LED activates only when probed with meter


I am kind of an EE novice. I have a few charlieplexed LEDs that I am soldering on to a board I did not design. The four charlieplexing pins lead to a resistor array which then goes to “the charlieplexing”. I wanted to test so I simply soldered in LED 2. I got code loaded on the atmega328p(tqfp package) and a clock soldered in. when I power it up, the pins voltage is going up and down, but the LED is not activating. What is weird, is that it works when I probe the activation pins with a multimeter. I tried to bridge it with just a wire and it doesn’t activate, only when I use a multimeter. the black marks on the attached file are where I am probing on the board.

boolean ledActive = false;

void setup() {                
  pinMode(5, OUTPUT);
  pinMode(6, INPUT);
  pinMode(7, OUTPUT);
  pinMode(8, OUTPUT);

// the loop routine runs over and over again forever:
void loop() {
void toggleLED(){
  ledActive = !ledActive;

Probing puts a (small) mechanical strain on the solder joints, maybe restoring contact. They may suffer from a bad (solder) contact (takes a trained eye to recognize), probably right there at the resistor array. If you can, resolder them (or return the board in case a warranty is still valid).

"Probing" with an ohmmeter can forward bias a P-N junction, too.

Absolutely true, and most mulitmeters also have ohm ranges.
As OP spoke of a multimeter, i assumed he was measuring voltages, but assumption is the mother of all …ups.
So what were you measuring ?