On Mon, 15 Jun 2009 22:21:27 +0100 Dave wrote:
DS> 2009/6/15 Robert Story <[email protected]>:
DS> > if internal to the cache helper,  that would mean OIDs in the conf file, 
since
DS> > that's all the cache helper is passed..
DS> 
DS> No - it means that MIB object names in the config file would be converted
DS> to numeric OIDs, presumably by the config parsing routine.
DS> 
DS> Internally the comparisons would be of numeric OIDs, yes.
DS> But the user visible config could use names or numbers.

True.. I was just think that we always tell people you don't need to load the
mibs to run the agent... 

DS> (But I do think this is a less than ideal approach - specifying the timeout
DS> in one place would be clearer, IMO)

Well the first one, the creating of the cache, is in the 'do not edit' code,
hiding the complexity from the user... but that's another argument. :-)

DS> > But persistence could be done later, don't you think?
DS> 
DS> Yes - that might be sensible.

Ok, so I think we're basically in agreement.. Unless anyone else wants to
chime in on whether or not handling the code in the cache helper is better
than directly in the tables? 

The only wrinkle I can think of might be some mib table doing it's own
caching, not using the helper, that also wants to take advantage of the new
configurable values.. Should they be left with hard-coded values, updated to
use cache handlers, or just use the same new selist/netsnmp_container api for
puling the value out directly?


------------------------------------------------------------------------------
Crystal Reports - New Free Runtime and 30 Day Trial
Check out the new simplified licensing option that enables unlimited
royalty-free distribution of the report engine for externally facing 
server and web deployment.
http://p.sf.net/sfu/businessobjects
_______________________________________________
Net-snmp-coders mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/net-snmp-coders

Reply via email to