Heyo!
Ok, thing of it like this for the first one (I had some homework on it and got it wrong, but in the end it made sense, so I might not be entirely factual but yah);
The first pendulum will swing for 1.9 seconds, and will reach its start 0.1 seconds ahead of the second. After two swings, it'll be 0.2 seconds ahead, 3 swings 0.3, 4 swings? 0.4.
The second one is 0.1 behind, and it'll be 0.1 behind for the first, 0.2 for the second, 0.3 for the third, 0.4 for the fourth. After 5 swings, we can see that the 1.9 second pendulum is approximately 0.5 seconds out of phase with the second one, in terms of being ahead, and relatively the 2 second pendulum is 0.5 seconds behind. So, they're not in phase. So, yah, we can see a pattern arising here. We need to find a time where they are in phase.
We've taken 0.1 seconds as the phase difference, after all 2-1.9 is 0.1. Each swing, Pendulum 1 is 0.1 seconds ahead of Pendulum 2. We need to figure out how many of these out-of-phase revolutions it'll take. So, if you've followed me so far, you can see that after 20 revolutions of pendulum 1, we'd be 20*0.1 seconds out, 2 seconds, which is in fact the time period of the second one; we're in phase.
Now, you can use pendulum two for this too. Pendulum 2 is 0.1 seconds behind, so, for each oscillation it'll be 0.1 seconds behind the first one. After 19 revolutions, we can see that pendulum 2 is in fact 1.9 seconds behind, which is an entire revolution of pendulum 1, which means it's in phase.
I'm so frickin' sorry for my terrible formatting, etc, but I hope I helped. Tell me if you want an analogy to help you with this, because I know I make no sense.