One of the last projects I managed at That Place was a provisioning engine that would not only set users up in the system, but also provide capacity monitoring for the servers. The goal was we'd a) fail a provisioning event when the servers were too full (not *entirely* full, but full enough that performance or capacity maximums were likely to be reached) and b) well before that point, provide feedback to the operations team ranging from "Service X is at threshold capacity (usually 50-60%, depending on the service); additional capacity will be needed soon" to "Hey, dipshit! Customers want to give us money, but they can't because the servers are full! Add a fucking server NOW!"
One of the few projects and teams from my latter days there of which I'm proud. Naturally, our management tried to make sure we didn't deliver it in a form that could be useful to *anyone*, and sadly most of my pride is from the fact that we defied them and delivered two complete working versions, one for internal consumption and one for sale to external ISPs. This, of course, meant the end of my career there. But I'd do it again every time. :)
![[personal profile]](https://www.dreamwidth.org/img/silk/identity/user.png){kind=small}
no subject
One of the few projects and teams from my latter days there of which I'm proud. Naturally, our management tried to make sure we didn't deliver it in a form that could be useful to *anyone*, and sadly most of my pride is from the fact that we defied them and delivered two complete working versions, one for internal consumption and one for sale to external ISPs. This, of course, meant the end of my career there. But I'd do it again every time. :)