The tooltip, when you hover over the downvote arrow, says "This question does not show any research effort; it is unclear or not useful". That's the closest you'll get to an authoritative answer as to why anyone downvoted a question or answer, absent a comment from a downvoter.
My thoughts on why I might downvote them, if I were in that sort of mood:
"do we need to patch": A mixture of too broad and no clear problem statement. You're mushing up all sorts of things in that question, and it's hard to tell exactly what you want help with.
"graphics are nan": asking "what could explain this?" is just begging for someone to answer "sunspots". The close reason this question has attracted sums up the problem very nicely: "Questions seeking installation, configuration or diagnostic help must include the desired end state, the specific problem or error, sufficient information about the configuration and environment to reproduce it, and attempted solutions. Questions without a clear problem statement are not useful to other readers and are unlikely to get good answers."
"Can gmond be configured to not need a restart?" This one seems fine to me, and it's the one that's gotten an upvote, so all seems well there. The downvote could very well be someone who saw your other questions and went on a derp spree.
"showing both hostnames and IPs" This one is another magnet for close votes for "lack of sufficient information to reproduce". Computers don't decide to do things differently all on their own. Something changed in your environment. What was it?
"adding more colours" This one seems reasonable.
Also, in general, there's no real need to link to the wikipedia page for ganglia in your questions... anyone here who needs to be told what ganglia is and can't work out how to search for it is probably not your desired target audience.
