Yep, mistyped … corrected it now.

Most being the keyword, not all. And honestly, when I’m developing, I don’t rely on what pm2 is telling me, but rather specifically what the browser sees (or doesn’t) and where the errors lie. pm2 doesn’t do that. I can get all errors and debug messages in one place: the browser’s console log. I don’t need to open yet another window just to look at pm2’s logs.

You need the browser’s console log. I forgot how to pull it up in Electron, but a search on the forums should give you an answer. Alternatively, you can open your favorite browser on a different machine (on the same network) and visit the rpi’s ip and MM’s port, like so: http://rpi_ip:8080 (fill in rpi_ip). Then hit F12 on the browser to open the console panel.

Lots and lots of them on Amazon. It really depends on your application. I have both 2, 4, and 5 ports around the house for various things. Hit Amazon, and search for ‘Anker wall charger’. Then look at their outputs and do the math. For example, if you see a 4-port, 40W, the calculation is:

40W / 4 ports = 10W per port
Each port supplies 5V, so 10W / 5V = 2Amps per port.

Read the descriptions on them. I generally only buy those that can supply upwards of 2.4Amps on at least 1 port (and 2Amps on the others). But they have models that can supply that much on all ports as well.

Break them up:

				feeds: [
					{
						title: "Spiegel",
						url: "http://www.spiegel.de/schlagzeilen/tops/index.rss",
					},
					{
						title: "Sueddeutsche",
						url: "http://rss.sueddeutsche.de/rss/Topthemen",
					},
					{
						title: "Tagesschau",
						url: "http://www.tagesschau.de/xml/rss2"
					}
				],

As @jc21 suggested, add the line debug: true to the module’s config section, and restart MM, then look at the console log.

That’s correct, and that’s because of their power outputs. You have to find out that powered and that has a nice high output. I actually run one of my rpis off of an Anker wall charger that outputs up to 2.4Amps per port.

Interesting proposition: adding the moon phases to the clock. Adding clock faces is very easy, however integrating one module into the other (a clock into the weather module, or a partial weather query inside the clock module), now you’re working to make a third beast of a module. A possible solution might be to have the clock module listen to broadcasts from the weather module and adjust its face accordingly.

A possible working method:
configure the clock to have an analog face (with a default face)
configure the clock to listen to weather broadcasts
configure the weather module to broadcast moon phases
if/when the clock receives such notification (hopefully only once a day), it will change its face to match and timing will be of importance here when the mirror first loads up - what if the weather module broadcasts before the clock is up and running?

Then there’s the question of, how detailed do you want it to be? Do you want a 28-day moon phase displayed? Or a simplified 8-stages version?

Something else to keep in mind, having an actual image will increase the payload of the module itself. Right now all of the analog faces are SVG files, which are rather small, and they can be scaled without losing resolution. You can’t easily do that with an actual picture. Either you are including very large images in case someone (like me) decides to have a screen with ONLY the clock on it, taking up the full screen, and still be able to scale it small for the “normal” folks. Personally my vote would be to use plain black and white SVG images. They’ll be small and easily scalable.

Anyway, that’s my 2 pesos.

And even if you find out that you need additional devices later, you are better off with a good quality self-powered USB hub. They’re cheap, they’re small, you could even remove the whole circuitry from the case for a much smaller form factor.

Not for MM, I can’t think of any.