I’m doing some investigation on how a BigFix client evaluates Analysis properties, both for educational purposes, and for troubleshooting one property that seems to be taking too long to evaluate (getting “Inspector Interrupted”).
Hunting around I’ve found some really interesting information on this forum and on IBM Support (search here for “client evaluation cycle”). One thing I haven’t found, however, is just a good general description of the client evaluation cycle, the things that are done in that cycle, and logs that are helpful. There are strong hints by looking at the “evaluationcycle of client” inspector, but I’d like something that might tie it all together.
Can someone either provide me pointers to already published work, or write up a handy overview that I and others can reference here in the forum?
In general the client performs two types of “evaluation cycles,” the first being the normal background fixlet evaluation, and the second being the Action evaluation (when an action becomes relevant).
There are two essential “lists” of fixlets that the client would go through, the action relevant list usually being a lot shorter, so the evaluation cycle does change depending on the mode it is in and can look quicker if its going through the action cycle.
Things that are generally done in the cycle are checking for relevance, checking for gathering, checking the processes running, reporting, checking for power states etc.