Logs

Every service, and StartOS itself, streams its logs live in the web interface. You can read them there, scroll back through older entries, and download a copy to keep or send to support. Nothing needs to be installed, and you do not need SSH or the command line.

Service Logs

Open the service from the dashboard and select the Logs tab. New lines appear as the service writes them.

This is where a misbehaving service explains itself: a failed health check, a backend it cannot reach, a setting it rejected, or a crash on startup.

OS and Kernel Logs

Go to System:

• OS Logs — StartOS itself: the daemon, service supervision, networking, and updates.
• Kernel Logs — the Linux kernel: hardware, disks, drivers, and out-of-memory events.

Reading a Log

The view follows the end of the log, so new lines appear as they are written. Scroll up to read older entries — more are loaded as you reach the top. Scroll to bottom returns to the end and resumes following.

Downloading Logs

Download, at the bottom of any log view, saves the most recent 10,000 entries as an HTML file that opens in any browser, colored the same as it appears onscreen. The file is named for what it came from — <service-id>-logs.html, os-logs.html, or kernel-logs.html.

Attach that file when you ask for help. It carries far more than the last error onscreen, and it is the fastest way to get an answer.

From the Command Line

The web interface is the primary way to read logs. start-cli covers the cases it cannot reach — a server that will not boot far enough to serve the UI, or a log you want to pipe into another tool:

start-cli package logs <ID>
start-cli server logs
start-cli server kernel-logs

In Diagnostic Mode, where StartOS has failed to start normally, the diagnostic page offers View logs. There is no download button there, so start-cli diagnostic logs is how to capture those. See the FAQ.

Log Level

Client and service connection failures, including refused backend ports and TLS errors, are logged at trace level rather than filling normal OS logs. Set the running daemon’s verbosity with start-cli server set-log-level <LEVEL>, choosing error, warn, info, debug, or trace:

start-cli server set-log-level trace

Read the output in System → OS Logs or with start-cli server logs --follow. The selected level applies only to start-core/start-os code. Dependency logs keep their startup filter, which shows warnings and errors by default. Trace output can still contain sensitive details and substantial volume. Return to the packaged backend default after diagnosis:

start-cli server set-log-level debug

The setting lasts until the daemon restarts, when its startup filter, including any explicit RUST_LOG settings, is restored. More-specific startup directives retain their usual precedence. Explicit debug/trace directives for HTTP/WebSocket dependencies can feed their diagnostics back into followed logs. The command requires the same authentication as other server commands and changes no saved configuration.