On Aug 7, 6:01 pm, Andrew O'Brien <andr...@oriel.com.au> wrote: > Profiling for monitoring and activity estimation purposes - isn't that the > point of the sFlow set of patches mentioned a few times on list?
My opinion on the sFlow patches has been that it shouldn't be a change to memcached. What can we do to memcached to make it easy for someone to deploy this at runtime on an existing system? (well, assuming the system doesn't have dtrace, anyway) The last conversation I had involved writing a shim engine to inject the sFlow logic. I don't know that I've heard any reason that wouldn't work. If there's something better we could do, let's do that thing.