Contact UsGet started
© 2024 Direct Cursus Technology L.L.C.

Yandex Application Load Balancer log reference

Written by
Updated at November 21, 2024

This section describes the various fields of the logs generated by the L7 load balancer.

Logs are provided in JSON format. A single log entry represents one client request to the load balancer.

For more information, see Logging.

Field Description
type Request type. The possible values are:
  • http: HTTP/1.0 or HTTP/1.1 without encryption.
  • https: HTTP/1.0 or HTTP/1.1 with TLS encryption (HTTPS or HTTP over TLS).
  • h2: HTTP/2.
  • h3: HTTP3.
  • grpc: gRPC without encryption.
  • grpcs: gRPC with TLS encryption.
time Request date and time: receipt of the first byte from the client by the load balancer. Written in the following ISO 8601 format: <YYYY>-<MM>-<DD>T<hh>:<mm>:<ss>Z, where Z stands for universal coordinated time, UTC.
http_method HTTP request method.
authority For HTTP/1.0 and HTTP/1.1: Host header value. For other supported protocols: :authority pseudo-header value.
request_uri Request path. For example, /api/v1/foo.
http_status HTTP response (status) code.
http_version HTTP version, e.g., HTTP/2.
rewritten_request_uri Request path following the substitution configured in the HTTP router's route. For example, /api/backend/v1/foo.
request_content_type Content-Type HTTP header value in the client's request: requested content type.
request_id X-Request-ID HTTP header value in the client's request.
user_agent User-Agent HTTP header value: client app ID string.
referer Referer HTTP header value in the client's request: request source URL.
response_content_type Content-Type HTTP header value in the load balancer's response: content requested type.
client_ip Client IP address.
client_port Client TCP port used by the connection between the client and the load balancer.
backend_ip Internal backend endpoint IP address: virtual machine from the target group or Object Storage bucket.
backend_port Port used by the load balancer to access the backend.
load_balancer_id ID of the network load balancer.
http_router_id HTTP router ID.
virtual_host_name Virtual host name.
route_name Route name.
backend_group_id Backend group ID.
backend_name Backend name.
request_headers_bytes Request header size in bytes.
request_body_bytes Request body size in bytes.
response_headers_bytes Response header size in bytes.
response_body_bytes Response body size in bytes.
error_details Error encountered by the load balancer when processing a request. The possible values are:
  • no_healthy_backend: No healthy backends. For more information, see Health checks.
  • backend_request_timeout: Backend request timeout specified in the route settings was exceeded.
  • balancer_reset: Connection between the client and the load balancer was reset.
  • backend_connection_closed_before_data_sent_to_client: Connection between the load balancer and the backend was reset.
  • failed_to_connect_to_backend: Load balancer failed to connect to the backend.
  • backend_connection_terminated: Backend closed the connection with the load balancer.
  • no_route: Client's request does not match any routes configured in the load balancer HTTP routers.
  • client_disconnected: Client closed the connection with the load balancer.
  • backend_stream_idle_timeout: Idle timeout specified in the route settings was exceeded: the connection between the load balancer and the backend was idle for too long with no data transmission.
  • backend_retry_limit_exceeded: Load balancer exceeded the maximum number of backend connection attempts.
  • client_protocol_error: Client protocol error.
tls_version TLS version, e.g., TLSv1.3.
cipher_suite Set of ciphers used for the TLS encryption of the connection between client and load balancer.
sni_hostname Domain name sent to the load balancer by the client using the TLS Server Name Indication (SNI) extension.
server_certificate_subject Domain name on the TLS certificate sent to the client by the load balancer.
grpc gRPC request information. Returned for gRPC requests only ("type": "grpc" or "type": "grpcs").
  method gRPC request method.
  service gRPC service name.
  status gRPC request status.
   code gRPC request error code. For more information about the gRPC statuses that correspond to HTTP codes, see google.rpc.Code.
   message gRPC request error description.
upstream_transport_failure_reason Error message regarding a transport-layer connection error between the load balancer and backend, such as in the TLS connection.
request_processing_times Information about the time it took to process a request. All field values are specified in seconds.
  request_time Total time it took to handle a request.
  response_start_time Time before the start of a response: between when the first request byte is received from the client by the load balancer and when the first byte of the response is sent by the load balancer to the client.
  request_processing_time Time the load balancer takes to process a request: between receiving the first byte from the client and sending the first byte to the backend.
  backend_processing_time Time it takes the backend to process a request: between the load balancer sending the first byte to and receiving the first byte from the backend.
  response_processing_time Time it takes the load balancer to process a response: between receiving the first byte from the backend and sending the first byte to the client.
  request_rx_time Time it takes the load balancer to receive a request: between the first and the last bytes received from the client.
  request_tx_time Time it takes the backend to receive a request: between the first and the last bytes sent by the load balancer.
  backend_response_time Wait time for a backend response: between the load balancer sending the last byte of a request to and receiving the first byte of the response from the backend.
  response_rx_time Time it takes the backend to send a response: between the first and the last bytes received by the load balancer.
  response_tx_time Time it takes the load balancer to send a response: between the first and the last bytes sent by the load balancer.
x_forwarded_for Value of the incoming request X-Forwarded-For (XFF) header containing the IP addresses of the forwarding nodes. client_ip comes first, followed by IP addresses of any intermediate nodes, if any. For example:
  • x_forwarded_for: 50.0.0.1: Request came directly from a single client. Application Load Balancer added its client_ip to the header.
  • x_forwarded_for: 50.0.0.1,40.0.0.1: Proxy server was used for the request. Application Load Balancer added the proxy IP address after client_ip in the header.
Previous
Audit Trails events
Next
Release notes