--- Robert Story <[EMAIL PROTECTED]> wrote: > On Tue, 24 May 2005 08:42:35 -0700 (PDT) Salsa > wrote: > S> Rightr now is another machine but I also used > S> localhost. Snmptrapd was indeed running and > received > S> traps until my subagent stopped. Then it said > that it > S> lost connection and tried to ping and reconnect. > > So the sub-agent detects a failed ping... it sounds > like the master get > busy for some reason. It the master being polled by > some process? Is there a > load spike right before the disconnect?
Yes, I believe it's something like that. But to me looks more like some kind of deadlock. No, the master is not being polled, at least it shouldn't be because I'm not monitoring my computer nor anything. I didn't notice any load spikes either. > > Try running the agent with debug and time-stamps to > try and establish a > time-line for reproducing the problem. In your > snmpd.conf, add: > > [snmp] doDebugging 1 > [snmp] debugTokens agent > [snmp] logTimestamp true > > Add the same thing in the conf file for your > sub-agent. Then try to establish > the sequence of events. I didn't get this one. I added those lines to snmpd.conf but I couldn't notice anything different. The line s are supossed to be added exactly like that? That's what I did (with the [snmp] and all). Where should I see the debug? Do I have to run the agent with any special option? > > S> I have to send 100 traps. > > Are you sending them all at once? Can you throttle > them back and send fewer at > a time? I'm working on this work-around. But I wish I could send them all. I mean, every time my callback function is called I'd like to send every pending alarm. If I send some, return and wait till my function is called again, then no problem occurs. > > Another thing you can try is increasing the timeout > for the sub-agent. > Unfortunately, there isn't way to do this without > re-compiling the libraries. > In agent/mibgroup/agentx/subagent.c, around line > 710, in > subagent_open_master_settion, set the sess.timeout > to something like "5 * > ONE_SEC" for a 5 second timeout, instead of the > default one second. > I'll try later. But What I did for now was to sleep (1) for every iteration. So I had at least one second between each trap sent. > -- > NOTE: messages sent directly to me, instead of the > lists, will be deleted > unless they are requests for paid consulting > services. > > Robert Story; NET-SNMP Junkie > Support: <http://www.net-snmp.org/> > <irc://irc.freenode.net/#net-snmp> > Archive: > <http://sourceforge.net/mailarchive/forum.php?forum=net-snmp-users> > > You are lost in a twisty maze of little standards, > all different. > -- Salsa __________________________________ Do you Yahoo!? Make Yahoo! your home page http://www.yahoo.com/r/hs ------------------------------------------------------- This SF.Net email is sponsored by Yahoo. Introducing Yahoo! Search Developer Network - Create apps using Yahoo! Search APIs Find out how you can build Yahoo! directly into your own Applications - visit http://developer.yahoo.net/?fr=offad-ysdn-ostg-q22005 _______________________________________________ Net-snmp-users mailing list Net-snmp-users@lists.sourceforge.net Please see the following page to unsubscribe or change other options: https://lists.sourceforge.net/lists/listinfo/net-snmp-users
