Tracer Startup Logs

Startup logs

Tracer startup logs capture all obtainable information at startup and log it as DATADOG TRACER CONFIGURATION, DATADOG TRACER DIAGNOSTICS, DATADOG ERROR, or DATADOG CONFIGURATION to simplify searching within your logs.

Some languages log to a separate file depending on language conventions and the safety of accessing Stdout or equivalent. In those cases, the location of logs are noted in the language tab below. Some languages don’t log diagnostics entries, also noted below.

CONFIGURATION logs are a JSON formatted representation of settings applied to your tracer. In languages where an Agent connectivity check is performed, the configuration JSON will also include an agent_error key, which indicates whether the Agent is reachable.

DIAGNOSTICS or ERROR log entries, in the languages that produce them, happen when the tracer encounters an error during application startup. If you see DIAGNOSTICS or ERROR log lines, confirm from the indicated log that settings and configurations are applied correctly.

If you do not see logs at all, ensure that your application logs are not silenced and that your log level is at least INFO where applicable.

PHP Info: Obtain the startup logs JSON string from a phpinfo() page next to “DATADOG TRACER CONFIGURATION”. Create the following PHP file and access it from a browser on the host machine.

<?php phpinfo(); ?>

Diagnostic information is displayed in a separate table to help diagnose common issues.

PHP Info

CLI SAPI:

Get the info from the CLI SAPI by running php --ri=ddtrace.

ddtrace


Datadog PHP tracer extension
For help, check out the documentation at https://docs.datadoghq.com/tracing/languages/php/
(c) Datadog 2020

Datadog tracing support => enabled
Version => 1.0.0-nightly
DATADOG TRACER CONFIGURATION => {"date":"2020-07-01T17:43:54Z","os_name":"Linux 49b1cb4bdd12 4.19.76-linuxkit #1 SMP Tue May 26 11:42:35 UTC 2020 x86_64","os_version":"4.19.76-linuxkit","version":"1.0.0-nightly","lang":"php","lang_version":"7.4.5","env":null,"enabled":true,"service":null,"enabled_cli":false,"agent_url":"https://localhost:8126","debug":false,"analytics_enabled":false,"sample_rate":1.000000,"sampling_rules":null,"tags":null,"service_mapping":null,"distributed_tracing_enabled":true,"priority_sampling_enabled":true,"dd_version":null,"architecture":"x86_64","sapi":"cli","ddtrace.request_init_hook":null,"open_basedir_configured":false,"uri_fragment_regex":null,"uri_mapping_incoming":null,"uri_mapping_outgoing":null,"auto_flush_enabled":false,"generate_root_span":true,"http_client_split_by_domain":false,"measure_compile_time":true,"report_hostname_on_root_span":false,"traced_internal_functions":null,"auto_prepend_file_configured":false,"integrations_disabled":null,"enabled_from_env":true,"opcache.file_cache":null,"agent_error":"Couldn't connect to server","ddtrace.request_init_hook_reachable":false}

                               Diagnostics
agent_error => Couldn't connect to server
ddtrace.request_init_hook_reachable => false

Directive => Local Value => Master Value
ddtrace.disable => Off => Off
...

Configuration:

If the tracer is in DEBUG mode, the startup logs will appear in the error_log once per process on the first request.

DATADOG TRACER CONFIGURATION - {"agent_error":"Couldn't connect to server","ddtrace.request_init_hook_reachable":false,"date":"2020-07-01T17:42:50Z","os_name":"Linux 49b1cb4bdd12 4.19.76-linuxkit #1 SMP Tue May 26 11:42:35 UTC 2020 x86_64","os_version":"4.19.76-linuxkit","version":"1.0.0-nightly","lang":"php","lang_version":"7.4.5","env":null,"enabled":true,"service":null,"enabled_cli":false,"agent_url":"https://localhost:8126","debug":false,"analytics_enabled":false,"sample_rate":1.000000,"sampling_rules":null,"tags":null,"service_mapping":null,"distributed_tracing_enabled":true,"priority_sampling_enabled":true,"dd_version":null,"architecture":"x86_64","sapi":"cgi-fcgi","ddtrace.request_init_hook":null,"open_basedir_configured":false,"uri_fragment_regex":null,"uri_mapping_incoming":null,"uri_mapping_outgoing":null,"auto_flush_enabled":false,"generate_root_span":true,"http_client_split_by_domain":false,"measure_compile_time":true,"report_hostname_on_root_span":false,"traced_internal_functions":null,"auto_prepend_file_configured":false,"integrations_disabled":null,"enabled_from_env":true,"opcache.file_cache":null}

Diagnostics:

Failed diagnostics for the PHP tracer print in the error_log if the tracer is in DEBUG mode.

DATADOG TRACER DIAGNOSTICS - agent_error: Couldn't connect to server
DATADOG TRACER DIAGNOSTICS - ddtrace.request_init_hook_reachable: false

Runtime:

Access the startup logs as a JSON string at runtime with \DDTrace\startup_logs().

echo \DDTrace\startup_logs() . PHP_EOL;

Connection errors

If your application or startup logs contain DIAGNOSTICS errors or messages that the Agent cannot be reached or connected to (varying depending on your language), it means the tracer is unable to send traces to the Datadog Agent.

If you have these errors, check that your Agent is set up to receive traces for ECS, Kubernetes, Docker or any other option, or contact support to review your tracer and Agent configuration.

See Connection Errors for information about errors indicating that your instrumented application cannot communicate with the Datadog Agent.

Configuration settings

If your logs contain only CONFIGURATION lines, a useful troubleshooting step is to confirm that the settings output by the tracer match the settings from your deployment and configuration of the Datadog Tracer. Additionally, if you are not seeing specific traces in Datadog, review the Compatibility Requirements section of the documentation to confirm these integrations are supported.

If an integration you are using is not supported, or you want a fresh pair of eyes on your configuration output to understand why traces are not appearing as expected in Datadog, contact support who can help you diagnose and create a Feature Request for a new integration.

Disabling startup logs

For each language, you can disable startup logs by setting the environment variable DD_TRACE_STARTUP_LOGS=false, but do this only if the logs emitted are posing a problem. If later you are sending debug logs, remember to enable startup logs and send all relevant logs together to speed up your support case triage.

Further Reading

Additional helpful documentation, links, and articles:

Troubleshooting APM Connection ErrorsDOCUMENTATION more more
PREVIEWING: guacbot/translation-pipeline