You have twenty nodes in your ASG and your health checks say they’re all fine – so why are 1/20th of your users seeing white screens? We’ve just added per-host data breakdown and filtering to TraceView to make it easier than ever to answer that type of question (and more)!
We’ve added a Hosts data table, which shows your most-trafficked, slowest, or most under load hosts, along with the entry point(s) for requests on each node. For instance, below we have a screenshot of a few of our app servers–they seem to be in the same ballpark for request handling, though because we have some sticky sessions, one of the hosts is receiving a few more requests than the others.
Clicking a hostname name is just like filtering on any other attribute in TraceView–now you can see detailed performance data for requests handled by that host, as well as the set of request traces gathered from it.
This breakdown even applies to data filtered by layers, and even individual calls like queries, cache operations, and remote procedure calls.
Check it out today, and as always, let us know what you think!