>>>>> On Tue, 29 Nov 2005 09:17:11 +0000, Dave Shield <[EMAIL PROTECTED]> said:
Dave> Though I'd probably prefer to see it as the culmination of Dave> a series of individual sub-tree tests. Something like: Dave> - does the system group work properly? Dave> - does the ip group work properly? Dave> - <etc, etc> Dave> - does the whatever-the-last-group-is work properly? Those would certainly be nice and should be done *as well*. However this: Dave> - does a full walk work properly? should be done too, since it will also verify for people that their particular installed 3rd party modules are behaving properly as well besides just the known ones. Also, as I tried to state previously (and poorly), I've seen some tables fail only when you start a getnext *before* the table itself. IE, snmpgetnext on ifTable mhight work, but doing an snmpgetnext on ifNumber.0 would fail (the lexicographically previous node). To make this worse, however, I actually wonder if it should be off by default. A huge walk requires a huge timeout and a huge delay in the make test suite. I think we need to start thinking about test levels, where this would be > than the default level to run. EG, default = 5 and this would be something like an 8. -- Wes Hardaker Sparta, Inc. ------------------------------------------------------- This SF.net email is sponsored by: Splunk Inc. Do you grep through log files for problems? Stop! Download the new AJAX search engine that makes searching your log files as easy as surfing the web. DOWNLOAD SPLUNK! http://ads.osdn.com/?ad_id=7637&alloc_id=16865&op=click _______________________________________________ Net-snmp-coders mailing list Net-snmp-coders@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/net-snmp-coders
