On Wed, Jan 18, 2006 at 02:38:11PM -0500, Robert Story wrote: > On Wed, 18 Jan 2006 20:21:29 +0100 Magnus wrote: > MF> It seems that delegated handlers that return ENDOFMIBVIEW as response to > MF> a GETNEXT request won't trigger any further advancement in the registry so > MF> they might end up returning a false ENDOFMIBVIEW (or, in some cases, an > MF> NOSUCHOBJECT or NOSUCHINSTANCE) > MF> > MF> Is this all intentional? > MF> What would be the right thing to return from my outstanding item? NULL? > > They shouldn't return ENDOFMIBVIEW, they should set the request error to > ENDOFMIBVIEW and return success/noerror.
Ok, but if I return SUCCESS and sets the request error to ENDOFMIBVIEW I still get the problem. On a GET, my code is made to set the request error to NOSUCHOBJECT. On a GETNEXT, my code is made to set the request error to ENDOFMIBVIEW. If I send a GETNEXT to the program it returns as follows: $ MIBS=ALL snmpgetnext -v2c -c public localhost 1.3.6.1.4.1.8072.2.1.2 NET-SNMP-EXAMPLES-MIB::netSnmpExampleSleeper.0 = No Such Object available on this agent at this OID One interesting point is that the only requests my program sees are GET-requests. Using a similar program that isn't using a delayed instance it works and the responsets I get consist of a GET followed by a GETNEXT. Both programs are agentx subagents and I am using net-snmp CVS HEAD. Delayed version: http://www.lysator.liu.se/~magfr/a2.c Direct version: http://www.lysator.liu.se/~magfr/a3.c /MF ------------------------------------------------------- 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://sel.as-us.falkag.net/sel?cmd=lnk&kid=103432&bid=230486&dat=121642 _______________________________________________ Net-snmp-coders mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/net-snmp-coders
