Get started
Yandex Serverless Containers

Viewing container monitoring charts

Written by

You can monitor containers at runtime using the monitoring tools in your management console. These tools display diagnostic information in the form of charts. Metric values are collected and charts are displayed by Yandex Monitoring.

The chart update period is 15 seconds.

Viewing monitoring charts

  1. In the management console, select the folder with your container.

  2. Select Serverless Containers.

  3. Select a container to view its monitoring charts.

  4. Go to the Monitoring tab.

  5. The following charts open on the page:

    • Requests: number of container calls.
    • Requests latency: average time for a container to process a call.
    • Errors: number of errors generated while a container was processing a call.

    You can select the period to display information for: an hour, three hours, a day, week, month, or a custom interval.

Custom metrics

To proceed to working with metrics, dashboards, and alerts, in Yandex Monitoring, in the upper-right corner, click Open in Monitoring.

Metric name Units Explanations
serverless.containers.inflight Invocations Number of concurrent container invocations.
serverless.containers.started_per_second Calls per second Frequency of container invocations.
serverless.containers.finished_per_second Calls per second Frequency of completing container invocation processing.
serverless.containers.errors_per_second Errors per second Frequency of errors when processing container invocations.
serverless.containers.initializations_per_second Initializations per second Frequency of initializing new container instances.
serverless.containers.execution_time_milliseconds Calls per second Histogram of the container invocation frequency distribution by request processing time in milliseconds. Request processing time intervals are shown in the bin label.
serverless.containers.used_memory_bytes Calls per second Histogram of the container invocation frequency distribution by RAM used in bytes. Intervals of the RAM used by the request are shown in the bin label.

Custom metrics labels

For more information about the labels concept, see the Yandex Monitoringdocumentation.

Label name Possible values Explanations
container Container name The label contains the name of the container that the metric values pertain to.
revision Revision ID The label contains the container revision ID that the metric values pertain to
bin Interval of histogram values For metrics represented by a histogram, the label value describes the interval of the measured value during which the event occurred.
In this article: