Yandex Cloud
Search
Contact UsGet started
  • Blog
  • Pricing
  • Documentation
  • All Services
  • System Status
    • Featured
    • Infrastructure & Network
    • Data Platform
    • Containers
    • Developer tools
    • Serverless
    • Security
    • Monitoring & Resources
    • ML & AI
    • Business tools
  • All Solutions
    • By industry
    • By use case
    • Economics and Pricing
    • Security
    • Technical Support
    • Customer Stories
    • Gateway to Russia
    • Cloud for Startups
    • Education and Science
  • Blog
  • Pricing
  • Documentation
Yandex project
© 2025 Yandex.Cloud LLC
Yandex SpeechKit Hybrid
  • System requirements
  • Service architecture
    • Service settings
    • Monitoring SpeechKit Hybrid services
    • Envoy monitoring
    • Logging levels
    • TLS termination
  • Pricing policy
  • SpeechKit Hybrid releases

In this article:

  • License server metrics
  • Common metrics
  • Metrics for the Cloud Billing license model
  • Metrics for the Yubikey license model
  • Envoy metrics
  1. Admin SpeechKit Hybrid
  2. Envoy monitoring

Envoy service monitoring

Written by
Yandex Cloud
Updated at July 8, 2024
  • License server metrics
    • Common metrics
    • Metrics for the Cloud Billing license model
    • Metrics for the Yubikey license model
  • Envoy metrics

Envoy services is a collective name of the Envoy and License server SpeechKit Hybrid components For more information about the SpeechKit Hybrid components, see SpeechKit Hybrid architecture.

Envoy services collect and store metrics in Prometheus format. For more information about metric types, see the Prometheus documentation.

License server metricsLicense server metrics

License server metrics are available at the URL:

<IP address>:<port>

Where:

  • <IP address>: IP address of the License server SpeechKit Hybrid component in your network.
  • <port>: Prometheus port. The default port is 8003, you can change it in SpeechKit Hybrid settings.

Common metricsCommon metrics

Metric Type Description Labels
active_services GAUGE Number of running SpeechKit Hybrid services type, service type: stt/tts.
log4j2_appender_total COUNTER Number of logs with this level. level, the logging level.
grpc_statuses COUNTER gRPC method operating states app (application), method (gRPC method), status (query status). To track the connection with Yandex Billing, in the method label, specify the values: yandex.cloud.ds.billing.BillingService/Register, yandex.cloud.ds.billing.BillingService/SendBillingData.
grpc_durations HISTOGRAM Histogram of the gRPC query processing time app (application), method (gRPC method), status (query status).
grpc_requests GAUGE Various metrics of gRPC queries app (application), method (gRPC method), status (query status).

Metrics for the Cloud Billing license modelMetrics for the Cloud Billing license model

Metric Type Description Labels
billing_agent_billing_units_sent COUNTER Count of billing packets sent to Yandex Billing unit_type, service type: stt/tts.
billing_agent_requests_sent COUNTER Number of queries sent to the Cloud Billing server SpeechKit Hybrid
billing_agent_entries_sent COUNTER Number of billing records sent
billing_agent_queue_size GAUGE Current Cloud Billing queue size SpeechKit Hybrid
billing_agent_billing_units_received COUNTER Number of billing packets received from stt/tts services unit_type, service type: stt/tts.
billing_agent_last_entry_serial GAUGE Serial number of the last billing record sent
billing_agent_last_entry_unixtime GAUGE Standard Unix time of the last billing record sent

Metrics for the Yubikey license modelMetrics for the Yubikey license model

Metric Type Description Labels
license_active GAUGE License status (0 — for inactive license, 1 — for active license)
license_parameters GAUGE License options type, parameter type: max_stt_connections: Maximum number of connections to stt service; max_tts_rps: Maximum RPS for the tts service

Envoy metricsEnvoy metrics

To get information about the Envoy service status, use the administrative interface at the URL:

<IP address>:9091

Where:

  • <IP address>: IP address of the Envoy SpeechKit Hybrid service in your network.

To learn more about the interface, see the Envoy documentation.

Envoy metrics in Prometheus format are available via the administrative interface at the URL:

<IP address>:9091/stats/prometheus

Note

To get only significant metrics that changed at least once, use the URL: <IP address>:9091/stats?format=prometheus&usedonly.

Informative Envoy metrics for SpeechKit Hybrid monitoring:

  • Upstream metrics for Envoy communications with the stt/tts service clusters.

  • Downstream metrics for Envoy communications with external resources. For example:

    Metric Type Description Labels
    downstream_cx_total COUNTER Number of HTTP connections during the service operation envoy_http_conn_manager_prefix
    downstream_cx_active GAUGE Number of active HTTP connections. envoy_cluster_name, cluster name: upstream_stt, upstream_tts for SpeechKit Hybrid services
    downstream_cx_rx_bytes_total COUNTER Amount of information received (in bytes) during the service operation envoy_http_conn_manager_prefix
    downstream_cx_tx_bytes_total COUNTER Amount of information sent (in bytes) during the service operation envoy_http_conn_manager_prefix
    downstream_rq_total COUNTER Numbers of queries sent during the service operation envoy_http_conn_manager_prefix
    downstream_rq_active GAUGE Number of active queries. envoy_http_conn_manager_prefix

Was the article helpful?

Previous
Monitoring SpeechKit Hybrid services
Next
Logging levels
Yandex project
© 2025 Yandex.Cloud LLC