Project

General

Profile

Actions

Feature #17

closed

LED support on ALIX, WRAP, etc.

Added by Jim Pingle over 15 years ago. Updated about 15 years ago.

Status:
Closed
Priority:
Very Low
Assignee:
Category:
Operating System
Target version:
Start date:
06/22/2009
Due date:
% Done:

100%

Estimated time:
Plus Target Version:
Release Notes:

Description

There is now hardware support for LEDs on several embedded platforms, but there isn't much to do with them yet.

A forum user wrote (a long time ago) a daemon in C that will make tie an LED to network activity for a given interface:
http://forum.pfsense.org/index.php/topic,3454.msg21194.html#msg21194

It still compiles and runs on a new ALIX, so I'd like to eventually make a UI for that portion.

I was also thinking that in 2.0 we might be able to have LED1 flash when writing the config, or perhaps blink or flash a pattern when there is a system alert. (Morse code is part of FreeBSD's LED driver, you can echo 'm...---...' > /dev/led/led3 and it will repeatedly flash that pattern.)

Actions #1

Updated by Jim Pingle over 15 years ago

  • Assignee set to Jim Pingle
Actions #2

Updated by Jim Pingle over 15 years ago

I see that there is already code which blinks on a pending alert, so that's already in and taken care of.

I'll probably write a small set of functions to wrap the LED support up a little, make it easier to call from other places.

Actions #3

Updated by Jim Pingle over 15 years ago

  • % Done changed from 20 to 70

I committed a LED wrapper library for 2.0, and I plan on making some random hooks for alerts here and there.

I also committed a blinkled package that will allow people to tie LED #2 and #3 to network interface activity.

Actions #4

Updated by Jim Pingle over 15 years ago

So I don't forget this later, in 2.0 I have it strobe on bootup but if there is an alert during the boot sequence the pattern gets messed up. The alerts should probably be on LED#1 anyhow.

Also, it may be a good idea to led_normalize() if it drops into interface assigment, but I need to test some more.

Actions #5

Updated by Jim Pingle about 15 years ago

  • Status changed from New to Closed
  • % Done changed from 70 to 100
  • Affected Version set to 2.0

I have already addressed the alert issue, and the interface assignment issue is apparently either fixed or no longer a concern (it worked when I just tried it).

Seeing as the bootup sequence and alerts are working, and the basic LED controls (and underlying library) are all there, it should be sufficient for 2.0.

The blinkled functionality for showing network activity on LEDs 2 and 3 can remain a package. It installs and runs fine on NanoBSD, I don't really see a need for it to be in the base system at this point.

Actions

Also available in: Atom PDF