This weeks question fits me well, as I am still unsure about the full details of how the fuck this all came together and was about to just rant about it anyway.

Ever since this companies network equipment and cabling has been updated, a lot of vital tools went down and bug out every now and then, at seemingly random times.

The codebase is a horrible mess to begin with and random things execute at random times and at random places spread all over different resources that get random hooks from random physical values etc.

Turns out (or at least what it so far seems like) all of them somehow sync their clock and other variables based on how many (valid-?) requests it gets per measured time and similar oddities, so when the network equipment got updated, that meant that multiple processes now could reach each other much faster and therefore threw off thousands of values and internal clocks.

There's a total of like 600 systems that are all "separate" from each other but all need to communicate in-sync for the production chain to properly work. Thankfully I didn't sign anything yet, so might actually just redirect them to somebody else, I am not ready to age 20 years, even for the amount that would pay.