Back with another summary of a week working at Telescope. With my duties as sheriff in the year view mirror now, it is now time to start writing some code that will land in 2.6 and Beyond. Or, at least that’s what I was hoping to do this week.
304 Debugging
I figured I would start off the week by trying to knock off an issue that is preventing my blog from getting picked up in the Telescope feed. This issue was first created back in Dec 2021 and there was already some discussion on the matter. The professor had given some advise on how to move forward
After confirming that nobody has tried that yet there was another issue that came up in that I am not that good at debugging on Visual Studio Code. I asked if there was any videos on the matter from the previous course, and found out that there is a .md document that explains how to set up a debugging session with Telescope. I was able to follow the instructions and get the debugging session to start. I was also able to add breakpoints and step in, out, and over functions. So there was defiantly some improvement for me in that department.
Unfortunately I still have not been able to identify the problem. The issue that I am currently stuck on is trying figuring out what endpoint I should be using to test things. I thought I had it figured out by going to this end point http://localhost/v1/posts/feeds/invalid and then picking one that had a 304 error message. I then changed the endpoint to be `http://localhost/v1/posts/feeds/${post.id}` which worked and brought me to the appropriate endpoint, but then my breakpoint would not trigger when I had it on the function. All this has taken me several hours, and it’s not really an issue that has anything to do with what I want to focus on for 3.0. Which is why I spent the rest of my week working on the Docker related issues.
Containing Images
As stated in last weeks blog, I want to work on Docker. I have self-assigned myself to multiple tasks that relate to it, and that I believe I should be able to implement by the time 3.0 comes around. During our meeting on Tuesday, the professor and I set up a meeting to discuss what needs to happen for a couple of the issues that will be my starting point.
Specifically issue#1743 is about using GitHub Actions to automate the process of building all of the repos images and then pushing them to a container registry. Well, I am not going to test this on every image we have to start. I will first start with an image that will have little impact on the project if something goes wrong with the feed-discovery service.
There was a lot of reading on my part this week. The professor had linked 3 articles that he thought would be good reading for before the meeting we had set up.
I and read all of those plush many more articles and documentation about the process. I had started to build the workflow by creating a new yaml file and copying the template found in the first link that the professor had shared. I started at the top and started making adjustments for it to work with the application. There was a couple of things that I was unsure of, and in my search for answers I came across this new workflow called push-to-ghcr. I plan on trying to implement push-to-ghcr over the weekend.
If I can build and push that image, I will then try to make a slim version of the image by using Docker-Slim. Having the image in a registry as well as being a fraction of the size could drastically cut down the load times for the application.
Having this work on the single services is just to prove that it can be done. Once the proof of concept has been made then we can go ahead and do the process for every single service that we use a docker image for. This gives me something to try and get landed for 2.6 and if it works it will give something I’m excited to implement for a couple of releases at minimum.
Tim's Blog for DPS 909 