On Tue, Mar 9, 2010 at 9:35 AM, Robert Story <rst...@freesnmp.com> wrote: > There was some discussion in the IRC channel this morning about the default > behavior of snmpwalk biting newbies. They will often run snmpwalk without > specifying a starting point, and by default snmpwalk will start the walk on > the mib-2 branch. This leaves users puzzled as to why they aren't getting > enterprise MIB objects. > > There were a few proposals for a possible fix: > > a) start at .1 > pros: will show all available objects > cons: will show all available objects
I like this option. Yes, it's changing a default, and it's possible that someone's relying on this default -- but is it really that likely? What's the scenario that some automated script somewhere wants all of mib-2 and nothing else? It seems more likely to me that automated uses of snmpwalk specify a much smaller subtree. I can't count the number of times I've had to tell users to add ".1" to their snmpwalk command line - mostly "I've added the MIBs for my device but snmpwalk still doesn't show the enterprise OIDs!". > b) start at mib-2, but print a message to stderr (much like we do when we > default to stderr for logging) > pros: user advised that only a portion of mib is being used > cons: slight change in behavior (can be mitigated with config token to > disable message) I could live with this option. It'll probably still result in confusion, since people may miss the warning (say, "snmpwalk foo | less", and the error message gets overwritten), but may reduce the confusion somewhat. > c) print error and exit if no start point specified (suggest mib-2 or .1) > pros: makes it very clear what's going on > cons: big change in behavior This one just feels annoying :-) Bill ------------------------------------------------------------------------------ Download Intel® Parallel Studio Eval Try the new software tools for yourself. Speed compiling, find bugs proactively, and fine-tune applications for parallel performance. See why Intel Parallel Studio got high marks during beta. http://p.sf.net/sfu/intel-sw-dev _______________________________________________ Net-snmp-coders mailing list Net-snmp-coders@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/net-snmp-coders
