Why linkburst makes long time server down？
cb23kg
✭✭
Dear Ingress staffs,
Even if it's not a sophisticated global network service, it usually measures the response time, and if it takes several times or dozens of times longer than usual, it determines the requests that are overloaded and cuts them off, then distributes the load by adding servers or bypassing regions to calm it down, and announcing it on social media etc. within half or one hour.
But there's no way it would be possible to leave it up to a country's employees' working hours for several hours while it's in trouble around the world. And I didn't see any improvement at all after last month's Europe for the third time.
We were aware of the starburst link limits through our predecessor's operations, but we dared to go for it in the hope that it might be an improvement. But nothings had changed...
Today, 24-hour server monitoring and automated or manual recovery can be outsourced, but you're not a weak startup that can't do that.
Is it because you can't do that despite having a globally popular game that is derived from the same service and shares some resources, or is it because ingress is not being taken seriously by the operations engineers or is it company-wide?
Or will the new system that has been replaced not be able to scale up?
I'm worried about whether you can continue to do this when large scale events and innovative services are released in the future.
Comments
I totally agree with you. I have serious doubt whether or not the game infrastructure can support large events such as XM anomaly.