Dave Shield wrote: >>Due to what you wrote running the tests the way they're done at present >>might be right or slightly wrong. They're right in that all OID's that >>are expected to exist will be treated as failed in test 47 and that in >>the tests 48 and 49 these OID's need not be handled as failed again. > > > Hmmmm.... > In general, we've tended to regard each test as completely independent > of the others. In principle, it'd be perfectly possible to run *just* > tests 48 and 49 (without running 47 first) So I'd be reluctant to > have the results of one test rely on a previous one.
Do you mean "not rely on a previous one"? That way I can use every test independent of any other (and in any order if necessary). > There's also the possibility of access control settings being such > that certain objects are only available for one SNMP version, and > not for others. > > > > >>But they're wrong in that even for tests 48 and 49 the snmp daemon >>doesn't return the expected value since those OID's don't exist. > > > My main problem with these most recent tests (T16xrfc1213), is that > I'm not clear what they are actually trying to test. As you're probably seen in CVS yourself those tests have been added from patch #689757 submitted by Ling Xiaofeng (adding has been done by Mike Slifcak about 10 months ago). They seem to test whether most of the variables in RFC-1213 MIB are accessible using protocols 1, 2c or 3 (I didn't check but I think that not all mib-II variables are checked currently). > The basic idea of the test suite is to tell whether the suite is > behaving as expected, with each test checking one particular style > of behaviour. > That doesn't appear to be true of these last three tests, so > it's unclear (to me at least), what constitutes a "successful" test. > > I'm wondering whether these might be better regarded as separate > from the main "testing/tests" block. My problem when building net-snmp binary tarballs is that "make test" stops with an error if any of those mib-II variables isn't available for any of my platforms (this is HP-UX 10.20/11.0/11i). I know that at least one of those variables is not available (ip.ipRoutingDiscards.0), and every time I'm running nsb-package the build process is aborted due to that non-error. I'd vote for a separate probably optional test block which tests the availability of most (or all) mib-II variables and maybe also other MIB variables (like the host MIB). This test block should in no case stop the build process, but it could become part of the delivery process as it is the case with the configuration summary that can be downloaded from the net-snmp files section. This way everyone can inform himself which mib variables are supported for a certain platform and which aren't available due to any reason (not implemented, not available for that platform etc.). > Dave Johannes -- Johannes Schmidt-Fischer InterFace AG phone +49 (0)89 / 610 49 - 207 Leipziger Str. 16 fax +49 (0)89 / 610 49 - 85 D-82008 Unterhaching mobile +49 (0)171/ 787 76 01 http://www.InterFace-AG.com mailto:[EMAIL PROTECTED] ------------------------------------------------------- SF.Net email is Sponsored by the Better Software Conference & EXPO September 19-22, 2005 * San Francisco, CA * Development Lifecycle Practices Agile & Plan-Driven Development * Managing Projects & Teams * Testing & QA Security * Process Improvement & Measurement * http://www.sqe.com/bsce5sf _______________________________________________ Net-snmp-coders mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/net-snmp-coders
