On Jan 9, 2008, at 10:39 AM, Jo Rhett wrote:

Verbose: output should make clear to the cfengine administrator what
their configuration syntax resolves to.

Debug: output should make clear to the cfengine hacker what their
code is really doing on the cfengine's administrator's system, when
he reports a bug.

In short, any query to cfengine-help should include Verbose output,
and any query to cfengine-bug should include Debug output.

Does this make any sense?

Sounds like a great distinction to me, but I think the first step would be standardizing what all the levels mean (Mark?). I mention this because its not as easy as a split between verbose and debug because there are different levels of debug (Debug, Debug1, Debug2, etc.). Would be nice to have each one defined as to how it pertains to the complexity or obscurity of the information. That way, rather than changing all the package debugging lines to Debug, we could put them at a `proper' level.

Speaking of logging in general, is there a usage guideline on CfLog(cfverbose) vs. Verbose()? The latter seems to just be macros for printf when verbosity is high enough, whereas the former is more powerful in that it checks to see if error messages exist, as well as automatically printing the context.

Eric

Attachment: PGP.sig
Description: This is a digitally signed message part

_______________________________________________
Bug-cfengine mailing list
[email protected]
https://cfengine.org/mailman/listinfo/bug-cfengine

Reply via email to