4 ms·
Yes, the actual cause, a problem on the wire between them and Amazon. UDP flood eating up available bandwidth. I didn't guess it either :-)
by idlewords 17y ago
Yes, the actual cause, a problem on the wire between them and Amazon. UDP flood eating up available bandwidth.
I didn't guess it either :-)
- keefe 17y agoSo, do you think setting a smaller max size on your EBS instances would have avoided this by spreading the traffic, so if you were using 1TB using 10 100GB ones instead and federating queries across them?