CM12 - Day 33 in the lab

We brought up a couple dozen clients on the site last night and today the CAS was just a sluggish pig. SMSExec wasn't doing anything (things like dataldr.log are dead since it gets that data from SQL from the primary sites now). But the CPUs were burning on SQL and even though this is a VM, the disk queue length was still under .05 (the underlying disks are like a dozen RAID 10 drives). So I shut down the VM and in Hyper-V manager, I cranked it from 2 GB min to 4 GB min and from 2 CPUs to 4. She's purring now.

Holy smokes, my primary was melting on CPU too. So I just gave it the same medicine. I was thinking the super lag between sites was the old-style transactional replication, but it was probably just the CPU limit.