RPC

HTTP RPC enables you to manage your Fluentd instance through HTTP endpoints. You can use this feature as a replacement of Unix signals.

It is especially useful for environments where signals are not supported well e.g. Windows. This requires Fluentd to start not with --no-supervisor command-line option.

Configuration

By default, HTTP RPC is not enabled. To use this feature, set the rpc_endpoint option:

<system>
  rpc_endpoint 127.0.0.1:24444
</system>

Now, you can manage your Fluentd instance using an HTTP client:

$ curl http://127.0.0.1:24444/api/plugins.flushBuffers
{"ok":true}

As evident from the output above, each endpoint returns a JSON object as its response.

HTTP Endpoints

| Endpoint | Replacement of | Description | Version | | :--- | :---: | :---: | | | /api/processes.interruptWorkers | SIGINT | Stops the daemon. | v1.0 | | /api/processes.killWorkers | SIGTERM | Stops the daemon. | v1.0 | | /api/processes.zeroDowntimeRestart | SIGUSR2 | Restarts Fluentd with zero-downtime. (Not supported on Windows) | v1.18 | | /api/processes.flushBuffersAndKillWorkers | SIGUSR1 and SIGTERM | Flushes buffer and stops the daemon. | v1.0 | | /api/plugins.flushBuffers | SIGUSR1 | Flushes the buffered messages. | v1.0 | | /api/config.reload | SIGHUP | Reloads configuration. | v1.0 | | /api/config.gracefulReload | --- | Reloads configuration. | v1.9 |

Appendix:

• /api/config.gracefulReload: This is the replacement of SIGUSR2 before v1.18. Please use /api/processes.zeroDowntimeRestart or /api/config.reload unless there is a special reason. See SIGUSR2 for details.

If this article is incorrect or outdated, or omits critical information, please let us know. Fluentd is an open-source project under Cloud Native Computing Foundation (CNCF). All components are available under the Apache 2 License.