the frontest of front ends

I woke up in the middle of the night with the most deranged way to display logs in an app. The scenario that spawned this was struggling to deal with being both developer and operator of apps with almost no goddamn means of being able to operate in a sustainable manner: everything kept breaking. We have no UAT; we have no access to our clusters, we had logs shipped somewhere, which they then stopped because it broke something else; the end users demanded to see logs, and really, what end user can you explain with ever-fraying patience, that it could be fetched out of the k8s pod logs and yeah, learning k8s is going to be a doozy for you guys, you do everything in Excel.

A junior of mine basically confessed to setting up the cheapest of cheap hacks on her local dev machine (shitty flask app calling shitty k8s commands to display for logs, which I’m not sure was successful given that I don’t think it ever occurred to the junior that cron did not have access to the k8s config) and I can’t help but be unreasonably angry that this is what it’s come to in this team to get stuff going. I’m amazed at the ingenuity - the ingenuity is unparalleled when it comes to the depths of despair - but the cost is pretty significant. The other junior didn’t even bother to check if UAT was working and went straight to prod, which was discovered only when they couldn’t get something to work. So now I’m lecturing on the Importance of UAT even when Our UAT is such a Shitshow I Can’t Blame You And This Really Sucks.

I’m pretty frustrated, if you can’t tell. Seems to have been a recurring thing for a while.