Lambda Labs Logs: Week 1

February 21, 2019

Part 1 - Individual Accomplishments this Week

Team Contribution Graph: https://github.com/Lambda-School-Labs/labs9-developer-profiles/graphs/contributors

Github Handle: mkerbleski

Front-End URL: https://ecstatic-dev-profiles.netlify.com/

Back_End URL: https://developer-profiles.herokuapp.com/

This week was a lot of set up for the web-page. While we got a lot done with setting up endpoints, and establishing the code base we were challenged by deploying the server to AWS (which we reverted to using Heroku) and the bigger challenge of deploying the back-end within a suborder. We are very close to overcoming both of those challenges. Personally most of the work that I did was in the back-end, establishing endpoints, deployment, and working on seeds and migrations. I also created a react component that renders the user profile card, with dummy data. Tools used include: AWS, Heroku, express, node.js, react, javascript, dotenv, & knex.

Another big take away from this week was how quickly the code base can change and narrowing down bugs when there are 5 programmers making changes at once. I think that it will be important to establish tests and test on every deployment as well as before a PR in order to keep everything current.

Tasks Pulled

Front End

User Card
Github: https://github.com/Lambda-School-Labs/labs9-developer-profiles/pull/44
Trello: https://trello.com/c/xxdFabGT/55-skelton-for-developer-detail

BackEnd

Set up server
Github : https://github.com/Lambda-School-Labs/labs9-developer-profiles/pull/5
Trello : https://trello.com/c/NfXJnQSw/38-set-up-server

Set up routes
Github : https://github.com/Lambda-School-Labs/labs9-developer-profiles/pull/6/files
Trello : https://trello.com/c/Opf02DqY/39-add-skill-post

Deployment
Github: https://github.com/Lambda-School-Labs/labs9-developer-profiles/pull/11
Trello: https://trello.com/c/J9bJIOKl/47-deploy-back-end

Detailed Analysis

Back End - Ticket 1

For the back-end server we wanted to try using AWS to host are server so that we would have other options for deploying besides solely Heroku. I made an account and after a lot of research on what was going on managed to make a MySQL database. And found that there was many more options and controls with amazon then there were with Heroku, as seen below.

I believe that I was connected to AWS at some point as the DB connection graph spiked to 1 as I made a migration locally. However testing endpoints at this point in time to retrieve any data was impossible as they were all broken.

With the deadline looming we decided to go with what we had done previously and to use the Heroku PostgreSQL server. In hindsight this was probably unnecessary because we were unable to get that server to work and concluded that how we had deployed our back end was not installing the dependencies. As well as our endpoints still not working.

After problem solving everything we could think of with the team. We decided to start from scratch. Team member Liz took the reins and figured out that the dependencies needed to go in the root package.json folder instead of the sub folder. While the endpoints still do not work, we have reestablished a live, remote db connection and the server is up and running!

While this task took many unexpected turns and is still left with some bugs to clean up, I have learned a ton about dev-ops, db and Amazon web services. As well as the value of teamwork, and the importance of taking a step back when blockers occur for a while.

Part 2 - Milestone Reflections

My experience forming a team was pretty effortless. From signups I recognized team-members that were hard working and people that I wanted to work with. I think I helped the team solidify as a group by taking the initiative at the beginning and volunteering to take on some initial tasks to get the ball rolling. One thing that I think may have caused friction, which has since been overcome, was probably struggling with back-end deployment by myself for a while, when I opened it up to the group the task at hand came together much smoother and everyone was up to date. Next week I am going to propose that we break down tasks into the smallest possible problems, and a more robust testing strategy so that PRs are merged with less conflicts and debugging in other parts of the code base are not effected.