无法查看通过 opentelemetry 发送到控制台的指标

Question

我正在尝试检查 opentelemetry 导出器是否发送了指标，并且我想首先在控制台上测试它，如本页所述 https://opentelemetry.io/docs/languages/python/exporters/#console ,

所以我做了这段代码：

from opentelemetry.sdk.resources import SERVICE_NAME, Resource

from opentelemetry import metrics
from opentelemetry.sdk.metrics import MeterProvider
from opentelemetry.sdk.metrics.export import PeriodicExportingMetricReader, ConsoleMetricExporter

resource = Resource(attributes={
    SERVICE_NAME: "your-service-name"
})

reader = PeriodicExportingMetricReader(ConsoleMetricExporter())
meterProvider = MeterProvider(resource=resource, metric_readers=[reader])
metrics.set_meter_provider(meterProvider)
meter = metrics.get_meter()
counter = meter.create_counter("my_metric", description="Example metric sent to console", unit="1")

l = [1, 2, 3, 4, 5]
for value in l:
    counter.add(value)

但我无法将指标发送到标准输出。这就是我所看到的：

{
    "resource_metrics": [
        {
            "resource": {
                "attributes": {
                    "service.name": "your-service-name"
                },
                "schema_url": ""
            },
            "scope_metrics": [
                {
                    "scope": {
                        "name": "__main__",
                        "version": "",
                        "schema_url": "",
                        "attributes": null
                    },
                    "metrics": [
                        {
                            "name": "my_metric",
                            "description": "Example metric sent to console",
                            "unit": "1",
                            "data": {
                                "data_points": [
                                    {
                                        "attributes": {},
                                        "start_time_unix_nano": 1731532392141560100,
                                        "time_unix_nano": 1731532392141560100,
                                        "value": 15
                                    }
                                ],
                                "aggregation_temporality": 2,
                                "is_monotonic": true
                            }
                        }
                    ],
                    "schema_url": ""
                }
            ],
            "schema_url": ""
        }
    ]
}

出了什么问题？

Answer 1

您似乎确实看到了指标。您有一个 data_point，其值为 15 - 是 [1, 2, 3, 4, 5] 的累积值。这些值会累积，因为您正在使用计数器。

"metrics": [
              {
                "name": "my_metric",
                "description": "Example metric sent to console",
                "unit": "1",
                "data": {
                  "data_points": [
                    {
                      "attributes": {},
                      "start_time_unix_nano": 1731532392141560100,
                      "time_unix_nano": 1731532392141560100,
                      "value": 15
                    }
                  ],
                  "aggregation_temporality": 2,
                  "is_monotonic": true
                }
              }
            ]

您可以将阅读器配置为每 0.5 秒导出一次，然后通过包含 time.sleep(1) 每秒添加到计数器中，以观察值的增加：

reader = PeriodicExportingMetricReader(ConsoleMetricExporter(), export_interval_millis=500)
meterProvider = MeterProvider(resource=resource, metric_readers=[reader])
metrics.set_meter_provider(meterProvider)
meter = metrics.get_meter("your-service-name")
counter = meter.create_counter("my_metric", description="Example metric sent to console", unit="1")

l = [1, 2, 3, 4, 5]
for value in l:
    counter.add(value)
    time.sleep(1)

然后你的出口商将生产类似的东西：

"data_points": [
             {
               "attributes": {},
               "start_time_unix_nano": 1731545432747512000,
               "time_unix_nano": 1731545433250621000,
               "value": 1,
               "exemplars": []
             }
],
...
"data_points": [
             {
               "attributes": {},
               "start_time_unix_nano": 1731545432747512000,
               "time_unix_nano": 1731545433757252000,
               "value": 3,
               "exemplars": []
             }
],
...
"data_points": [
             {
               "attributes": {},
               "start_time_unix_nano": 1731545432747512000,
               "time_unix_nano": 1731545434259877000,
               "value": 3,
               "exemplars": []
             }
],
...
"data_points": [
             {
               "attributes": {},
               "start_time_unix_nano": 1731545432747512000,
               "time_unix_nano": 1731545434763509000,
               "value": 6,
               "exemplars": []
             }
],

随着你的计数器累积。

如果您想观察样本数据中的值

[1, 2, 3, 4, 5]

，您可以使用带有 gauge.set(value) 的仪表而不是计数器。