Details
-
AboutFrustrated Sysadmin
-
Skillslinux, python, ruby, puppet, ansible
-
LocationSt. Helens, UK
-
Website
-
Github
Joined devRant on 10/21/2019
Join devRant
Do all the things like
++ or -- rants, post your own rants, comment on others' rants and build your customized dev avatar
Sign Up
Pipeless API
![](/static/devrant/img/pipeless-devrant-banner-white.png)
From the creators of devRant, Pipeless lets you power real-time personalized recommendations and activity feeds using a simple API
Learn More
-
I have two contributions so far in my first hacktoberfest. I have created a readme and updated a .gitignore to ignore VSCode’s .vscode file, just in case anyone else who wanted to contribute that uses VSCode doesn’t have to worry about the .vscode file3
-
So recently my open source project took off and got trending on GitHub (680 starts and 225 forks). This was the first time a project of mine really gained some traction and invested more of my time and weekends to maintain this project - I wrote comprehensive docs, contributing guidelines and reviewed PRs and made sure I commented on every single one of them. Sure, it isn't easy to review 50 PRs a day after coming home from work but the excitement of seeing this project becoming trending fueled me.
First 2 weeks it was good. I would come home from work and have dinner and sit down to maintain the project. Whenever contributors would be stuck, I would help them and write comments on each PR.
But the problem started since last week. People just really want to see their contribution activity graph get populated and hence they would make stupid PRs and literally no one followed contributing guidelines - I mentioned in that that the code should adhere to Pep8 styling but no one gave a shit. Each day I would spend reviewing PR with crappy formatted code and no sign of Pep8, and even some will just file PR and add a fucking docstring to every function or add paragraph of comments. Also, the PR quality was bad with unsquashed commits amounting to 10 or 20 or even sometimes 50.
I wrote the contributing guidelines doc and in that I mentioned every source that contributors could find helpful like how to squash commits, how to file a PR and Pep8 and not to write useless comments. Seriously people, grow up!6 -
"You gave us bad code! We ran it and now production is DOWN! Join this bridgeline now and help us fix this!"
So, as the author of the code in question, I join the bridge... And what happens next, I will simply never forget.
First, a little backstory... Another team within our company needed some vendor client software installed and maintained across the enterprise. Multiple OSes (Linux, AIX, Solaris, HPUX, etc.), so packaging and consistent update methods were a a challenge. I wrote an entire set of utilities to install, update and generally maintain the software; intending all the time that this other team would eventually own the process and code. With this in mind, I wrote extensive documentation, and conducted a formal turnover / training season with the other team.
So, fast forward to when the other team now owns my code, has been trained on how to use it, including (perhaps most importantly) how to send out updates when the vendor released upgrades to the agent software.
Now, this other team had the responsibility of releasing their first update since I gave them the process. Very simple upgrade process, already fully automated. What could have gone so horribly wrong? Did something the vendor supplied break their client?
I asked for the log files from the upgrade process. They sent them, and they looked... wrong. Very, very wrong.
Did you run the code I gave you to do this update?
"Yes, your code is broken - fix it! Production is down! Rabble, rabble, rabble!"
So, I go into our code management tool and review the _actual_ script they ran. Sure enough, it is my code... But something is very wrong.
More than 2/3rds of my code... has been commented out. The code is "there"... but has been commented out so it is not being executed. WT-actual-F?!
I question this on the bridge line. Silence. I insist someone explain what is going on. Is this a joke? Is this some kind of work version of candid camera?
Finally someone breaks the silence and explains.
And this, my friends, is the part I will never forget.
"We wanted to look through your code before we ran the update. When we looked at it, there was some stuff we didn't understand, so we commented that stuff out."
You... you didn't... understand... my some of the code... so you... you didn't ask me about it... you didn't try to actually figure out what it did... you... commented it OUT?!
"Right, we figured it was better to only run the parts we understood... But now we ran it and everything is broken and you need to fix your code."
I cannot repeat the things I said next, even here on devRant. Let's just say that call did not go well.
So, lesson learned? If you don't know what some code does? Just comment that shit out. Then blame the original author when it doesn't work.
You just cannot make this kind of stuff up.105