Whenever the engineering support steps in, the problem gets solved fast or the issue gets acknowledged. However, over the last 15 months, ~50% of times our GCP

Whenever the engineering support steps in, the problem gets solved fast or the issue gets acknowledged. However, over the last 15 months, ~50% of times our GCP support cases end up in frustration with no resolution. We have more success finding issues in the corresponding GitHub repo and opening an issue there. We have the expensive off-the-shelve support option (I think 450/seat/month) for 1h response. In most cases, we spend more time back and forth with support than it would take you to figure it out. I'm talking about issues that span over weeks with tens of hours spent. We end up reiterating the original support case problem (i.e. the support engineer doesn't bother reading the actual problem) whenever the engineer changes. We've had: P1s where the support engineer told us we'll get an update the following day, only to figure out that there was a breaking release on their side that exactly matched our description. While investigating a load balancer issue, the support engineer looked at the LB logs, saw a ton of logs coming from penetration scans (e.g. GET /phpmyadmin), and suggested that the solution was to open up those addresses.