martinwoodward/dasdeployer

Raspberry Pi powered manual release approval gate for Azure Pipelines. Mit viele blinkenlighten

Martin Woodward (he/him)

Title: VP of Developer Relations, GitHub

Bio Short: Martin is the Vice President of Developer Relations at GitHub where he helps developers and open source communities to create delightful things. Prior to that he was part of the team at Microsoft building the tooling for DevOps teams inside and outside the company where he helped change how the company built software, by helping them change how they work with open source communities and how they talk with developers.

Bio Long: Martin is the Vice President of Developer Relations at GitHub where he helps developers and open source communities to create delightful things. Prior to that he was part of the team at Microsoft building the tooling for DevOps teams inside and outside the company where he helped change how the company built software, by helping them change how they work with open source communities and how they talk with developers. He was the original creator of the Microsoft org on GitHub. While there he also helped set up the .NET Foundation bringing in other companies like Amazon, Google, Samsung and RedHat to help drive the future direction of the open source platform. Martin joined Microsoft after they acquired his small five person start-up that was based around the Eclipse Java-based platform and was part of the acquisition team for GitHub.

For photos, please see the headshots folder.

dasdeployer's Issues

Questions I'm having as I go through this build

Martin suggested on Twitter I open this up as an issue to track and for other people to see them as well.

I'm not sure exactly how the Pi is being powered. I'm not sure if you're running a +5v from the PSU to the 5v HAT pin on the Pi, or running the +5v into your perfboard, and letting 3.3V to 5V shifter board convert that to 3.3V to power the Pi, or both? Looking at the pictures from the Ignite presentation (Slide 21) it looks like there are a couple of wires going into the PSU and (other than the mains input) they all look to be on the 5V terminals, but I can't tell from the picture where they all go or even if they're all being used. It also looks like no matter what there's +5v going into the perfboard but no ground coming out, so it's being grounded via the Pi, so the Pi has to be connected to the ground on the PSU from somewhere? I've been able to power the Pi by plugging the +5v and ground from my PSU into the 5V and Ground pins on the Pi respectively. My understanding on the 5V rail on the Pi is that all the 5V is connected, so once you're running power into the 5V of the Pi, the "external" 5V going into the perf board can either come directly from the PSU or from the other 5V pin on the Pi, electrically there appears to be no difference.

Another question with the Toggle switches, I thought this would be relatively simple and I'd be able to figure it out myself and I mostly did, but apparently these switches are pretty common and also not well documented, so lots of people smarter than me have had some problems with them. I went with a different model than what was recommended but it looks like the leads are labelled the same way. I attached the top lead with the ground symbol to a ground terminal, the positive lead to IO12 (the "dev" switch led in the code) and the remaining lead (looks like a headlight? I can't find any reference for what that symbol means) to IO16 (the "dev" toggle in the code). Once I had it wired up, I was able to go to a Python terminal and initialize IO12 as a light and turn it on and off (once I'd put the switch in the On position) and initialize IO16 as a button and read when it was pressed (flipped on) and released (flipped off). Though somehow when it was flipped off there LED was faintly on, though not as bright as when the switch was flipped on and the LED properly turned on. As if a little bit of current was flowing through from the switch detection. So I'm wondering if your switches are faintly lit when they're off, or if I reversed IO12 and IO16, or if the code is dealing with this condition somehow? Played with this again, I was right that the positive terminal goes to 12 and the lamp terminal goes to 16, and I see in the code that you initialize the switch with the pull_up off, which prevents the faint glow even when the LED is off.

Hardware diagram

Any update on the hardware diagram/images?