找成功次数的总数。node_xfs_inode_operation_missed_totalXFS节点操作查找失败次数的总数。node_xfs_inode_operation_reclaims_totalXFS节点操作回收次数的总数。node_xfs_inode_operation_recycled_totalXFS节点操作回收再利用次数的总数。node_xfs_read_calls_totalXFS读取调用次数的总数。node_xfs_vnode_active_totalXFS活动节点数量的总数。node_xfs_vnode_allocate_totalXFS节点分配次数的总数。node_xfs_vnode_get_totalXFS获取节点次数的总数。node_xfs_vnode_hold_totalXFS保留节点次数的总数。node_xfs_vnode_reclaim_totalXFS节点回收次数的总数。node_xfs_vnode_release_totalXFS释放节点次数的总数。node_xfs_vnode_remove_totalXFS删除节点次数的总数。node_xfs_write_calls_totalXFS写入调用次数的总数。process_cpu_seconds_total进程的CPU时间总量。process_max_fds进程的最大文件描述符数。process_open_fds进程的已打开文件描述符数。process_resident_memory_bytes进程的常驻内存大小字节。process_start_time_seconds进程的启动时间秒。process_virtual_memory_bytes进程的虚拟内存大小字节。process_virtual_memory_max_bytes进程的最大虚拟内存大小字节。promhttp_metric_handler_errors_totalPrometheus HTTP指标处理器的错误次数。promhttp_metric_handler_requests_in_flightPrometheus HTTP指标处理器当前正在处理的请求数。promhttp_metric_handler_requests_totalPrometheus HTTP指标处理器的请求总数。
常用指标
1. CPU Cores 物理 CPU 的核数 cat /proc/cpuinfo| grep cpu cores| uniq
typeSinglestat
Unit: short
metrics:count(count(node_cpu_seconds_total{instance~$node,job~$job}) by (cpu))2. Total RAM 内存大小 cat /proc/meminfo | grep MemTotal
typeSinglestat
Unit: bytes
metrics:node_memory_MemTotal_bytes{instance~$node,job~$job}3. Total SWAP 交换分区的大小 cat /proc/swaps
typeSinglestat
Unit: bytes
metrics:node_memory_SwapTotal_bytes{instance~$node,job~$job}4. Total RootFS 根文件系统总空间
typeSinglestat
Unit bytes
metrics:node_filesystem_size_bytes{instance~$node,job~$job,mountpoint/,fstype!rootfs}5. System Load (1m avg) 系统一分钟内的负载 cat /proc/loadavg 第一列单核 cpu 的load小于1 表示没有等待的任务 等于1 表示系统
已经没有额外的资源跑更多进程了大于1表示进程拥堵在等待资源
typeSinglestat
Unit: short
metrics:node_load1{instance~$node,job~$job}6. Uptime 系统正常运行的时间
typeSinglestat
Unit: seconds (s)
metrics:node_time_seconds{instance~$node,job~$job} - node_boot_time_seconds{instance~$node,job~$job}node_time_seconds 当前系统时间
node_boot_time_seconds 系统启动时间
Basic CPU / Mem / Disk Gauge1. CPU Busy :收集所有 cpu 内核 busy 状态占比
type: Singlestat
Unit: perent(0-100)
(所有 cpu使用情况 - 5分钟内 cpu 空闲的平均值) / 所有 cpu使用情况
metrics:(((count(count(node_cpu_seconds_total{instance~$node,job~$job}) by (cpu))) - avg(sum by (mode)(irate(node_cpu_seconds_total{modeidle,instance~$node,job~$job}[5m])))) * 100) / count(count(node_cpu_seconds_total{instance~$node,job~$job}) by (cpu))最大值 100%2. Used RAM Memory free -m
type: Singlestat
Unit: perent(0-100)
已使用的内存占比包括Buffer缓存和Cached缓存
metrics:((node_memory_MemTotal_bytes{instance~$node,job~$job} - node_memory_MemFree_bytes{instance~$node,job~$job}) / (node_memory_MemTotal_bytes{instance~$node,job~$job} )) * 100node_memory_MemFree_bytes 空闲内存已使用的内存占比不包括Buffer缓存和Cached缓存
metrics:100 - ((node_memory_MemAvailable_bytes{instance~$node,job~$job} * 100) / node_memory_MemTotal_bytes{instance~$node,job~$job})MemAvailable: Free Buffers Cached - 不可回收的部分。不可回收部分包括共享内存段tmpfsramfs等3. Used SWAP: 交换分区使用率
type: Singlestat
Unit: perent(0-100)
metrics:((node_memory_SwapTotal_bytes{instance~$node,job~$job} - node_memory_SwapFree_bytes{instance~$node,job~$job}) / (node_memory_SwapTotal_bytes{instance~$node,job~$job} )) * 100node_memory_SwapFree_bytes 交换分区的空闲大小4. Used Root FS 根文件系统使用率
type: Singlestat
Unit: perent(0-100)
metrics:100 - ((node_filesystem_avail_bytes{instance~$node,job~$job,mountpoint/,fstype!rootfs} * 100) / node_filesystem_size_bytes{instance~$node,job~$job,mountpoint/,fstype!rootfs})node_filesystem_avail_bytes 文件系统可用空间5. CPU System Load (1m avg) 一分钟内 CPU 所有内核的平均负载率
type: Singlestat
Unit: perent(0-100)
metrics:avg(node_load1{instance~$node,job~$job}) / count(count(node_cpu_seconds_total{instance~$node,job~$job}) by (cpu)) * 100node_load1 : 系统一分钟内的负载6. CPU System Load (5m avg) 五分钟内 CPU 所有内核的平均负载率
type: Singlestat
Unit: perent(0-100)
metrics:avg(node_load5{instance~$node,job~$job}) / count(count(node_cpu_seconds_total{instance~$node,job~$job}) by (cpu)) * 100node_load5 : 指5分钟内cpu的负载
Basic CPU / Mem Graph1. CPU Basic cpu 的基本信息 /proc/stat
type: Graph
Unit: shortBusy System: cpu 处于核心态的占比
metrics:sum by (instance)(rate(node_cpu_seconds_total{modesystem,instance~$node,job~$job}[5m])) * 100Busy User: cpu 处于用户态的占比
metrics:sum by (instance)(rate(node_cpu_seconds_total{modeuser,instance~$node,job~$job}[5m])) * 100Busy Iowait: cpu 处于 io 等待的时间占比
metrics:sum by (instance)(rate(node_cpu_seconds_total{modeiowait,instance~$node,job~$job}[5m])) * 100Busy IRQs: cpu 处于中断状态占比
metrics:sum by (instance)(rate(node_cpu_seconds_total{mode~.*irq,instance~$node,job~$job}[5m])) * 100Idle: cpu 处于空闲状态占比
metrics:sum by (mode)(rate(node_cpu_seconds_total{modeidle,instance~$node,job~$job}[5m])) * 100Busy Other: cpu 处于其他状态占比(非系统状态、非用户态、非io等待状态、非空闲态、非中断状态)
metrics:sum (rate(node_cpu_seconds_total{mode!idle,mode!user,mode!system,mode!iowait,mode!irq,mode!softirq,instance~$node,job~$job}[5m])) * 1002. Memory Basic 内存基本信息
type: Graph
Unit: shortRAM Total: 内存大小
metrics:node_memory_MemTotal_bytes{instance~$node,job~$job}RAM Used: 已使用的内存大小(内存总量-空闲的内存大小-Buffer缓存和Cached缓存占的内存大小)
metrics:node_memory_MemTotal_bytes{instance~$node,job~$job} - node_memory_MemFree_bytes{instance~$node,job~$job} - (node_memory_Cached_bytes{instance~$node,job~$job} node_memory_Buffers_bytes{instance~$node,job~$job})RAM Cache Buffer: Cached缓存占的内存大小
metrics:node_memory_Cached_bytes{instance~$node,job~$job} node_memory_Buffers_bytes{instance~$node,job~$job}RAM Free: 空闲的内存大小
metrics:node_memory_MemFree_bytes{instance~$node,job~$job}SWAP Used: 已使用的交换内存的大小
metrics:(node_memory_SwapTotal_bytes{instance~$node,job~$job} - node_memory_SwapFree_bytes{instance~$node,job~$job})
交换分区的总大小- 空闲大小
Basic Net / Disk Info1. Network Traffic Basic 每个接口的基本网络信息
type: Graph
Unit: bytesrecv {{device}} 各个网络接口的下载量
recv lo: 本地环回接口
recv eth0: 以太网接口
recv docker0: docker0 网络接口
metrics:rate(node_network_receive_bytes_total{instance~$node,job~$job}[5m])trans {{device}} 各个网络接口的上传量
metrics:rate(node_network_transmit_bytes_total{instance~$node,job~$job}[5m])2. Disk Space Used Basic 所有挂载的文件系统的磁盘空间占比
type: Graph
unit: perent(0-100)
metrics:100 - ((node_filesystem_avail_bytes{instance~$node,job~$job,device!~rootfs} * 100) / node_filesystem_size_bytes{instance~$node,job~$job,device!~rootfs})
CPU Memory Net Disk1. CPU
type: Graph
Unit: short
max: 100
min: 0
Label: PercentageSystem - cpu 在内核模式下执行的进程占比
metrics:sum by (mode)(irate(node_cpu_seconds_total{modesystem,instance~$node,job~$job}[5m])) * 100User - cpu 在用户模式下执行的正常进程占比
metrics:sum by (mode)(irate(node_cpu_seconds_total{modeuser,instance~$node,job~$job}[5m])) * 100Nice - cpu 在用户模式下执行的 nice 进程占比
metrics:sum by (mode)(irate(node_cpu_seconds_total{modenice,instance~$node,job~$job}[5m])) * 100Idle - cpu 在空闲模式下的占比
metrics:sum by (mode)(irate(node_cpu_seconds_total{modeidle,instance~$node,job~$job}[5m])) * 100Iowait - cpu 在 io 等待的占比
metrics:sum by (mode)(irate(node_cpu_seconds_total{modeiowait,instance~$node,job~$job}[5m])) * 100Irq - cpu 在服务中断的占比
metrics:sum by (mode)(irate(node_cpu_seconds_total{modeirq,instance~$node,job~$job}[5m])) * 100Softirq - cpu 在服务软中断的占比
metrics:sum by (mode)(irate(node_cpu_seconds_total{modesoftirq,instance~$node,job~$job}[5m])) * 100Steal - 在 VM 中运行时其他 VM 占用的本 VM 的 cpu 的占比
metrics:sum by (mode)(irate(node_cpu_seconds_total{modesteal,instance~$node,job~$job}[5m])) * 100Guest - 运行各种 VM 使用的 CPU 占比
metrics:sum by (mode)(irate(node_cpu_seconds_total{modeguest,instance~$node,job~$job}[5m])) * 1002. Memory Stack 内存堆栈 /proc/meminfo
type: Graph
Unit: bytes
min: 0
Label: BytesApps - 用户空间应用程序使用的内存
metrics:node_memory_MemTotal_bytes{instance~$node,job~$job} - node_memory_MemFree_bytes{instance~$node,job~$job}
- node_memory_Buffers_bytes{instance~$node,job~$job} - node_memory_Cached_bytes{instance~$node,job~$job}
- node_memory_Slab_bytes{instance~$node,job~$job} - node_memory_PageTables_bytes{instance~$node,job~$job}
- node_memory_SwapCached_bytes{instance~$node,job~$job}PageTables - 用于在虚拟和物理内存地址之间映射的内存
metrics:node_memory_PageTables_bytes{instance~$node,job~$job}SwapCache - 用于跟踪已从交换区中提取出来但尚未修改的页面的内存
metrics:node_memory_SwapCached_bytes{instance~$node,job~$job}Slab - 内核用于缓存数据结构以供自己使用的内存如 inodedentry 等缓存
metrics:node_memory_Slab_bytes{instance~$node,job~$job}Cache - 频繁访问的文件数据或内容的缓存
metrics:node_memory_Cached_bytes{instance~$node,job~$job}Buffers - 块设备例如硬盘缓存
metrics:node_memory_Buffers_bytes{instance~$node,job~$job}Unused - 未使用的内存大小
metrics:node_memory_MemFree_bytes{instance~$node,job~$job}Swap - 交换分区使用的空间
metrics:(node_memory_SwapTotal_bytes{instance~$node,job~$job} - node_memory_SwapFree_bytes{instance~$node,job~$job})Harware Corrupted - 内核识别为已损坏或不工作的内存量
metrics:node_memory_HardwareCorrupted_bytes{instance~$node,job~$job}3. Network Traffic 各个网络接口的传输速率
type: Graph
Unit: bytes/sec
Label: Bytes out(-)/in(){{device}} - Receive 各个网络接口下载速率
metrics:irate(node_network_receive_bytes_total{instance~$node,job~$job}[5m]){{device}} - Transmit 各个网络接口上传速率
metrics:irate(node_network_transmit_bytes_total{instance~$node,job~$job}[5m])4. Disk Space Used 所有挂载的文件系统的磁盘空间大小
type: Graph
Unit: bytes
min: 0
Label: Bytes
metrics:node_filesystem_size_bytes{instance~$node,job~$job,device!~rootfs} - node_filesystem_avail_bytes{instance~$node,job~$job,device!~rootfs}5. Disk IOps 磁盘读写
type: Graph
Unit: I/O ops/sec (iops)
Label: IO read(-)/write(){{device}} - Reads completed 磁盘的读取速率五分钟内
metrics:irate(node_disk_reads_completed_total{instance~$node,job~$job,device~[a-z]*[a-z]}[5m]){{device}} - Writes completed 磁盘的写入速率五分钟内
metrics:irate(node_disk_writes_completed_total{instance~$node,job~$job,device~[a-z]*[a-z]}[5m])6. I/O Usage Read / Write
type: Graph
Unit: bytes
Label: Bytes read(-)/write()成功读取的字节数五分钟内
metrics:irate(node_disk_read_bytes_total{instance~$node,job~$job,device~[a-z]*[a-z]}[5m])成功写入的字节数五分钟内
metrics:irate(node_disk_written_bytes_total{instance~$node,job~$job,device~[a-z]*[a-z]}[5m])7. I/O Usage Times 使用 I/O 的毫秒数
type: Graph
Unit: ms
Label: Milliseconds
metrics:irate(node_disk_io_time_seconds_total{instance~$node,job~$job,device~[a-z]*[a-z]} [5m])
Memory Detail Meminfo /proc/meminfo1. Memory Active / Inactive
type: Graph
Unit: bytes
Label: BytesInactive - 最近使用较少的内存, 优先被回收利用 /proc/meminfo Inactive
metrics:node_memory_Inactive_bytes{instance~$node,job~$job}Active - 最近被频繁使用的内存除非绝对必要否则通常不会回收 /proc/meminfo Active
metrics:node_memory_Active_bytes{instance~$node,job~$job}2. Memory Commited
type: Graph
Unit: bytes
Label: BytesCommitted_AS - 当前系统已经分配的内存量包括已分配但尚未使用的内存大小 /proc/meminfo Committed_AS
metrics:node_memory_Committed_AS_bytes{instance~$node,job~$job}CommitLimit - 当前系统可分配的内存量 /proc/meminfo CommitLimit
metrics:node_memory_CommitLimit_bytes{instance~$node,job~$job}3. Memory Active / Inactive Detail
type: Graph
Unit: bytes
Label: BytesInactive_file - LRU list 上长时间未被访问过的与文件对应的内存页 /proc/meminfo LRU_INACTIVE_FILE
metrics:node_memory_Inactive_file_bytes{instance~$node,job~$job}Inactive_anon - 上长时间未被访问过的匿名页和交换区缓存包括 tmpfs /proc/meminfo LRU_INACTIVE_ANON
metrics:node_memory_Inactive_anon_bytes{instance~$node,job~$job}Active_file - LRU list 最近被访问过的与文件对应的内存页 /proc/meminfo LRU_ACTIVE_FILE
metrics:node_memory_Active_file_bytes{instance~$node,job~$job}Active_anon - 最近被访问过的匿名页和交换区缓存包括 tmpfs /proc/meminfo LRU_ACTIVE_ANON
metrics:node_memory_Active_anon_bytes{instance~$node,job~$job}4. Memory Writeback and Dirty
type: Graph
Unit: bytes
Label: BytesWriteback - 正准备主动回写硬盘的缓存页 /proc/meminfo Writeback
metrics:node_memory_Writeback_bytes{instance~$node,job~$job}WritebackTmp - FUSE用于临时写回缓冲区的内存 /proc/meminfo WritebackTmp
metrics:node_memory_WritebackTmp_bytes{instance~$node,job~$job}Dirty - 需要写回磁盘的数据大小 /proc/meminfo Dirty
metrics:node_memory_Dirty_bytes{instance~$node,job~$job}5. Memory Shared and Mapped
type: Graph
Unit: bytes
Label: BytesMapped - mapped 缓存页占用的内存 /proc/meminfo Mapped
metrics:node_memory_Mapped_bytes{instance~$node,job~$job}Shmem - 共享内存 /proc/meminfo Shared
metrics:node_memory_Shmem_bytes{instance~$node,job~$job}6. Memory Slab
type: Graph
Unit: bytes
Label: BytesSUnreclaim - 通过slab分配的内存中不可回收的部分 /proc/meminfo SUnreclaim
metrics:node_memory_SUnreclaim_bytes{instance~$node,job~$job}SReclaimable - 通过slab分配的内存中可回收的部分 /proc/meminfo SReclaimable
metrics:node_memory_SReclaimable_bytes{instance~$node,job~$job}7. Memory Vmalloc
type: Graph
Unit: bytes
Label: BytesVmallocChunk - vmalloc 可分配的最大的逻辑连续的内存大小 /proc/meminfo VmallocChunk
metrics:node_memory_VmallocChunk_bytes{instance~$node,job~$job}VmallocTotal - vmalloc 可使用的总内存大小 /proc/meminfo VmallocTotal
metrics:node_memory_VmallocTotal_bytes{instance~$node,job~$job}VmallocUsed - vmalloc 已用的总内存大小 /proc/meminfo VmallocUsed
metrics:node_memory_VmallocUsed_bytes{instance~$node,job~$job}8. Memory Bounce /proc/meminfo Bounce
type: Graph
Unit: bytes
Label: BytesBounce - bounce buffers 占用的内存
metrics:node_memory_Bounce_bytes{instance~$node,job~$job}9. Memory Anonymous
type: Graph
Unit: bytes
Label: BytesAnonHugePages - AnonHugePages 占用的内存大小 /proc/meminfo AnonHugePages
metrics:node_memory_AnonHugePages_bytes{instance~$node,job~$job}AnonPages - 用户进程中匿名内存页大小 /proc/meminfo AnonPages
metrics:node_memory_AnonPages_bytes{instance~$node,job~$job}1
shell
0. Memory Kernel /proc/meminfo KernelStack
type: Graph
Unit: bytes
Label: BytesKernelStack - 内核栈大小常驻内存不可回收
metrics:node_memory_KernelStack_bytes{instance~$node,job~$job}1
shell
1. Memory HugePages Counter
type: Graph
Unit: short
Label: PagesHugePages_Free - 系统当前总共拥有的空闲 HugePages 数目 /proc/meminfo HugePages_Free
metrics:node_memory_HugePages_Free{instance~$node,job~$job}HugePages_Rsvd - 系统当前总共保留的HugePages数目更具体点就是指程序已经向系统申请但是由于程序还没有实质的HugePages读写操作因此系统尚未实际分配给程序的HugePages数目 /proc/meminfo HugePages_Rsvd
metrics:node_memory_HugePages_Rsvd{instance~$node,job~$job}HugePages_Surp - 指超过系统设定的常驻HugePages数目的数目 /proc/meminfo HugePages_Surp
metrics:node_memory_HugePages_Surp{instance~$node,job~$job}1
shell
2. Memory HugePages Size
type: Graph
Unit: bytes
Label: BytesHugePages - 系统当前总共拥有的HugePages数目 /proc/meminfo HugePages
metrics:node_memory_HugePages_Total{instance~$node,job~$job}Hugepagesize - 每一页 HugePages 的大小 /proc/meminfo Hugepagesize
metrics:node_memory_Hugepagesize_bytes{instance~$node,job~$job}1
shell
3. Memory DirectMap /proc/meminfo DirectMap
type: Graph
Unit: bytes
Label: BytesDirectMap1G - 映射为 1G 的内存页的内存数量
metrics:node_memory_DirectMap1G{instance~$node,job~$job}DirectMap2M - 映射为 2M 的内存页的内存数量
metrics:node_memory_DirectMap2M_bytes{instance~$node,job~$job}DirectMap4K - 映射为 4kB 的内存页的内存数量
metrics:node_memory_DirectMap4k_bytes{instance~$node,job~$job}1
shell
4. Memory Unevictable and MLocked
type: Graph
Unit: bytes
Label: BytesUnevictable - 不可被回收的内存 /proc/meminfo Unevictable
metrics:node_memory_Unevictable_bytes{instance~$node,job~$job}MLocked - 被mlock()系统调用锁定的内存大小 /proc/meminfo MLocked
metrics:node_memory_Mlocked_bytes{instance~$node,job~$job}1
shell
5. Memory NFS /proc/meminfo NFS_Unstable
type: Graph
Unit: bytes
Label: BytesNFS Unstable - 发给NFS server但尚未写入硬盘的缓存页
metrics:node_memory_NFS_Unstable_bytes{instance~$node,job~$job}
Memory Detail Vmstat1. Memory Pages In / Out
type: Graph
Unit: short
Label: PagesPagesin - 数据从硬盘读到物理内存的速率(5分钟内) /proc/vmstat pgpgin
metrics:irate(node_vmstat_pgpgin{instance~$node,job~$job}[5m])Pagesout - 数据从物理内存写到硬盘的速率(5分钟内) /proc/vmstat pgpgout
metrics:irate(node_vmstat_pgpgout{instance~$node,job~$job}[5m])2. Memory Pages Swap In / Out
type: Graph
Unit: short
Label: PagesPswpin - 数据从磁盘交换区装入内存的速率(5分钟内) /proc/vmstat pswpin
metrics:irate(node_vmstat_pswpin{instance~$node,job~$job}[5m])Pswpout - 数据从内存转储到磁盘交换区的速率(5分钟内) /proc/vmstat pswpout
metrics:irate(node_vmstat_pswpout{instance~$node,job~$job}[5m])3. Memory Page Operations
type: Graph
Unit: short
Label: PagesPgdeactivate - 激活的平均页数(5分钟内) /proc/vmstat pgdeactivate
metrics:irate(node_vmstat_pgdeactivate{instance~$node,job~$job}[5m])Pgfree - 释放的平均页数(5分钟内) /proc/vmstat pgfree
metrics:irate(node_vmstat_pgfree{instance~$node,job~$job}[5m])Pgactivate - 未激活的平均页数(5分钟内) /proc/vmstat pgactivate
metrics:irate(node_vmstat_pgactivate{instance~$node,job~$job}[5m])4. Memory Page Faults
type: Graph
Unit: short
Label: FaultsPgfault - 一级页面和二级页面的平均错误数(5分钟内) /proc/vmstat pgfault
metrics:irate(node_vmstat_pgfault{instance~$node,job~$job}[5m])Pgmajfault - 一级页面的平均错误数(5分钟内) /proc/vmstat pgmajfault
metrics:irate(node_vmstat_pgmajfault{instance~$node,job~$job}[5m])Pgminfault - 二级页面的平均错误数(5分钟内)
metrics:irate(node_vmstat_pgfault{instance~$node,job~$job}[5m]) - irate(node_vmstat_pgmajfault{instance~$node,job~$job}[5m])5. Memory Pages Reclaimed
type: Graph
Unit: short
Label: PagesKswapd_inodesteal - 由 kswapd 回收用于其它目的的平均页面数(5分钟内) /proc/vmstat kswapd_inodesteal
metrics:irate(node_vmstat_kswapd_inodesteal{instance~$node,job~$job}[5m])Pgindesteal - 由 inode 释放回收的平均页面数(5分钟内) /proc/vmstat pgindesteal
metrics:irate(node_vmstat_pginodesteal{instance~$node,job~$job}[5m])6. Memory Calls Reclaimed
type: Graph
Unit: short
Label: CellsPageoutrun - 由 kswapd调用来回收的平均页面数(5分钟内) /proc/vmstatpageoutrun
metrics:irate(node_vmstat_pageoutrun{instance~$node,job~$job}[5m])Allocstall - 请求直接回收的平均页面数(5分钟内) /proc/vmstat allocstall
metrics:irate(node_vmstat_allocstall{instance~$node,job~$job}[5m])Zone_reclaim_failed - 内存域回收失败的平均页面数(5分钟内) /proc/vmstat zone_reclaim_failed
metrics:irate(node_vmstat_zone_reclaim_failed{instance~$node,job~$job}[5m])7. Memory Page Rotate /proc/vmstat pgrotated
type: Graph
Unit: short
Label: PagesPgrotated - 轮换的平均页面数(5分钟内)
metrics:irate(node_vmstat_pgrotated{instance~$node,job~$job}[5m])8. Memory Page Drop
type: Graph
Unit: short
Label: CellsDrop_pagecache - 调用释放缓存的平均页面数(5分钟内) /proc/vmstat drop_pagecache
metrics:node_vmstat_drop_pagecache{instance~$node,job~$job}Drop_slab - 调用释放 slab 缓存的平均页面数(5分钟内) /proc/vmstat drop_slab
metrics:node_vmstat_drop_slab{instance~$node,job~$job}9. Memory Scan Slab /proc/vmstat slabs_scanned
type: Graph
Unit: shortSlabs_scanned - 被扫描的 Slab 页面的平均页面数(5分钟内)
metrics:irate(node_vmstat_slabs_scanned{instance~$node,job~$job}[5m])1
shell
0. Memory Unevictable Pages
type: Graph
Unit: short
Label: PagesUnevictable_pgs_cleared - Unevictable pages cleared
metrics:irate(node_vmstat_unevictable_pgs_cleared{instance~$node,job~$job}[5m])Unevictable_pgs_culled - Unevictable pages culled
metrics:irate(node_vmstat_unevictable_pgs_culled{instance~$node,job~$job}[5m])Unevictable_pgs_mlocked - Unevictable pages mlocked
metrics:irate(node_vmstat_unevictable_pgs_mlocked{instance~$node,job~$job}[5m])Unevictable_pgs_munlocked - Unevictable pages munlocked
metrics:irate(node_vmstat_unevictable_pgs_munlocked{instance~$node,job~$job}[5m])Unevictable_pgs_rescued- Unevictable pages rescued
metrics:irate(node_vmstat_unevictable_pgs_rescued{instance~$node,job~$job}[5m])Unevictable_pgs_scanned - Unevictable pages scanned
metrics:irate(node_vmstat_unevictable_pgs_scanned{instance~$node,job~$job}[5m])unevictable_pgs_stranded - Unevictable pages stranded
metrics:irate(node_vmstat_unevictable_pgs_stranded{instance~$node,job~$job}[5m])1
shell
1. Memory Page Allocation
type: Graph
Unit: short
Label: PagesPgalloc_dma - DMA 存储区分配的平均页数(5分钟内) /proc/vmstat pgalloc_dma
metrics:irate(node_vmstat_pgalloc_dma{instance~$node,job~$job}[5m])Pgalloc_dma32 - DMA32 存储区分配的平均页数(5分钟内) /proc/vmstat pgalloc_dma32
metrics:irate(node_vmstat_pgalloc_dma32{instance~$node,job~$job}[5m])Pgalloc_movable - movable 存储区分配的平均页数(5分钟内) /proc/vmstat pgalloc_movable
metrics:irate(node_vmstat_pgalloc_movable{instance~$node,job~$job}[5m])Pgalloc_normal - 普通存储区分配的平均页数(5分钟内) /proc/vmstat pgalloc_normal
metrics:irate(node_vmstat_pgalloc_normal{instance~$node,job~$job}[5m])1
shell
2. Memory Page Refill
type: Graph
Unit: short
Label: PagesPgrefill_dma - DMA 再填充的平均页数(5分钟内) /proc/vmstat pgrefill_dma
metrics:irate(node_vmstat_pgrefill_dma{instance~$node,job~$job}[5m])Pgrefill_dma32 - DMA32 存储区再填充的平均页数(5分钟内) /proc/vmstat pgrefill_dma32
metrics:irate(node_vmstat_pgrefill_dma32{instance~$node,job~$job}[5m])Pgrefill_movable - movable 存储区再填充的平均页数(5分钟内) /proc/vmstat pgrefill_movable
metrics:irate(node_vmstat_pgrefill_movable{instance~$node,job~$job}[5m])Pgrefill_normal - 普通存储区再填充的平均页数(5分钟内) /proc/vmstat pgrefill_normal
metrics:irate(node_vmstat_pgrefill_normal{instance~$node,job~$job}[5m])1
shell
3. Memory Page Steal Direct
type: Graph
Unit: short
Label: PagesPgsteal_direct_dma - DMA 存储区被直接回收用于其它目的的平均页面数(5分钟内) /proc/vmstat pgsteal_direct_dma
metrics:irate(node_vmstat_pgsteal_direct_dma{instance~$node,job~$job}[5m])Pgsteal_direct_dma32 - DMA32 存储区被直接回收用于其它目的的平均页面数(5分钟内) /proc/vmstat pgsteal_direct_dma32
metrics:irate(node_vmstat_pgsteal_direct_dma32{instance~$node,job~$job}[5m])Pgsteal_direct_movable - movable 存储区被直接回收用于其它目的的平均页面数(5分钟内) /proc/vmstat pgsteal_direct_movable
metrics:irate(node_vmstat_pgsteal_direct_movable{instance~$node,job~$job}[5m])Pgsteal_direct_normal - 普通存储区被直接回收用于其它目的的平均页面数(5分钟内) /proc/vmstat pgsteal_direct_normal
metrics:irate(node_vmstat_pgsteal_direct_normal{instance~$node,job~$job}[5m])1
shell
4. Memory Page Steal Kswapd
type: Graph
Unit: short
Label: PagesPgsteal_kswapd_dma - kswapd 后台进程回收 DMA 存储区用于其它目的的平均页面数(5分钟内) /proc/vmstat pgsteal_kswapd_dma
metrics:irate(node_vmstat_pgsteal_kswapd_dma{instance~$node,job~$job}[5m])Pgsteal_kswapd_dma32 - kswapd 后台进程回收 DMA32 存储区用于其它目的的平均页面数(5分钟内) /proc/vmstat pgsteal_kswapd_dma32
metrics:irate(node_vmstat_pgsteal_kswapd_dma32{instance~$node,job~$job}[5m])Pgsteal_kswapd_movable - kswapd 后台进程回收 movable 存储区用于其它目的的平均页面数(5分钟内 /proc/vmstat pgsteal_kswapd_movable
metrics:irate(node_vmstat_pgsteal_kswapd_movable{instance~$node,job~$job}[5m])Pgsteal_kswapd_normal - swapd后台进程回收普通存储区用于其它目的的平均页面数(5分钟内 /proc/vmstat pgsteal_kswapd_normal
metrics:irate(node_vmstat_pgsteal_kswapd_normal{instance~$node,job~$job}[5m])1
shell
5. Memory Scan Direct
type: Graph
Unit: short
Label: PagesPgscan_direct_dma - DMA 存储区被直接回收的平均页面数(5分钟内) /proc/vmstat pgscan_direct_dma
metrics:irate(node_vmstat_pgscan_direct_dma{instance~$node,job~$job}[5m])Pgscan_direct_dma32 - DMA32 存储区被直接回收的平均页面数(5分钟内) /proc/vmstat pgscan_direct_dma32
metrics:irate(node_vmstat_pgscan_direct_dma32{instance~$node,job~$job}[5m])Pgscan_direct_movable - movable 存储区被直接回收的平均页面数(5分钟内) /proc/vmstat pgscan_direct_movable
metrics:irate(node_vmstat_pgscan_direct_movable{instance~$node,job~$job}[5m])Pgscan_direct_normal - 普通存储区被直接回收的平均页面数(5分钟内) /proc/vmstat pgscan_direct_normal
metrics:irate(node_vmstat_pgscan_direct_normal{instance~$node,job~$job}[5m])Pgscan_direct_throttle - throttle 存储区被直接回收的平均页面数(5分钟内) /proc/vmstat pgscan_direct_throttle
metrics:irate(node_vmstat_pgscan_direct_throttle{instance~$node,job~$job}[5m])1
shell
6. Memory Scan Kswapd
type: Graph
Unit: short
Label: PagesPgscan_kswapd_dma - kswapd 后台进程扫描的 DMA 存储区平均页面数(5分钟内) /proc/vmstat pgscan_kswapd_dma
metrics:irate(node_vmstat_pgscan_kswapd_dma{instance~$node,job~$job}[5m])Pgscan_kswapd_dma32 - kswapd 后台进程扫描的 DMA32 存储区平均页面数(5分钟内) /proc/vmstat pgscan_kswapd_dma32
metrics:irate(node_vmstat_pgscan_kswapd_dma32{instance~$node,job~$job}[5m])Pgscan_kswapd_movable - kswapd 后台进程扫描的 movable 存储区平均页面数(5分钟内) /proc/vmstat pgscan_kswapd_movable
metrics:irate(node_vmstat_pgscan_kswapd_movable{instance~$node,job~$job}[5m])Pgscan_kswapd_normal - kswapd 后台进程扫描的普通存储区平均页面数(5分钟内 /proc/vmstat pgscan_kswapd_normal
metrics:irate(node_vmstat_pgscan_kswapd_normal{instance~$node,job~$job}[5m])1
shell
7. Memory Page Compact
type: Graph
Unit: short
Label: PagesCompact_free_scanned - 扫描由压缩守护程序释放的页面 /proc/vmstat compact_free_scanned
metrics:irate(node_vmstat_compact_free_scanned{instance~$node,job~$job}[5m])Compact_isolated - 用于内存压缩隔离的页面 /proc/vmstat compact_isolated
metrics:irate(node_vmstat_compact_isolated{instance~$node,job~$job}[5m])Compact_migrate_scanned - 通过内存压缩守护程序扫描以进行迁移的页面 /proc/vmstat compact_migrate_scanned
metrics:irate(node_vmstat_compact_migrate_scanned{instance~$node,job~$job}[5m])1
shell
8. Memory Compactions 内存紧缩
type: Graph
Unit: short
Label: CompactionsCompact_fail - 高阶分配的内存碎片整理失败的页面数(5分钟内) /proc/vmstat compact_fail
metrics:irate(node_vmstat_compact_fail{instance~$node,job~$job}[5m])Compact_stall - 开始执行内存碎片失败的页面数(5分钟内) /proc/vmstat compact_stall
metrics:irate(node_vmstat_compact_stall{instance~$node,job~$job}[5m])Compact_sucess - 高阶分配的内存碎片整理成功的页面数(5分钟内)
metrics:irate(node_vmstat_compact_success{instance~$node,job~$job}[5m])1
shell
9. Memory Kswapd Watermark
type: Graph
Unit: short
Label: CounterKswapd_high_wmark_hit_quickly - 剩余内存达到 high 的水位线的时间 /proc/vmstat kswapd_high_wmark_hit_quickly
metrics:node_vmstat_kswapd_high_wmark_hit_quickly{instance~$node,job~$job}Kswapd_low_wmark_hit_quickly - - 剩余内存达到 low 的水位线的时间 /proc/vmstat kswapd_low_wmark_hit_quickly
metrics:node_vmstat_kswapd_low_wmark_hit_quickly{instance~$node,job~$job}2
shell
0. Memory Buddy Alloc
type: Graph
Unit: short
Label: AllocationsHtlb_buddy_alloc_fail - buddy 给 hugetlb 分配失败的次数 /proc/vmstat htlb_buddy_alloc_fail
metrics:node_vmstat_htlb_buddy_alloc_fail{instance~$node,job~$job}Htlb_buddy_alloc_success - buddy 给 hugetlb 分配成功的次数 /proc/vmstat htlb_buddy_alloc_success
metrics:node_vmstat_htlb_buddy_alloc_success{instance~$node,job~$job}2
shell
1. Memory Numa Allocations
type: Graph
Unit: short
Label: AllocationsNuma_foreign - 计划使用其他节点内存但是却使用本地内存次数 /proc/vmstat numa_foreign
metrics:irate(node_vmstat_numa_foreign{instance~$node,job~$job}[5m])Numa_hit - 使用本节点内存次数 /proc/vmstat numa_hit
metrics:irate(node_vmstat_numa_hit{instance~$node,job~$job}[5m])Numa_interleave - 交叉分配使用的内存中使用本节点的内存次数 /proc/vmstat numa_interleave
metrics:irate(node_vmstat_numa_interleave{instance~$node,job~$job}[5m])Numa_local - 在本节点运行的程序使用本节点内存次数 /proc/vmstat numa_local
metrics:irate(node_vmstat_numa_local{instance~$node,job~$job}[5m])Numa_miss - 计划使用本节点内存而被调度到其他节点次数 /proc/vmstat numa_miss
metrics:irate(node_vmstat_numa_miss{instance~$node,job~$job}[5m])Numa_other - 在其他节点运行的程序使用本节点内存次数 /proc/vmstat numa_other
metrics:irate(node_vmstat_numa_other{instance~$node,job~$job}[5m])2
shell
2. Memory Numa Page Migrations
type: Graph
Unit: short
Label: PagesNuma_pages_migrated - NUMA page 数 /proc/vmstat numa_pages_migrated
metrics:irate(node_vmstat_numa_pages_migrated{instance~$node,job~$job}[5m])Pgmigrate_fail - 迁移失败的页面数 /proc/vmstat pgmigrate_fail
metrics:irate(node_vmstat_pgmigrate_fail{instance~$node,job~$job}[5m])Pgmigrate_success - 成功迁移的页面数 /proc/vmstat pgmigrate_success
metrics:irate(node_vmstat_pgmigrate_success{instance~$node,job~$job}[5m])2
shell
3. Memory Numa Hints
type: Graph
Unit: short
Label: HintsNuma_hint_faults - NUMA hint faults trapped
metrics:irate(node_vmstat_numa_hint_faults{instance~$node,job~$job}[5m])Numa_hint_faults_local - Hinting faults to local nodes
metrics:irate(node_vmstat_numa_hint_faults_local{instance~$node,job~$job}[5m])2
shell
4. Memory Numa Table Updates
type: Graph
Unit: short
Label: UpdatesNuma_pte_updates - NUMA page table entry updates
metrics:irate(node_vmstat_numa_pte_updates{instance~$node,job~$job}[5m])Numa_huge_pte_updates - NUMA huge page table entry updates
metrics:irate(node_vmstat_numa_huge_pte_updates{instance~$node,job~$job}[5m])2
shell
5. Memory THP Splits
type: Graph
Unit: short
Label: SplitsThp_split - 大型页面分割成多个常规页面 /proc/vmstat thp_split
metrics:irate(node_vmstat_thp_split{instance~$node,job~$job}[5m])2
shell
6. Memory Workingset
type: Graph
Unit: short
Label: CounterWorkingset_activate - Page activations to form the working set
metrics:irate(node_vmstat_workingset_activate{instance~$node,job~$job}[5m])Workingset_nodereclaim - NUMA node working set page reclaims
metrics:irate(node_vmstat_workingset_nodereclaim{instance~$node,job~$job}[5m])Workingset_refault - Refaults of previously evicted pages
metrics:irate(node_vmstat_workingset_refault{instance~$node,job~$job}[5m])2
shell
7. Memory THP Allocations
type: Graph
Unit: short
Label: AllocationsThp_collapse_alloc - Transparent huge page collapse allocations
metrics:irate(node_vmstat_thp_collapse_alloc{instance~$node,job~$job}[5m])Thp_collapse_alloc_failed - Transparent huge page collapse allocation failures
metrics:irate(node_vmstat_thp_collapse_alloc_failed{instance~$node,job~$job}[5m])Thp_zero_page_alloc - Transparent huge page zeroed page allocations
metrics:irate(node_vmstat_thp_zero_page_alloc{instance~$node,job~$job}[5m])Thp_zero_page_alloc_failed - Transparent huge page zeroed page allocation failures
metrics:irate(node_vmstat_thp_zero_page_alloc_failed{instance~$node,job~$job}[5m])Thp_fault_alloc - Transparent huge page fault allocations
metrics:irate(node_vmstat_thp_fault_alloc{instance~$node,job~$job}[5m])Thp_fault_fallback - Transparent huge page fault fallbacks
metrics:irate(node_vmstat_thp_fault_fallback{instance~$node,job~$job}[5m])
Memory Detail Vmstat Counters1. Memory Page Active
type: Graph
Unit: short
Label: PagesActive_anon - pages最近被使用过的匿名虚拟内存页 /proc/vmstat nr_active_anon
metrics:node_vmstat_nr_active_anon{instance~$node,job~$job}Active_file - 最近被使用过的文件虚拟内存页 /proc/vmstat nr_active_file
metrics:node_vmstat_nr_active_file{instance~$node,job~$job}2. Memory Page Reclaimed / Unreclaimed
type: Graph
Unit: short
Label: PagesReclaimable - 可回收的 slab 虚拟内存页 /proc/vmstat nr_slab_reclaimable
metrics:node_vmstat_nr_slab_reclaimable{instance~$node,job~$job}Unreclaimable - 不可回收的 slab 虚拟内存页 /proc/vmstat nr_slab_unreclaimable
metrics:node_vmstat_nr_slab_unreclaimable{instance~$node,job~$job}3. Memory Page Inactive
type: Graph
Unit: short
Label: PagesInactive_anon - 每个 NUMA node 的每个域中的长时间未被访问过的匿名内存页 /proc/vmstat nr_inactive_anon
metrics:node_vmstat_nr_inactive_anon{instance~$node,job~$job}Inactive_file - 每个 NUMA node 的每个域中的长时间未被访问过的与文件对应的内存页 /proc/vmstat nr_inactive_file
metrics:node_vmstat_nr_inactive_file{instance~$node,job~$job}4. Memory Page Dirty / Bounce
type: Graph
Unit: short
Label: PagesDirty - 脏页数 /proc/vmstat nr_dirty
metrics:node_vmstat_nr_dirty{instance~$node,job~$job}Bounce - Bounce buffer 页面数 /proc/vmstat nr_bounce
metrics:node_vmstat_nr_bounce{instance~$node,job~$job}5. Memory Page Free / Written
type: Graph
Unit: short
Label: PagesFree_pages - 空闲页数 /proc/vmstat nr_free_pages
metrics:node_vmstat_nr_free_pages{instance~$node,job~$job}Written - 每个 NUMA node 的每个域中写出的页面 /proc/vmstat nr_written
metrics:node_vmstat_nr_written{instance~$node,job~$job}6.Memory Page Shmem / Mapped
type: Graph
Unit: short
Label: PagesShmem - 共享内存页数 /proc/vmstat nr_shmem
metrics:node_vmstat_nr_shmem{instance~$node,job~$job}Mapped - 每个 NUMA node 的每个域 mapped 缓存页的页数 /proc/vmstat nr_mapped
metrics:node_vmstat_nr_mapped{instance~$node,job~$job}7.Memory Page Unevictable / MLock
type: Graph
Unit: short
Label: PagesUnevictable - 不可回收的页数 /proc/vmstat nr_unevictable
metrics:node_vmstat_nr_unevictable{instance~$node,job~$job}Mlock - 被 mlock()系统调用锁定的页数 /proc/vmstat nr_mlock
metrics:node_vmstat_nr_mlock{instance~$node,job~$job}8.Memory Page Writeback
type: Graph
Unit: short
Label: PagesWriteback - 回写页数 /proc/vmstat nr_writeback
metrics:node_vmstat_nr_writeback{instance~$node,job~$job}Writeback_temp - 临时回写页数 /proc/vmstat nr_writeback_temp
metrics:node_vmstat_nr_writeback_temp{instance~$node,job~$job}9.Memory Page Kernel_stack
type: Graph
Unit: short
Label: PagesKernel_stack - 内核栈的页数 /proc/vmstat nr_kernel_stack
metrics:node_vmstat_nr_kernel_stack{instance~$node,job~$job}1
shell
0.Memory Page Dirty Threshold
type: Graph
Unit: short
Label: PagesDirty_background_threshold - 脏页后台回写阈值 /proc/vmstat nr_dirty_background_threshold
metrics:node_vmstat_nr_dirty_background_threshold{instance~$node,job~$job}Dirty_threshold - 脏页限制阈值 /proc/vmstat nr_dirty_threshold
metrics:node_vmstat_nr_dirty_threshold{instance~$node,job~$job}1
shell
1.Memory Page File_pages
type: Graph
Unit: short
Label: PagesFile_pages - 每个 NUMA node 的每个域文件缓存页的页数 /proc/vmstat nr_file_pages
metrics:node_vmstat_nr_file_pages{instance~$node,job~$job}1
shell
2.Memory Page Page_table_pages
type: Graph
Unit: short
Label: PagesPage_table_pages - 每个 NUMA node 的每个域页面表的页数 /proc/vmstat nr_page_table_pages
metrics:node_vmstat_nr_page_table_pages{instance~$node,job~$job}1
shell
3.Memory Page Unstable / Dirtied
type: Graph
Unit: short
Label: PagesUnstable - 每个 NUMA node 的每个域中处于不稳定页面的页数 /proc/vmstat nr_unstable
metrics:node_vmstat_nr_unstable{instance~$node,job~$job}Dirtied - 每个 NUMA node 的每个域中进入脏页面的页数 /proc/vmstat nr_dirtied
metrics:node_vmstat_nr_dirtied{instance~$node,job~$job}1
shell
4.Memory Page Isolated
type: Graph
Unit: short
Label: PagesIsolated_anon - 每个 NUMA node 的每个域中隔离的匿名内存页面的页数 /proc/vmstat nr_isolated_anon
metrics:node_vmstat_nr_isolated_anon{instance~$node,job~$job}Isolated_file - 每个 NUMA node 的每个域中隔离的文件存储页面的页数 /proc/vmstat nr_isolated_file
metrics:node_vmstat_nr_isolated_file{instance~$node,job~$job}1
shell
5.Memory Page Alloc_batch
type: Graph
Unit: short
Label: PagesAlloc_batch - 每个 NUMA node 的每个域中由于内存不足分配给其他域的页面 /proc/vmstat nr_alloc_batch
metrics:node_vmstat_nr_alloc_batch{instance~$node,job~$job}1
shell
6.Memory Page Misc
type: Graph
Unit: short
Label: PagesFree_cma - 每个 NUMA node 的每个域中空闲的连续内存分配器页面 /proc/vmstat nr_free_cma
metrics:node_vmstat_nr_free_cma{instance~$node,job~$job}Vmscan_write - LRU 内存回收写入的页面 /proc/vmstat nr_vmscan_write
metrics:node_vmstat_nr_vmscan_write{instance~$node,job~$job}Immediate_reclaim - 每个 NUMA node 的每个域中当回写结束时优先回收的页面 /proc/vmstat nr_vmscan_immediate_reclaim
metrics:node_vmstat_nr_vmscan_immediate_reclaim{instance~$node,job~$job}1
shell
7.Memory Page Anon
type: Graph
Unit: short
Label: PagesAnon_pages - 每个 NUMA node 的每个域中匿名 mapped 缓存页 /proc/vmstat nr_anon_pages
metrics:node_vmstat_nr_anon_pages{instance~$node,job~$job}Anon_transparent_hugepages - 每个 NUMA node 的每个域中 THP(Transparent Huge Pages) /proc/vmstat nr_anon_transparent_hugepages
metrics:node_vmstat_nr_anon_transparent_hugepages{instance~$node,job~$job}
System Detail1. Context Switches / Interrupts
type: Graph
Unit: short
Label: CounterContext switches - CPU 的 context switch 平均次数(5分钟内)
metrics:irate(node_context_switches_total{instance~$node,job~$job}[5m])Interrupts - 服务的平均中断总数(5分钟内)
metrics:irate(node_intr_total{instance~$node,job~$job}[5m])2. System Load
type: Graph
Unit: short
Label: LoadLoad 1m - 系统1分钟内的平均负载
metrics:node_load1{instance~$node,job~$job}Load 5m - 系统5分钟内的平均负载
metrics:node_load5{instance~$node,job~$job}Load 15m - 系统15分钟内的平均负载
metrics:node_load15{instance~$node,job~$job}3. Interrupts Detail /proc/interrupts
type: Graph
Unit: short
Label: Counter{{ type }} - {{ info }} - 当前系统的软中断列表和对应的中断号平均中断次数(5分钟内)
metrics:irate(node_interrupts_total{instance~$node,job~$job}[5m])4. File Descriptors
type: Graph
Unit: short
Label: DescriptorsMaximum open file descriptors - 最大打开文件描述符数
metrics:process_max_fds{instance~$node,job~$job}Open file descriptors - 打开文件描述符的数量
metrics:process_open_fds{instance~$node,job~$job}5. Entropy
type: Graph
Unit: short
Label: EntropyEntropy available to random number generators
metrics:node_entropy_available_bits{instance~$node,job~$job}6. Processes State
type: Graph
Unit: short
Label: ProcessesProcesses blocked - 当前被阻塞的任务的数目 /proc/stat procs_blocked
metrics:node_procs_blocked{instance~$node,job~$job}Processes in runnable state - 当前运行队列的任务的数目 /proc/stat procs_running
metrics:node_procs_running{instance~$node,job~$job}7. Processes Forks
type: Graph
Unit: short
Label: Forks / secProcesses forks second - 每秒创建的进程个数
metrics:rate(node_forks_total{instance~$node,job~$job}[5m])8. Processes Memory
type: Graph
Unit: bytes
Label: Bytes进程占用的虚拟内存的大小:
metrics:process_virtual_memory_bytes{instance~$node,job~$job}进程常驻内存大小:
metrics:process_resident_memory_bytes{instance~$node,job~$job}9. Time Syncronized Status
type: Graph
Unit: short
Label: Counter时钟是否与一个可靠的服务器同步:
metrics:node_timex_sync_status{instance~$node,job~$job}本地时钟调整频率:
metrics:node_timex_frequency_adjustment_ratio{instance~$node,job~$job}1
shell
0. Time Syncronized Drift
type: Graph
Unit: seconds
Label: Seconds估算误差(秒):
metrics:node_timex_estimated_error_seconds{instance~$node,job~$job}本地系统和参考时钟之间的时间偏移:
metrics:node_timex_offset_seconds{instance~$node,job~$job}最大误差(秒):
metrics:node_timex_maxerror_seconds{instance~$node,job~$job}1
shell
1. Hardware temperature monitor 硬件的温度监控
type: Graph
Unit: Celsius(摄氏度)
Label: Temperature{{ chip }} {{ sensor }} temp -
metrics:node_hwmon_temp_celsius{instance~$node,job~$job}{{ chip }} {{ sensor }} Critical Alarm
metrics:node_hwmon_temp_crit_alarm_celsius{instance~$node,job~$job}{{ chip }} {{ sensor }} Critical
metrics:node_hwmon_temp_crit_celsius{instance~$node,job~$job}{{ chip }} {{ sensor }} Critical Historical
metrics:node_hwmon_temp_crit_hyst_celsius{instance~$node,job~$job}{{ chip }} {{ sensor }} Max
metrics:node_hwmon_temp_max_celsius{instance~$node,job~$job}
Disk Datail /proc/diskstats1. Disk IOps Completed
type: Graph
Unit: I/O ops/sec(iops)
Label: IO read(-)/write(){{device}} - Reads completed: 每个磁盘分区每秒读完成次数
metrics:irate(node_disk_reads_completed_total{instance~$node,job~$job}[5m]){{device}} - Writes completed: 每个磁盘分区每秒写完成次数
metrics:irate(node_disk_writes_completed_total{instance~$node,job~$job}[5m])2. Disk R/W Data
type: Graph
Unit: bytes/sec
Label: Bytes read(-)/write(){{device}} - Read bytes 每个磁盘分区每秒读取的比特数
metrics:irate(node_disk_read_bytes_total{instance~$node,job~$job}[5m]){{device}} - Written bytes 每个磁盘分区每秒写入的比特数
metrics:irate(node_disk_written_bytes_total{instance~$node,job~$job}[5m])3. Disk R/W Time
type: Graph
Unit: Milliseconds(ms)
Label: Millisec. read(-)/write(){{device}} - Read time ms 每个磁盘分区读花费的毫秒数
metrics:irate(node_disk_read_time_seconds_total{instance~$node,job~$job}[5m]){{device}} - Write time ms 每个磁盘分区写操作花费的毫秒数
metrics:irate(node_disk_write_time_seconds_total{instance~$node,job~$job}[5m])4. Disk IOs Weighted
type: Graph
Unit: Milliseconds(ms)
Label: Milliseconds{{device}} - IO time weighted 每个磁盘分区输入/输出操作花费的加权毫秒数
metrics:irate(node_disk_io_time_weighted_seconds_total{instance~$node,job~$job}[5m])5. Disk R/W Merged
type: Graph
Unit: I/O ops/sec(iops)
Label: I/Os{{device}} - Read merged 每个磁盘分区每秒合并读完成次数
metrics:irate(node_disk_reads_merged_total{instance~$node,job~$job}[5m]){{device}} - Write merged 每个磁盘分区每秒合并写完成次数
metrics:irate(node_disk_writes_merged_total{instance~$node,job~$job}[5m])6. Milliseconds Spent Doing I/Os
type: Graph
Unit: Milliseconds(ms)
Label: Milliseconds{{device}} - IO time ms 每个磁盘分区输入/输出操作花费的毫秒数
metrics:irate(node_disk_io_time_seconds_total{instance~$node,job~$job}[5m])7. Disk IOs Current in Progress
type: Graph
Unit: I/O ops/sec(iops)
Label: I/Os{{device}} - IO now 每个磁盘分区每秒正在处理的输入/输出请求数
metrics:irate(node_disk_io_now{instance~$node,job~$job}[5m])8. Open Error File
type: Graph
Unit: short
Label: ErrorsTextfile scrape error (1 true) 打开文件错误的个数
metrics:node_textfile_scrape_error{instance~$node,job~$job}
FileSystem Detail /proc/filesystems1. Filesystem space available
type: Graph
Unit: bytes
Label: Bytes{{mountpoint}} - 挂载的文件系统可用空间
metrics:node_filesystem_avail_bytes{instance~$node,job~$job,device!~rootfs}{{mountpoint}} - 挂载的文件系统剩余空间
metrics:node_filesystem_free_bytes{instance~$node,job~$job,device!~rootfs}{{mountpoint}} - 挂载的文件系统占用空间
metrics:node_filesystem_size_bytes{instance~$node,job~$job,device!~rootfs}2. File Nodes Free
type: Graph
Unit: short
Label: File Nodes{{mountpoint}} - 挂载的文件系统空闲的文件节点个数
metrics:node_filesystem_files_free{instance~$node,job~$job,device!~rootfs}3. File Descriptor
type: Graph
Unit: short
Label: Files最大打开文件描述符数
metrics:node_filefd_maximum{instance~$node,job~$job}打开文件描述符数:
metrics:node_filefd_allocated{instance~$node,job~$job}4. File Nodes Size
type: Graph
Unit: short
Label: File Nodes{{mountpoint}} - File nodes total挂载的文件系统的文件节点大小
metrics:node_filesystem_files{instance~$node,job~$job,device!~rootfs}5. Filesystem in ReadOnly
type: Graph
Unit: short
Label: Read Only{{mountpoint}} - ReadOnly 只读模式挂载的文件系统
metrics:node_filesystem_readonly{instance~$node,job~$job,device!~rootfs}
Network Traffic Detail /proc/net/dev1. Network Traffic by Packets
type: Graph
Unit: packets/sec
Label: Packets out (-) / in (){{device}} - Receive 各个接口每秒接收的数据包总数
metrics:irate(node_network_receive_packets_total{instance~$node,job~$job}[5m]){{device}} - Transmit 各个接口每秒发送的数据包总数
metrics:irate(node_network_transmit_packets_total{instance~$node,job~$job}[5m])2. Network Traffic Errors
type: Graph
Unit: packets/sec
Label: Packets out (-) / in (){{device}} - Receive errors 监测到各个接口每秒接收的错误数据包总数
metrics:irate(node_network_receive_errs_total{instance~$node,job~$job}[5m]){{device}} - Rransmit errors 监测到各个接口每秒发送的错误数据包总数
metrics:irate(node_network_transmit_errs_total{instance~$node,job~$job}[5m])3. Network Traffic Drop
type: Graph
Unit: packets/sec
Label: Packets out (-) / in (){{device}} - Receive drop 各个接口每秒接收的丢弃的数据包总数
metrics:irate(node_network_receive_drop_total{instance~$node,job~$job}[5m]){{device}} - Transmit drop 各个接口每秒发送的丢弃的数据包总数
metrics:irate(node_network_transmit_drop_total{instance~$node,job~$job}[5m])4. Network Traffic Compressed
type: Graph
Unit: packets/sec
Label: Packets out (-) / in (){{device}} - Receive compressed 各个接口每秒接收的压缩数据包总数
metrics:irate(node_network_receive_compressed_total{instance~$node,job~$job}[5m]){{device}} - Transmit compressed 各个接口每秒发送的压缩数据包总数
metrics:irate(node_network_transmit_compressed_total{instance~$node,job~$job}[5m])5. Network Traffic Multicast
type: Graph
Unit: packets/sec
Label: Packets out (-) / in (){{device}} - Receive multicast 各个接口每秒接收的多播包数
metrics:irate(node_network_receive_multicast_total{instance~$node,job~$job}[5m])6. Network Traffic Fifo
type: Graph
Unit: packets/sec
Label: Packets out (-) / in (){{device}} - Receive fifo 各个接口每秒接收的 fifo 包总数
metrics:irate(node_network_receive_fifo_total{instance~$node,job~$job}[5m]){{device}} - Transmit fifo 各个接口每秒发送的 fifo 包总数
metrics:irate(node_network_transmit_fifo_total{instance~$node,job~$job}[5m])7. Network Traffic Frame
type: Graph
Unit: packets/sec
Label: Packets out (-) / in (){{device}} - Receive frame 各个接口每秒接收的帧数
metrics:irate(node_network_receive_frame_total{instance~$node,job~$job}[5m])8. Network Traffic Carrier
type: Graph
Unit: short
Label: Counter{{device}} - Statistic transmit_carrier 由各个接口检测到的载波损耗的数量
metrics:irate(node_network_transmit_carrier_total{instance~$node,job~$job}[5m])9. Network Traffic Colls
type: Graph
Unit: short
Label: Counter{{device}} - Transmit colls 各个接口上检测到的冲突数
metrics:irate(node_network_transmit_colls_total{instance~$node,job~$job}[5m])1
shell
0. NF Contrack
type: Graph
Unit: short
Label: EntriesNF conntrack entries 跟踪连接数
metrics:node_nf_conntrack_entries{instance~$node,job~$job}NF conntrack limit
metrics:node_nf_conntrack_entries_limit{instance~$node,job~$job}1
shell
1. ARP Entries
type: Graph
Unit: short
Label: Entries{{ device }} - ARP entries 各个接口上 ARP 表中包的统计
metrics:node_arp_entries{instance~$node,job~$job}
Network Sockstat proc/net/sockstat1. Sockstat TCP
type: Graph
Unit: short
Label: SocketsTCP_alloc - 已分配已建立、已申请到sk_buff的TCP套接字数量
metrics:node_sockstat_TCP_alloc{instance~$node,job~$job}TCP_inuse - 正在使用正在侦听的TCP套接字数量
metrics:node_sockstat_TCP_inuse{instance~$node,job~$job}TCP_mem - TCP 套接字缓冲区使用量
metrics:node_sockstat_TCP_mem{instance~$node,job~$job}TCP_orphan - 无主不属于任何进程的TCP连接数无用、待销毁的TCP socket数
metrics:node_sockstat_TCP_orphan{instance~$node,job~$job}TCP_tw - 等待关闭的TCP连接数
metrics:node_sockstat_TCP_tw{instance~$node,job~$job}2. Sockstat UDP
type: Graph
Unit: short
Label: SocketsUDPLITE_inuse - 正在使用的 UDP-Lite 套接字数量
metrics:node_sockstat_UDPLITE_inuse{instance~$node,job~$job}UDP_inuse - 正在使用的 UDP 套接字数量
metrics:node_sockstat_UDP_inuse{instance~$node,job~$job}UDP_mem - UDP 套接字缓冲区使用量
metrics:node_sockstat_UDP_mem{instance~$node,job~$job}3. Sockstat Used
type: Graph
Unit: short
Label: SocketsSockets_used - 已使用的所有协议套接字总量
metrics:node_sockstat_sockets_used{instance~$node,job~$job}4. Sockstat Memory Size
type: Graph
Unit: bytes
Label: BytesTCP_mem_bytes - TCP 套接字缓冲区比特数
metrics:node_sockstat_TCP_mem_bytes{instance~$node,job~$job}UDP_mem_bytes - UDP 套接字缓冲区比特数
metrics:node_sockstat_UDP_mem_bytes{instance~$node,job~$job}5. Sockstat FRAG / RAW
type: Graph
Unit: short
Label: SocketsFRAG_inuse - 正在使用的 Frag 套接字数量
metrics:node_sockstat_FRAG_inuse{instance~$node,job~$job}FRAG_memory - 使用的 Frag 缓冲区
metrics:node_sockstat_FRAG_memory{instance~$node,job~$job}RAW_inuse - 正在使用的 Raw 套接字数量
metrics:node_sockstat_RAW_inuse{instance~$node,job~$job}
Network Netstat /proc/net/netstat1. Netstat IP In / Out
type: Graph
Unit: short
Label: Datagrams out (-) / in ()InReceives - 接收到的 ip 数据报
metrics:irate(node_netstat_Ip_InReceives{instance~$node,job~$job}[5m])DefaultTTL - 接收的默认生存时间的 IP 数据报
metrics:irate(node_netstat_Ip_DefaultTTL{instance~$node,job~$job}[5m])InDelivers - 传递的 IP 数据报
metrics:irate(node_netstat_Ip_InDelivers{instance~$node,job~$job}[5m])OutRequests - 发送的 ip 数据报
metrics:irate(node_netstat_Ip_OutRequests{instance~$node,job~$job}[5m])2. Netstat IP In / Out
type: Graph
Unit: short
Label: Octets out (-) / in ()InOctets - 接收到的 ip 数据报octets
metrics:irate(node_netstat_IpExt_InOctets{instance~$node,job~$job}[5m])OutOctets - 发送的 ip 数据报octets
metrics:irate(node_netstat_IpExt_OutOctets{instance~$node,job~$job}[5m])3. Netstat IP Bcast
type: Graph
Unit: short
Label: Datagrams out (-) / in ()InBcastPkts - 接收的 IP 广播数据报报文
metrics:irate(node_netstat_IpExt_InBcastPkts{instance~$node,job~$job}[5m])OutBcastPkts - 发送的 IP 广播数据报报文
metrics:irate(node_netstat_IpExt_OutBcastPkts{instance~$node,job~$job}[5m])4. Netstat IP Bcast Octets
type: Graph
Unit: short
Label: Octets out (-) / in ()InBcastOctets - 接收的 IP 广播数据报 octet 数
metrics:irate(node_netstat_IpExt_InBcastOctets{instance~$node,job~$job}[5m])OutBcastOctets - 发送的 IP 广播数据报 octet 数
metrics:irate(node_netstat_IpExt_OutBcastOctets{instance~$node,job~$job}[5m])5. Netstat IP Mcast
type: Graph
Unit: short
Label: Datagrams out (-) / in ()InMcastPkts - 接收的 IP 多播数据报报文
metrics:irate(node_netstat_IpExt_InMcastPkts{instance~$node,job~$job}[5m])OutMcastPkts - 发送的 IP 多播数据报报文
metrics:irate(node_netstat_IpExt_OutMcastPkts{instance~$node,job~$job}[5m])6. Netstat IP Mcast Octets
type: Graph
Unit: short
Label: Octets out (-) / in ()InMcastOctets - 接收的 IP 多播数据报octet 数
metrics:irate(node_netstat_IpExt_InMcastOctets{instance~$node,job~$job}[5m])OutMcastOctets - 发送的 IP 多播数据报报文 octet 数
metrics:irate(node_netstat_IpExt_OutMcastOctets{instance~$node,job~$job}[5m])7. Netstat IP Forwarding
type: Graph
Unit: short
Label: DatagramsForwDatagrams - IP 转发报文数
metrics:irate(node_netstat_Ip_ForwDatagrams{instance~$node,job~$job}[5m])Forwarding - IP 转发
metrics:irate(node_netstat_Ip_Forwarding{instance~$node,job~$job}[5m])8. Netstat IP Fragmented
type: Graph
Unit: short
Label: DatagramsFragCreates - 创建的 IP 分片报文数
metrics:irate(node_netstat_Ip_FragCreates{instance~$node,job~$job}[5m])FragFails - 失败的 IP 分片报文数
metrics:irate(node_netstat_Ip_FragFails{instance~$node,job~$job}[5m])FragOKs - 成功的 IP 分片报文数
metrics:irate(node_netstat_Ip_FragOKs{instance~$node,job~$job}[5m])9. Netstat IP ECT / CEP
type: Graph
Unit: short
Label: DatagramsInCEPkts - 拥塞转发的数据报
metrics:irate(node_netstat_IpExt_InCEPkts{instance~$node,job~$job}[5m])InECT0Pkts - 接收到的带有 ECT(0) 代码点的 ip 数据报
metrics:irate(node_netstat_IpExt_InECT0Pkts{instance~$node,job~$job}[5m])InECT1Pkt - 接收到的带有 ECT(1) 代码点的 ip 数据报
metrics:irate(node_netstat_IpExt_InECT1Pkts{instance~$node,job~$job}[5m])InNoECTPkts - 接收到的带有 NOECT 的 ip 数据报
metrics:irate(node_netstat_IpExt_InNoECTPkts{instance~$node,job~$job}[5m])1
shell
0. Netstat IP Reasambled
type: Graph
Unit: short
Label: DatagramsReasmFails - IP 重组失败的数据报
metrics:irate(node_netstat_Ip_ReasmFails{instance~$node,job~$job}[5m])ReasmOKs - IP 重组成功的数据报
metrics:irate(node_netstat_Ip_ReasmOKs{instance~$node,job~$job}[5m])ReasmReqds - 需要进行 IP 重组的数据报
metrics:irate(node_netstat_Ip_ReasmReqds{instance~$node,job~$job}[5m])ReasmTimeout - IP 重组超时的数据报
metrics:irate(node_netstat_Ip_ReasmTimeout{instance~$node,job~$job}[5m])1
shell
1. Netstat IP Errors / Discards
type: Graph
Unit: short
Label: Datagrams out (-) / in ()InDiscards - 接收的丢弃的 ip 数据报
metrics:irate(node_netstat_Ip_InDiscards{instance~$node,job~$job}[5m])InHdrErrors - IP inhdrerrors
metrics:irate(node_netstat_Ip_InHdrErrors{instance~$node,job~$job}[5m])InUnknownProtos - 由于未知协议而丢弃的 IP 数据报
metrics:irate(node_netstat_Ip_InUnknownProtos{instance~$node,job~$job}[5m])OutDiscards - IP outdiscards
metrics:irate(node_netstat_Ip_OutDiscards{instance~$node,job~$job}[5m])OutNoRoutes - 由于没有输出路由而丢弃的 IP 数据报
metrics:irate(node_netstat_Ip_OutNoRoutes{instance~$node,job~$job}[5m])InNoRoutes - 由于转发路径中没有路由而丢弃的 IP 数据报
metrics:irate(node_netstat_IpExt_InNoRoutes{instance~$node,job~$job}[5m])InCsumErrors - 具有校验和错误的 IP 数据报
metrics:irate(node_netstat_IpExt_InCsumErrors{instance~$node,job~$job}[5m])InTruncatedPkts - 由于帧没有携带足够的数据而丢弃的 IP 数据报
metrics:irate(node_netstat_IpExt_InTruncatedPkts{instance~$node,job~$job}[5m])InAddrErrors - 由于内部地址错误而丢弃的 IP 数据报
metrics:irate(node_netstat_Ip_InAddrErrors{instance~$node,job~$job}[5m])
Network Netstat TCP /proc/net/snmp1. TCP Segments
type: Graph
Unit: short
Label: Segments out (-) / in ()InCsumErrors - 接收的带有校验和错误的报文数(5分钟内)
metrics:irate(node_netstat_Tcp_InCsumErrors{instance~$node,job~$job}[5m])InErrs - TCP 接收的错误报文数(5分钟内)(例如错误的校验和)
metrics:irate(node_netstat_Tcp_InErrs{instance~$node,job~$job}[5m])InSegs - TCP 接收的目前所有建立连接的错误报文数(5分钟内)(例如错误的校验和)
metrics:irate(node_netstat_Tcp_InSegs{instance~$node,job~$job}[5m])OutRsts - TCP 发送的报文数(5分钟内)包括 RST flag
metrics:irate(node_netstat_Tcp_OutRsts{instance~$node,job~$job}[5m])OutSegs - TCP 发送的报文数(5分钟内)包括当前连接的段但是不包括重传的段)
metrics:irate(node_netstat_Tcp_OutSegs{instance~$node,job~$job}[5m])RetransSegs - TCP 重传报文数(5分钟内)
metrics:irate(node_netstat_Tcp_RetransSegs{instance~$node,job~$job}[5m])2. TCP Connections
type: Graph
Unit: short
Label: ConnectionsCurrEstab - 当前状态为 ESTABLISHED 或 CLOSE-WAIT 的 TCP 连接数
metrics:node_netstat_Tcp_CurrEstab{instance~$node,job~$job}MaxConn - 限制实体可以支持的 TCP 最大连接总数
metrics:node_netstat_Tcp_MaxConn{instance~$node,job~$job}3. TCP Retransmission
type: Graph
Unit: milliseconds
Label: MillisecondsRtoAlgorithm - TCP 重传超时时间
metrics:node_netstat_Tcp_RtoAlgorithm{instance~$node,job~$job}RtoMax - TCP允许的重传超时的最大值以毫秒为单位
metrics:node_netstat_Tcp_RtoMax{instance~$node,job~$job}RtoMin - TCP允许的重传超时的最小值以毫秒为单位
metrics:node_netstat_Tcp_RtoMin{instance~$node,job~$job}4. TCP Segments
type: Graph
Unit: short
Label: ConnectionsActiveOpens - 已从 CLOSED 状态直接转换到 SYN-SENT 状态的 TCP 平均连接数(5分钟内)
metrics:irate(node_netstat_Tcp_ActiveOpens{instance~$node,job~$job}[5m])AttemptFails - 从 SYN-SENT 和 SYN-RCVD 转换到 CLOSED 状态的 TCP 平均连接数(5分钟内)
metrics:irate(node_netstat_Tcp_AttemptFails{instance~$node,job~$job}[5m])EstabResets - 从 ESTABLISHED 状态或 CLOSE-WAIT 状态直接转换到 CLOSED 状态的 TCP 平均连接数(5分钟内)
metrics:irate(node_netstat_Tcp_EstabResets{instance~$node,job~$job}[5m])PassiveOpens - 已从 LISTEN 状态直接转换到 SYN-RCVD 状态的 TCP 平均连接数(5分钟内)
metrics:irate(node_netstat_Tcp_PassiveOpens{instance~$node,job~$job}[5m])
Network Netstat TCP Linux MIPs1. TCP Aborts / Tiemouts
type: Graph
Unit: short
Label: ConnectionsTCPAbortOnClose - 由于用户关闭中止的连接数
metrics:irate(node_netstat_TcpExt_TCPAbortOnClose{instance~$node,job~$job}[5m])TCPAbortOnData - 由于意外数据而中止的连接数
metrics:irate(node_netstat_TcpExt_TCPAbortOnData{instance~$node,job~$job}[5m])TCPAbortOnLinger - 关闭后在徘徊状态中止的连接数
metrics:irate(node_netstat_TcpExt_TCPAbortOnLinger{instance~$node,job~$job}[5m])TCPAbortOnMemory - 连接到 socket 之前中止的连接数
metrics:irate(node_netstat_TcpExt_TCPAbortOnMemory{instance~$node,job~$job}[5m])TCPAbortOnTimeout - 由于超时中止的连接数
metrics:irate(node_netstat_TcpExt_TCPAbortOnTimeout{instance~$node,job~$job}[5m])TCPAbortFailed - 由于内存不足连接中止但未发送RST的连接数
metrics:irate(node_netstat_TcpExt_TCPAbortFailed{instance~$node,job~$job}[5m])TCPTimeouts - 其他 TCP 连接超时的连接数
metrics:irate(node_netstat_TcpExt_TCPTimeouts{instance~$node,job~$job}[5m])2. TCP Delayed ACK
type: Graph
Unit: short
Label: CounterDelayedACKLocked - 由于 socket 锁定 延时ACK 进一步延迟的数量
metrics:irate(node_netstat_TcpExt_DelayedACKLocked{instance~$node,job~$job}[5m])DelayedACKLost - 快速回复 ACK 模式被激活的数量
metrics:irate(node_netstat_TcpExt_DelayedACKLost{instance~$node,job~$job}[5m])DelayedACKs - 发送延迟 AC K的数量
metrics:irate(node_netstat_TcpExt_DelayedACKs{instance~$node,job~$job}[5m])3. TCP SynCookie / Challenge
type: Graph
Unit: short
Label: Counter out (-) / in ()SyncookiesFailed - 接收的无效的 SYN cookies 的数量
metrics:irate(node_netstat_TcpExt_SyncookiesFailed{instance~$node,job~$job}[5m])SyncookiesRecv - 接收的 SYN cookies 的数量
metrics:irate(node_netstat_TcpExt_SyncookiesRecv{instance~$node,job~$job}[5m])SyncookiesSent - 发送的 SYN cookies 的数量
metrics:irate(node_netstat_TcpExt_SyncookiesSent{instance~$node,job~$job}[5m])SynChallenge - 发送的 SYNChallenge 数量
metrics:irate(node_netstat_TcpExt_TCPSYNChallenge{instance~$node,job~$job}[5m])TCPChallengeACK - 发送的 Challenge ACK 数量
metrics:irate(node_netstat_TcpExt_TCPChallengeACK{instance~$node,job~$job}[5m])4. TCP LOSS
type: Graph
Unit: short
Label: CounterTCPLossFailures - 处于 Loss 状态下的 TCP 包数量
metrics:irate(node_netstat_TcpExt_TCPLossFailures{instance~$node,job~$job}[5m])TCPLossProbeRecovery - 恢复的 TCP 丢失探测定时器的数量
metrics:irate(node_netstat_TcpExt_TCPLossProbeRecovery{instance~$node,job~$job}[5m])TCPLossProbes - 发送的 TCP 丢失探测定时器的数量
metrics:irate(node_netstat_TcpExt_TCPLossProbes{instance~$node,job~$job}[5m])TCPLossUndo - 在部分确认后拥塞窗口没有缓慢启动而恢复的数量
metrics:irate(node_netstat_TcpExt_TCPLossUndo{instance~$node,job~$job}[5m])TCPLostRetransmit - TCP 包丢失重传的数量
metrics:irate(node_netstat_TcpExt_TCPLostRetransmit{instance~$node,job~$job}[5m])5. TCP DROPS
type: Graph
Unit: short
Label: CounterListenDrops - 监听队列连接丢弃数
metrics:irate(node_netstat_TcpExt_ListenDrops{instance~$node,job~$job}[5m])LockDroppedIcmps - 因 socket 锁定而丢弃的 ICMP 数据包数量
metrics:irate(node_netstat_TcpExt_LockDroppedIcmps{instance~$node,job~$job}[5m])TCPDeferAcceptDrop - 在 SYN_RECV 状态下由 socket 接收的丢弃的 ACK 帧
metrics:irate(node_netstat_TcpExt_TCPDeferAcceptDrop{instance~$node,job~$job}[5m])TCPBacklogDrop - 由于 socket 接收队列已满丢弃的TCP数据包数量
metrics:irate(node_netstat_TcpExt_TCPBacklogDrop{instance~$node,job~$job}[5m])OutOfWindowIcmps - 由于 out-of-window 丢弃的 ICMP 包数量
metrics:irate(node_netstat_TcpExt_OutOfWindowIcmps{instance~$node,job~$job}[5m])TCPMinTTLDrop - 在 minTTL 条件下丢弃的 TCP数据包的数量
metrics:irate(node_netstat_TcpExt_TCPMinTTLDrop{instance~$node,job~$job}[5m])6. TCP Retrans
type: Graph
Unit: short
Label: CounterTCPForwardRetrans - 使用 F-RTO 重新传输丢失的数据包的数量
metrics:irate(node_netstat_TcpExt_TCPForwardRetrans{instance~$node,job~$job}[5m])TCPSlowStartRetrans - 在慢启动后重传丢失的数据包数量
metrics:irate(node_netstat_TcpExt_TCPSlowStartRetrans{instance~$node,job~$job}[5m])TCPSynRetrans - SYN-SYN/ACK重传以分解 SYN 中的重传快速/超时重传
metrics:irate(node_netstat_TcpExt_TCPSynRetrans{instance~$node,job~$job}[5m])TCPSpuriousRTOs - TCP 虚假 RTOs 数量
metrics:irate(node_netstat_TcpExt_TCPSpuriousRTOs{instance~$node,job~$job}[5m])TCPSpuriousRtxHostQueues - Times detected that the fast clone is not yet freed in tcp_transmit_skb()
metrics:irate(node_netstat_TcpExt_TCPSpuriousRtxHostQueues{instance~$node,job~$job}[5m])TCPFullUndo - 重传 undoRetransmits that undid the CWND reduction
metrics:irate(node_netstat_TcpExt_TCPFullUndo{instance~$node,job~$job}[5m])TCPRetransFail - tcp_retransmit_skb() 调用失败的数量
metrics:irate(node_netstat_TcpExt_TCPRetransFail{instance~$node,job~$job}[5m])TCPPartialUndo - 使用 Hoe heuristic 部分恢复拥塞窗口
metrics:irate(node_netstat_TcpExt_TCPPartialUndo{instance~$node,job~$job}[5m])7. TCP Pruned
type: Graph
Unit: short
Label: CounterPruneCalled - 由于 socket 缓冲区溢出而从接收队列中删除的数据包数量
metrics:irate(node_netstat_TcpExt_PruneCalled{instance~$node,job~$job}[5m])RcvPruned - 从接收队列中删除的数据包数量
metrics:irate(node_netstat_TcpExt_RcvPruned{instance~$node,job~$job}[5m])OfoPruned - 由于 socket 缓冲区溢出从无序队列中删除的数据包数量
metrics:irate(node_netstat_TcpExt_OfoPruned{instance~$node,job~$job}[5m])8. TCP Direct Copy
type: Graph
Unit: short
Label: CounterTCPDirectCopyFromBacklog - 接收的来自 accept queue 的数据包
metrics:irate(node_netstat_TcpExt_TCPDirectCopyFromBacklog{instance~$node,job~$job}[5m])TCPDirectCopyFromPrequeue - 接收的来自 TCP prequeue 的数据包
metrics:irate(node_netstat_TcpExt_TCPDirectCopyFromPrequeue{instance~$node,job~$job}[5m])9. TCP TimeWait
type: Graph
Unit: short
Label: CounterTW - 在快速计时器中完成 TIME_WAITTCP 套接字
metrics:irate(node_netstat_TcpExt_TW{instance~$node,job~$job}[5m])TWKilled - 在慢速计时器中完成 TIME_WAITTCP 套接字
metrics:irate(node_netstat_TcpExt_TWKilled{instance~$node,job~$job}[5m])TWRecycled - 按时间戳回收的 TIME_WAIT 套接字
metrics:irate(node_netstat_TcpExt_TWRecycled{instance~$node,job~$job}[5m])TCPTimeWaitOverflow - 发生 TIME_WAIT 溢出的数量
metrics:irate(node_netstat_TcpExt_TCPTimeWaitOverflow{instance~$node,job~$job}[5m])1
shell
0. TCP PAWS
type: Graph
Unit: short
Label: CounterPAWSActive - 由于 TCP 时间戳PAWS而拒绝激活的连接数
metrics:irate(node_netstat_TcpExt_PAWSActive{instance~$node,job~$job}[5m])PAWSEstab - 由于 TCP 时间戳PAWS而拒绝建立连接的数据包数量
metrics:irate(node_netstat_TcpExt_PAWSEstab{instance~$node,job~$job}[5m])PAWSPassive - 由于 TCP 时间戳PAWS而被拒绝的被动连接数
metrics:irate(node_netstat_TcpExt_PAWSPassive{instance~$node,job~$job}[5m])1
shell
1. TCP SACK
type: Graph
Unit: short
Label: CounterTCPSackRecovery - 使用 Sack 恢复丢失的包
metrics:irate(node_netstat_TcpExt_TCPSackRecovery{instance~$node,job~$job}[5m])
TCPSackRecoveryFail - 使用 Sack 恢复丢失的包失败
metrics:irate(node_netstat_TcpExt_TCPSackRecoveryFail{instance~$node,job~$job}[5m])TCPSackShiftFallback
metrics:irate(node_netstat_TcpExt_TCPSackShiftFallback{instance~$node,job~$job}[5m])TCPSackShifted
metrics:irate(node_netstat_TcpExt_TCPSackShifted{instance~$node,job~$job}[5m])TCPSackDiscard
metrics:irate(node_netstat_TcpExt_TCPSACKDiscard{instance~$node,job~$job}[5m])TCPSackFailures
metrics:irate(node_netstat_TcpExt_TCPSackFailures{instance~$node,job~$job}[5m])TCPSackMerged
metrics:irate(node_netstat_TcpExt_TCPSackMerged{instance~$node,job~$job}[5m])
TCPSACKReneging
metrics:irate(node_netstat_TcpExt_TCPSACKReneging{instance~$node,job~$job}[5m])TCPSACKReorder
metrics:irate(node_netstat_TcpExt_TCPSACKReorder{instance~$node,job~$job}[5m])1
shell
2. TCP DSACK
type: Graph
Unit: short
Label: CounterTCPDSACKIgnoredOld - 在重新传输时丢弃具有重复 SACK 的数据包
metrics:irate(node_netstat_TcpExt_TCPDSACKIgnoredOld{instance~$node,job~$job}[5m])TCPDSACKOfoRecv - 接收到无序的 DSACK 数据包
metrics:irate(node_netstat_TcpExt_TCPDSACKOfoRecv{instance~$node,job~$job}[5m])TCPDSACKOfoSent - 发送的无序的 DSACK 数据包
metrics:irate(node_netstat_TcpExt_TCPDSACKOfoSent{instance~$node,job~$job}[5m])TCPDSACKOldSent - 发送的旧 DSACKs 数据包
metrics:irate(node_netstat_TcpExt_TCPDSACKOldSent{instance~$node,job~$job}[5m])TCPDSACKRecv - 接收的 DSACK 数据包
metrics:irate(node_netstat_TcpExt_TCPDSACKRecv{instance~$node,job~$job}[5m])TCPDSACKUndo
metrics:irate(node_netstat_TcpExt_TCPDSACKUndo{instance~$node,job~$job}[5m])TCPDSACKIgnoredNoUndo
metrics:irate(node_netstat_TcpExt_TCPDSACKIgnoredNoUndo{instance~$node,job~$job}[5m])1
shell
3. TCP FastOpen / FastRetrans
type: Graph
Unit: short
Label: CounterTCPFastOpenActive - 成功的出站 TFO 连接
metrics:irate(node_netstat_TcpExt_TCPFastOpenActive{instance~$node,job~$job}[5m])TCPFastOpenActiveFail - 收到的 SYN-ACK 数据包未确认 SYN 数据包中发送的数据并导致无 SYN 数据的重传
metrics:irate(node_netstat_TcpExt_TCPFastOpenActiveFail{instance~$node,job~$job}[5m])TCPFastOpenCookieReqd - 请求设置 TFO 但没有 cookie 的入站 SYN 数据包
metrics:irate(node_netstat_TcpExt_TCPFastOpenCookieReqd{instance~$node,job~$job}[5m])TCPFastOpenListenOverflow - TFO 监听队列溢出
metrics:irate(node_netstat_TcpExt_TCPFastOpenListenOverflow{instance~$node,job~$job}[5m])TCPFastOpenPassive - 成功的入站 TFO 连接
metrics:irate(node_netstat_TcpExt_TCPFastOpenPassive{instance~$node,job~$job}[5m])TCPFastOpenPassiveFail - 带有TFO cookie 的无效的入站 SYN 数据包
metrics:irate(node_netstat_TcpExt_TCPFastOpenPassiveFail{instance~$node,job~$job}[5m])TCPFastRetrans - 丢失快速重传的数据包
metrics:irate(node_netstat_TcpExt_TCPFastRetrans{instance~$node,job~$job}[5m])1
shell
4. TCP HP
type: Graph
Unit: short
Label: CounterTCPHPAcks - 接收到的不包含数据的 Acks
metrics:irate(node_netstat_TcpExt_TCPHPAcks{instance~$node,job~$job}[5m])TCPHPHits - HP 数据包
metrics:irate(node_netstat_TcpExt_TCPHPHits{instance~$node,job~$job}[5m])TCPHPHitsToUser
metrics:irate(node_netstat_TcpExt_TCPHPHitsToUser{instance~$node,job~$job}[5m])1
shell
5. TCP ZeroWindow
type: Graph
Unit: short
Label: CounterTCPToZeroWindowAdv
metrics:irate(node_netstat_TcpExt_TCPToZeroWindowAdv{instance~$node,job~$job}[5m])TCPWantZeroWindowAdv
metrics:irate(node_netstat_TcpExt_TCPWantZeroWindowAdv{instance~$node,job~$job}[5m])TCPFromZeroWindowAdv
metrics:irate(node_netstat_TcpExt_TCPFromZeroWindowAdv{instance~$node,job~$job}[5m])1
shell
6. TCP Reorder
type: Graph
Unit: short
Label: CounterTCPFACKReorder - 如果在需要更新时判断支持FACK使用 TCPFACKReorder 计数器
metrics:irate(node_netstat_TcpExt_TCPFACKReorder{instance~$node,job~$job}[5m])TCPTSReorder - 如果是被一个partial ack确认后需要更新reorder值,使用 TCPTSReorder 计数器
metrics:irate(node_netstat_TcpExt_TCPTSReorder{instance~$node,job~$job}[5m])1
shell
7. TCP Reno
type: Graph
Unit: short
Label: CounterTCPRenoFailures - reno 后快速重传超时的数量
metrics:irate(node_netstat_TcpExt_TCPRenoFailures{instance~$node,job~$job}[5m])TCPRenoRecovery
metrics:irate(node_netstat_TcpExt_TCPRenoRecovery{instance~$node,job~$job}[5m])TCPRenoRecoveryFail
metrics:irate(node_netstat_TcpExt_TCPRenoRecoveryFail{instance~$node,job~$job}[5m])TCPRenoReorder
metrics:irate(node_netstat_TcpExt_TCPRenoReorder{instance~$node,job~$job}[5m])1
shell
8. TCP ReqQ
type: Graph
Unit: short
Label: CounterTCPReqQFullDoCookies
metrics:irate(node_netstat_TcpExt_TCPReqQFullDoCookies{instance~$node,job~$job}[5m])TCPReqQFullDrop
metrics:irate(node_netstat_TcpExt_TCPReqQFullDrop{instance~$node,job~$job}[5m])1
shell
9. TCP Out of order
type: Graph
Unit: short
Label: CounterTCPOFODrop - 在 OFO 中排队但由于达到了 socket rcvbuf 限制而丢弃的数据包
metrics:irate(node_netstat_TcpExt_TCPOFODrop{instance~$node,job~$job}[5m])TCPOFOMerge - OFO 中与其他数据包合并的数据包
metrics:irate(node_netstat_TcpExt_TCPOFOMerge{instance~$node,job~$job}[5m])TCPOFOQueue - OFO 队列的数据包
metrics:irate(node_netstat_TcpExt_TCPOFOQueue{instance~$node,job~$job}[5m])2
shell
0. TCP MD5
type: Graph
Unit: short
Label: CounterTCPMD5NotFound - 希望收到带 MD5 选项的包但是包里面没有 MD5 选项
metrics:irate(node_netstat_TcpExt_TCPMD5NotFound{instance~$node,job~$job}[5m])TCPMD5Unexpected - 不希望收到带 MD5 选项的包但是包里面有 MD5 选项
metrics:irate(node_netstat_TcpExt_TCPMD5Unexpected{instance~$node,job~$job}[5m])2
shell
1. TCP Prequeued
type: Graph
Unit: short
Label: CounterTCPPrequeued
metrics:irate(node_netstat_TcpExt_TCPPrequeued{instance~$node,job~$job}[5m])TCPPrequeueDropped - prequeue 队列丢弃的数据包
metrics:irate(node_netstat_TcpExt_TCPPrequeueDropped{instance~$node,job~$job}[5m])2
shell
2. TCP Rcv
type: Graph
Unit: short
Label: CounterTCPRcvCoalesce - 在接收队列中崩溃的数据包
metrics:irate(node_netstat_TcpExt_TCPRcvCoalesce{instance~$node,job~$job}[5m])TCPRcvCollapsed - 由于低的 socket 缓冲区在接收队列中崩溃的数据包
metrics:irate(node_netstat_TcpExt_TCPRcvCollapsed{instance~$node,job~$job}[5m])2
shell
3. TCP Original Data
type: Graph
Unit: short
Label: CounterTCPOrigDataSent - 带有原始数据的传出数据包
metrics:irate(node_netstat_TcpExt_TCPOrigDataSent{instance~$node,job~$job}[5m])2
shell
4. TCP Filters
type: Graph
Unit: short
Label: CounterArpFilter - 过滤的 Arp 数据包
metrics:irate(node_netstat_TcpExt_ArpFilter{instance~$node,job~$job}[5m])IPReversePathFilter - 从非直连网络到达的数据包
metrics:irate(node_netstat_TcpExt_IPReversePathFilter{instance~$node,job~$job}[5m])2
shell
5. TCP Pure ACK
type: Graph
Unit: short
Label: CounterTCPPureAcks - 接收到不包含的数据负载的 ACKs
metrics:irate(node_netstat_TcpExt_TCPPureAcks{instance~$node,job~$job}[5m])2
shell
6. TCP Auto Corking
type: Graph
Unit: short
Label: CounterTCPAutoCorking - Tcp 自动闭塞
metrics:irate(node_netstat_TcpExt_TCPAutoCorking{instance~$node,job~$job}[5m])2
shell
7. TCP Issues
type: Graph
Unit: short
Label: CounterBusyPollRxPackets - 低延迟应用程序获取的数据包
metrics:irate(node_netstat_TcpExt_BusyPollRxPackets{instance~$node,job~$job}[5m])EmbryonicRsts - Resets received for embryonic SYN_RECV sockets
metrics:irate(node_netstat_TcpExt_EmbryonicRsts{instance~$node,job~$job}[5m])ListenOverflows - 监听 socket 的队列溢出
metrics:irate(node_netstat_TcpExt_ListenOverflows{instance~$node,job~$job}[5m])TCPSchedulerFailed
metrics:irate(node_netstat_TcpExt_TCPSchedulerFailed{instance~$node,job~$job}[5m])TCPMemoryPressures
metrics:irate(node_netstat_TcpExt_TCPMemoryPressures{instance~$node,job~$job}[5m])
Network Netstat UDP /proc/net/snmp1. UDP In / Out
type: Graph
Unit: short
Label: Datagrams out (-) / in ()InDatagrams - 平均接收的 UDP 数据包5分钟内
metrics:irate(node_netstat_Udp_InDatagrams{instance~$node,job~$job}[5m])OutDatagrams - 平均发送的 UDP 数据包5分钟内
metrics:irate(node_netstat_Udp_OutDatagrams{instance~$node,job~$job}[5m])2. UDP Errors
type: Graph
Unit: short
Label: Datagrams out (-) / in ()InCsumErrors - 具有校验和错误的 UDP 数据包的平均数5分钟内
metrics:irate(node_netstat_Udp_InCsumErrors{instance~$node,job~$job}[5m])InErrors - 本机端口未监听之外的其他原因引起的 UDP 入包无法送达(应用层)的平均数5分钟内
metrics:irate(node_netstat_Udp_InErrors{instance~$node,job~$job}[5m])RcvbufErrors - 接收缓冲区溢出的 UDP 包的平均数5分钟内
metrics:irate(node_netstat_Udp_RcvbufErrors{instance~$node,job~$job}[5m])SndbufErrors - 发送缓冲区溢出的 UDP 包的平均数5分钟内
metrics:irate(node_netstat_Udp_SndbufErrors{instance~$node,job~$job}[5m])NoPorts - 未知端口接收 UDP 数据包的平均数5分钟内
metrics:irate(node_netstat_Udp_NoPorts{instance~$node,job~$job}[5m])3. UDP Lite In / Out
type: Graph
Unit: short
Label: Datagrams out (-) / in ()InDatagrams - 平均接收的 UDP-Lite 数据包5分钟内
metrics:irate(node_netstat_UdpLite_InDatagrams{instance~$node,job~$job}[5m])OutDatagrams - 平均发送的 UDP-Lite 数据包5分钟内
metrics:irate(node_netstat_UdpLite_OutDatagrams{instance~$node,job~$job}[5m])4. UDP Lite Errors
type: Graph
Unit: short
Label: Datagrams out (-) / in ()InCsumErrors - 具有校验和错误的 UDP-Lite 数据包的平均数5分钟内
metrics:irate(node_netstat_UdpLite_InCsumErrors{instance~$node,job~$job}[5m])InErrors - 本机端口未监听之外的其他原因引起的 UDP-Lite 入包无法送达(应用层)的平均数5分钟内
metrics:irate(node_netstat_UdpLite_InErrors{instance~$node,job~$job}[5m])RcvbufErrors - 接收缓冲区溢出的 UDP-Lite 包的平均数5分钟内
metrics:irate(node_netstat_UdpLite_RcvbufErrors{instance~$node,job~$job}[5m])SndbufErrors - 发送缓冲区溢出的 UDP-Lite 包的平均数5分钟内
metrics:irate(node_netstat_UdpLite_InErrors{instance~$node,job~$job}[5m])NoPorts - 未知端口接收 UDP-Lite 数据包的平均数5分钟内
metrics:irate(node_netstat_UdpLite_NoPorts{instance~$node,job~$job}[5m])
Network Netstat ICMP /proc/net/snmp1. ICMP Errors 1
type: Graph
Unit: short
Label: Messages out (-) / in ()InErrors - 接收的 ICMP 错误的报文例如ICMP校验和错误、长度错误等
metrics:irate(node_netstat_Icmp_InErrors{instance~$node,job~$job}[5m])OutErrors - 由于 ICMP 错误未发送的报文例如缺少缓存
metrics:irate(node_netstat_Icmp_OutErrors{instance~$node,job~$job}[5m])InDestUnreachs - 接收终点不可达的报文
metrics:irate(node_netstat_Icmp_InDestUnreachs{instance~$node,job~$job}[5m])OutDestUnreachs - 发送终点不可达的报文
metrics:irate(node_netstat_Icmp_OutDestUnreachs{instance~$node,job~$job}[5m])InType11 - 时间超时报文
metrics:irate(node_netstat_IcmpMsg_InType11{instance~$node,job~$job}[5m]). ICMP Errors 2
type: Graph
Unit: short
Label: Messages out (-) / in ()InCsumErrors - 具有校验和错误 ICMP 报文
metrics:irate(node_netstat_Icmp_InCsumErrors{instance~$node,job~$job}[5m])InTimeExcds - 接收时间超时报文
metrics:irate(node_netstat_Icmp_InTimeExcds{instance~$node,job~$job}[5m])OutTimeExcds - 发送时间超时报文
metrics:irate(node_netstat_Icmp_OutTimeExcds{instance~$node,job~$job}[5m])InParmProbs - 接收参数错误报文
metrics:irate(node_netstat_Icmp_InParmProbs{instance~$node,job~$job}[5m])OutParmProb - 发送参数错误报文
metrics:irate(node_netstat_Icmp_OutParmProbs{instance~$node,job~$job}[5m])InSrcQuenchs - 接收源点抑制报文
metrics:irate(node_netstat_Icmp_InSrcQuenchs{instance~$node,job~$job}[5m])OutSrcQuenchs - 发送源点抑制报文
metrics:irate(node_netstat_Icmp_OutSrcQuenchs{instance~$node,job~$job}[5m])3. ICMP In / Out - Messages / Redirects
type: Graph
Unit: short
Label: Messages out (-) / in ()InMsgs - 接收的报文数
metrics:irate(node_netstat_Icmp_InMsgs{instance~$node,job~$job}[5m])InRedirects - 接收的 ICMP 重定向报文
metrics:irate(node_netstat_Icmp_InRedirects{instance~$node,job~$job}[5m])OutMsgs - 发送的报文数
metrics:irate(node_netstat_Icmp_OutMsgs{instance~$node,job~$job}[5m])OutRedirects - 发送的 ICMP 重定向报文
metrics:irate(node_netstat_Icmp_OutRedirects{instance~$node,job~$job}[5m])4. ICMP Timestamps
type: Graph
Unit: short
Label: Messages out (-) / in ()InTimestampReps - 接收(应答)时间戳
metrics:irate(node_netstat_Icmp_InTimestampReps{instance~$node,job~$job}[5m])InTimestamps - 接收(请求)时间戳
metrics:irate(node_netstat_Icmp_InTimestamps{instance~$node,job~$job}[5m])utTimestampReps - 发送(应答)时间戳
metrics:irate(node_netstat_Icmp_OutTimestampReps{instance~$node,job~$job}[5m])OutTimestamps - 发送(请求)时间戳
metrics:irate(node_netstat_Icmp_OutTimestamps{instance~$node,job~$job}[5m])5. ICMP Echos
type: Graph
Unit: short
Label: Messages out (-) / in ()InEchoReps - 接收回显(应答)消息
metrics:irate(node_netstat_Icmp_InEchoReps{instance~$node,job~$job}[5m])InEchos - 接收回显(请求)消息
metrics:irate(node_netstat_Icmp_InTimestamps{instance~$node,job~$job}[5m])OutEchoReps - 发送回显(应答)消息
metrics:irate(node_netstat_Icmp_OutEchoReps{instance~$node,job~$job}[5m])OutEchos - 发送回显(请求)消息
metrics:irate(node_netstat_Icmp_OutEchos{instance~$node,job~$job}[5m])6. ICMP Masks
type: Graph
Unit: short
Label: Messages out (-) / in ()InAddrMaskReps - 接收地址掩码(应答)消息
metrics:irate(node_netstat_Icmp_InAddrMaskReps{instance~$node,job~$job}[5m])InAddrMasks - 接收地址掩码(请求)消息
metrics:irate(node_netstat_Icmp_InAddrMasks{instance~$node,job~$job}[5m])OutAddrMaskReps - 发送地址掩码(应答)消息
metrics:irate(node_netstat_Icmp_OutAddrMaskReps{instance~$node,job~$job}[5m])OutAddrMasks - 发送地址掩码(请求)消息
metrics:irate(node_netstat_Icmp_OutAddrMasks{instance~$node,job~$job}[5m])
Node Exporter1. Node Exporter Scrape Time
type: Graph
Unit: seconds
Label: Seconds{{collector}} - 各个收集器持续时间
metrics:node_scrape_collector_duration_seconds{instance~$node,job~$job}2. Node Exporter Scrape Success
type: Graph
Unit: short
Label: Counter{{collector}} - 各个收集器正常工作数量
metrics:node_scrape_collector_success{instance~$node,job~$job}
