On Fri, 2005-10-07 at 20:02 -0400, Robert Story wrote: > DS> "agentSecName" is for backward compatibility with previous releases. > DS> "iquerySecName" felt to be a more appropriate name. > > Then I strongly suggest that some error checking be introduced to check for > both being used, and issue a deprecated warning for the old name.
My immediate reaction to that suggestion was generally positive, but now I'm not so sure. Partly on grounds of practicality, since this is handled by a standard config handling routine (which doesn't care about resetting values). And partly by comparison with the current code, which will accept two different 'agentSecName' settings without complaint. There's nothing inherently different between that situation, and a pair of 'agentSecName'/'iquerySecName' settings. I'm inclined to rely on proper documentation instead. > DS> I also spotted a related change to the CVS code recently, whereby you > DS> seemed to provide a default value for this setting. Care to explain why? > > Sure. Because the code you checked in called get_ds_* for the pointer, but > didn't check if it was NULL before using it. I didn't have time to figure out > the correct fix, so I just made sure the ptr wouldn't be null. ;-) Good catch. I've now replaced this with a more appropriate fix - only creating an internal query session if there was a username specified. Dave ------------------------------------------------------- This SF.Net email is sponsored by: Power Architecture Resource Center: Free content, downloads, discussions, and more. http://solutions.newsforge.com/ibmarch.tmpl _______________________________________________ Net-snmp-coders mailing list Net-snmp-coders@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/net-snmp-coders
