Henrik Stoerner wrote:

This doesn't mean that I won't consider adding new stuff before the 4.3.0 release, but right now the plan is to get 4.3.0 shipped with the current set of features. But if I've missed someone's favourite patch or feature request, do let me know.

I am testing the latest snapshot now and I really like what I've seen, especially the hobbit-topchanges.sh script. This is

Major new features

• Flap detection of statuses that change color rapidly. The status is kept at the most critical level until it stops flapping.

Is the flap detetection going to have some kind of configurable parameters? And will we get some kind of indication that flapping is occurring?

• Split NCV support - graph data from NCV can be split into multiple RRD databases allowing for varying number of datasets.

I'm very happy to see this one! :)

• RRD database parameters are now configurable (i.e. number of datapoints stored, whether to store min/max values etc). Note that this only applies to newly created RRD files, not existing ones.

How do I toggle the min/max for new RRDs?

Display things

• The trends page default data-period can be configured to something other than the default 48-hour view, and the user can select a different period on-the-fly.

Another really cool and useful improvement.

Two features I've been asked for but I know either can't be done for 4.3.0 or maybe even at all are:

1. I get asked for this all the time: A way to mark "big" events on the graphs, so that we can have some marker inside or text outside the graph that gives some context to sweeping, overall trend changes. For example, if the average CPU IOwait on a database server drops from 35% to 10% after a code release, a person who is looking at that graph and wasn't involved in that code release would immediately know why the performance improved because a brief explanation of what took place during that time frame would be displayed along side.

2. An easier way to customize the colors or style of the Hobbit webpages. I tried to take this on one afternoon and found myself going through lots of source code, since it appeared that it was hard coded in quite a few places.

Tom

Tom Georgoulias Sr. Systems Engineer McClatchy Interactive tomg at mcclatchyinteractive.com