Before I set up load balancing, when my single instance reached 95% load, it slowly crawled-became unusable. Is there a way to make sure I don’t have any users encounter In this case, because they loop to overloaded instances in some way?
http:// technet.microsoft.com/en-us/library/cc772183(v=ws.10).aspx
I am using Azure to host cloud services, which I recently modified This service can be scaled across multiple instances, including the session cache worker role. My question is why I see extreme load (over 90%) on one instance, but not on other instances (at all 15-20% in other examples)? Should I be worried?
Before I set up load balancing, when my single instance reached 95% load, it slowly crawled-became unusable. Is there a way to make sure I don’t have any users encounter In this case, because they somehow loop to the overloaded instance?
We found that when a load-balanced instance failed over, we encountered a similar situation; what we saw was all load transfers, but not Will rebalance. We found that turning off keep alive for a few minutes allows the load to spread again, and then we can turn it back on.
http://technet.microsoft.com/en-us/library/ cc772183(v=ws.10).aspx