System
System 采集器收集系统负载、正常运行时间、CPU 核心数量以及登录的用户数。
配置¶
成功安装 DataKit 并启动后，会默认开启 System 采集器，无需手动开启。
采集器配置¶
进入 DataKit 安装目录下的 conf.d/host 目录，复制 system.conf.sample 并命名为 system.conf。示例如下：
[[inputs.system]]
##(optional) collect interval, default is 10 seconds
interval = '10s'
[inputs.system.tags]
# some_tag = "some_value"
# more_tag = "some_other_value"
配置好后，重启 DataKit 即可。
可通过 ConfigMap 方式注入采集器配置 或 配置 ENV_DATAKIT_INPUTS 开启采集器。
也支持以环境变量的方式修改配置参数（需要在 ENV_DEFAULT_ENABLED_INPUTS 中加为默认采集器）：
-
ENV_INPUT_SYSTEM_INTERVAL
采集器重复间隔时长
字段类型: Duration
采集器配置字段:
interval默认值: 10s
-
ENV_INPUT_SYSTEM_TAGS
自定义标签。如果配置文件有同名标签，将会覆盖它
字段类型: Map
采集器配置字段:
tags示例: tag1=value1,tag2=value2
指标¶
以下所有数据采集，默认会追加名为 host 的全局 tag（tag 值为 DataKit 所在主机名），也可以在配置中通过 [inputs.system.tags] 指定其它标签：
system¶
Basic information about system operation.
- 标签
| Tag | Description |
|---|---|
| host | hostname |
- 指标列表
| Metric | Description |
|---|---|
| cpu_total_usage | The percentage of used CPU. Type: float Unit: percent,percent |
| load1 | CPU load average over the past 1 minute. Type: float Unit: N/A |
| load15 | CPU load average over the past 15 minutes. Type: float Unit: N/A |
| load15_per_core | CPU single core load average over the past 15 minutes. Type: float Unit: N/A |
| load1_per_core | CPU single core load average over the past 1 minute. Type: float Unit: N/A |
| load5 | CPU load average over the past 5 minutes. Type: float Unit: N/A |
| load5_per_core | CPU single core load average over the last 5 minutes. Type: float Unit: N/A |
| memory_usage | The percentage of used memory. Type: float Unit: percent,percent |
| n_cpus | CPU logical core count. Type: int Unit: count |
| n_users | User number. Type: int Unit: count |
| process_count | Number of Processes running on the machine. Type: int Unit: count |
| uptime | System uptime. Type: int Unit: time,s |
conntrack¶
Connection track metrics (Linux only).
- 标签
| Tag | Description |
|---|---|
| host | hostname |
- 指标列表
| Metric | Description |
|---|---|
| entries | Current number of connections. Type: int Unit: count |
| entries_limit | The size of the connection tracking table. Type: int Unit: count |
| stat_drop | The number of packets dropped due to connection tracking failure. Type: int Unit: count |
| stat_early_drop | The number of partially tracked packet entries dropped due to connection tracking table full. Type: int Unit: count |
| stat_found | The number of successful search entries. Type: int Unit: count |
| stat_ignore | The number of reports that have been tracked. Type: int Unit: count |
| stat_insert | The number of packets inserted. Type: int Unit: count |
| stat_insert_failed | The number of packages that failed to insert. Type: int Unit: count |
| stat_invalid | The number of packets that cannot be tracked. Type: int Unit: count |
| stat_search_restart | The number of connection tracking table query restarts due to hash table size modification. Type: int Unit: count |
filefd¶
System file handle metrics (Linux only).
- 标签
| Tag | Description |
|---|---|
| host | hostname |
- 指标列表
| Metric | Description |
|---|---|
| allocated | The number of allocated file handles. Type: int Unit: count |
| maximum_mega | The maximum number of file handles, unit M(10^6). Type: float Unit: count |
FAQ¶
为什么没有 cpu_total_usage 指标？¶
CPU 部分采集功能不支持部分平台，如 macOS。