"<filename_str>"

Enter the file name on the FortiMail unit.

<process_name>

Enter the name of a process, such as smtpd or mailstatd.

<signal_int>

Enter a standard POSIX system signal number, such as:

• 2 (SIGINT) or 15 (SIGTERM) — Quit with an interrupt signal. Normal termination for many processes. Depending on its state, the process may handle the quit gracefully such as saving state and releasing system resources instead of an immediate SIGKILL.

• 3 (SIGQUIT) — Quit. Also output a core dump file.

• 6 (SIGABRT) — Quit. Also output a crash log file. Unlike SIGSEGV, processes usually trigger this signal themselves.

• 9 (SIGKILL) — Force quit.Unlike SIGINT, the process cannot ignore SIGKILL or handle the signal gracefully. Administrators may trigger this signal to terminate zombie processes.

• 11 (SIGSEGV) — Force quit like SIGKILL. Also output the segmentation fault in the crash log file.

If a process crashes, crash logs and/or core dumps (trace files) should be automatically generated. Do send a signal in that case.

If processes have not crashed, but Fortinet Support requires detailed troubleshooting information, Support may ask you to send signal 6 or 11 to manually trigger a trace log.

<tftp-server_ipv4>

Enter the IP address of a TFTP server.

coredump

Display a kernel core dump file.

crashlog

Display a process crash log.

debug {enable | disable}

Enable or disable output of debug messages in the CLI display.

Note: diagnose debug setting duration <minutes_int> can only be entered while debugging is enabled. When you disable debugging or the duration time elapses, then the settings are reset.

duration <minutes_int>

Enter how long in minutes to record debug information.

filter "<filter_str>"

Enter text that you want to find to search and filter output to show only matching lines.

format {default | datetime | file-name | func-name | line-number | millisecond | pid | trace-name}

Select the format of the debug file. default includes both the time and PID.

grep <filter_str>

Enter text that you want to find to search and filter output to show only matching lines.

For example, you could enter grep smtpd to show only lines with the SMTP process name.

level {{0..8} | d | D | l | L | v | V}

Select an amount or category of detail to include in the debug log, either:

• a number, such as 0 or 8

• d — Debug messages.

• D — Verbose debug messages.

• l — Library level debug messages.

• L — Verbose library level debug messages.

• v — Very verbose debug messages.

• V — Very verbose library level debug messages.

To display the current detail level, enter:

diagnose debug application <process_name> display

(For diagnose debug commands for the kernel, CLI, or database, to display the current level, you omit the level number instead.)

level clear

Resets the saved verbosity level of debug information.

level display

Displays the current and saved verbosity level of debug information.

level save

Saves the current verbosity level of debug information so that it persists after reboot. To undo this, enter diagnose debug settinglevel clear.

output {default | file | stdout | terminal}

Select the display location of the command results, either:

• default — Both the terminal and a debug log file.

• file — Debug log file.

• stdout — Monitor attached to the FortiMail appliance.

• terminal — Local console or remote terminal emulator.

rating [<refresh-interval-seconds_int>]

Enter how frequently in seconds to refresh the display for the list of FortiGuard Antispam rating query servers.

If you omit the refresh interval, the CLI output gives a list of the servers and then immediately returns you to the command prompt.

rotate <files_int> <megabytes_int>

Enter the maximum number and file size in megabytes (MB) of debug files to keep.