This shows a profound lack of understanding into why operations does things the way it does. Our primary task, above all else, is to make sure that the infrast

This shows a profound lack of understanding into why operations does things the way it does. Our primary task, above all else, is to make sure that the infrastructure and thus the application keep running. We often take the blame if a developer's app is deployed into prod, runs for 3 days, and then dies. When their app dies in prod, I'm the first person to get paged (and there's a really good chance your developers don't even carry a pager). > You use hand-overs, procedures & documentation requirements to slow us down. Which we do not to irritate you, but because we need to support your application in production. I need to know if/when/how your application needs to be shut down, because I've shut down things like Tomcat before and gotten yelled at because they've done something weird that prevents a normal shutdown from being safe. > You hide behind bureaucracy and use ISO standardization as an excuse. Developers often don't realize this, but there's a lot more communication that needs to happen on our side to make a change than yours. Every change I want to make needs to be verified with half of infrastructure to make sure it won't impact their processes. Security has enabled an IDS that then broke applications by screwing with TCP negotiation on database connections. We've had SMB authentication be broken by reverse proxies. Everything we do can impact someone else, and can impact all of the however many apps we manage. > You make the infrastructure overly complex, using SPOF as an argument. This is an almost scary comment. The infrastructure isn't overly complex, it's just as complex as it needs to be to support that ancient app from 1995 that none of the developers want to take ownership of that has no HA or clustering built in, but is still a required service for 75% of the business. That HA isn't there for that fancy new app with clustering built in that was designed to not require hardware redundancy. You get paid to write software so the business can work more efficiently, I get paid to make sure that your software stays up so the business can use it.