<html xmlns:v="urn:schemas-microsoft-com:vml" xmlns:o="urn:schemas-microsoft-com:office:office" xmlns:w="urn:schemas-microsoft-com:office:word" xmlns:m="http://schemas.microsoft.com/office/2004/12/omml" xmlns="http://www.w3.org/TR/REC-html40"><head><meta http-equiv=Content-Type content="text/html; charset=iso-8859-1"><meta name=Generator content="Microsoft Word 14 (filtered medium)"><style></style></head><body bgcolor=white lang=IT link=blue vlink=purple><div class=WordSection1>Hello<o:p></o:p><o:p> </o:p>Probably I solve the problem creating a script that manage the host/ip received from the clients...I have to test it more, and more J<o:p></o:p><o:p> </o:p>May anyone send me the snmptt.ini file ?<o:p></o:p><o:p> </o:p>I still don’t understand some behaviour….<o:p></o:p><o:p> </o:p>Marco<o:p></o:p><o:p> </o:p><div><div style='border:none;border-top:solid #B5C4DF 1.0pt;padding:3.0pt 0cm 0cm 0cm'>Da: H�lzel, Arnold [mailto:arnold.holzel@kpn.com] Inviato: mercoled� 12 settembre 2012 11:47 A: Nagios Users List Oggetto: Re: [Nagios-users] R: R: configure receiving snmp traps<o:p></o:p></div></div><o:p> </o:p>Hello Marco,<o:p></o:p><o:p> </o:p>You can check your snmptt.ini file (/etc/snmp/snmptt.ini) there you can configure SNMPTT to strip the domain names. See the “dns_enable”, “strip_domain” and the “stip_domain_list” options, maybe that is the problem?<o:p></o:p><o:p> </o:p><![if !supportLists]>- <![endif]>Our systems send their SNMP messages to our monitoring server<o:p></o:p><![if !supportLists]>- <![endif]>the monitoring server does a DNS lookup, strips the domain name and sends the data to a “submit_trap” script<o:p></o:p><![if !supportLists]>- <![endif]>the “submit_trap” script converts the hostname to uppercase and send the data to Nagios.<o:p></o:p><o:p> </o:p>Hope it helps<o:p></o:p><o:p> </o:p><div>Met vriendelijke groet/with kind regards, <o:p></o:p>Arnold H�lzel<o:p></o:p></div><o:p> </o:p><div><div style='border:none;border-top:solid #B5C4DF 1.0pt;padding:3.0pt 0cm 0cm 0cm'>From: Marco Borsani <a href="mailto:[mailto:m.borsani@it.net]">[mailto:m.borsani@it.net]</a> Sent: woensdag 12 september 2012 11:20 To: 'Nagios Users List' Subject: [Nagios-users] R: R: configure receiving snmp traps<o:p></o:p></div></div><o:p> </o:p>Hi<o:p></o:p>Yes, I supposed this could be the problem.<o:p></o:p><o:p> </o:p><![if !supportLists]>1) <![endif]>I am sending trap from a client as:<o:p></o:p>snmptrap -v 1 -c public <Nagios_IP> 1.2.3.4 <Client hostname as it is configured on Nagios> 3 0 ''<o:p></o:p><o:p> </o:p><![if !supportLists]>2) <![endif]>Snmptrapd on Nagios receive the trap:<o:p></o:p>2012-09-12 11:08:23 <REAL CLIENT hostname fully qualified> [CLIENT IP] (via UDP: [CLIENT IP]:52203->[NAGIOS IP]) TRAP, SNMP v1, community public iso.2.3.4 Link Up Trap (0) Uptime: 36 days, 19:43:55.96<o:p></o:p><o:p> </o:p><![if !supportLists]>3) <![endif]>A script like “submit_trap” pass it to Nagios<o:p></o:p><o:p> </o:p><![if !supportLists]>4) <![endif]>Nagios log file:<o:p></o:p>[1347440926] Warning: Passive check result was received for service 'TRAP_service' on host <REAL CLIENT HOSTNAME>, but the host could not be found!<o:p></o:p><o:p> </o:p>In my configuration is not a “simple” problem of uppercase/lowercase, but I need to pass exactly the hostnames I need … Before sending the trap or change them as soon as the trap arrived??<o:p></o:p><o:p> </o:p><o:p> </o:p>Any idea?<o:p></o:p>Marco<o:p></o:p><o:p> </o:p><o:p> </o:p><o:p> </o:p><div><div style='border:none;border-top:solid #B5C4DF 1.0pt;padding:3.0pt 0cm 0cm 0cm'>Da: H�lzel, Arnold <a href="mailto:[mailto:arnold.holzel@kpn.com]">[mailto:arnold.holzel@kpn.com]</a> Inviato: marted� 11 settembre 2012 17:35 A: Nagios Users List Oggetto: Re: [Nagios-users] R: configure receiving snmp traps<o:p></o:p></div></div><o:p> </o:p>Hello Marco,<o:p></o:p><o:p> </o:p>What is the case (uppercase or lowercase) of the system names in Nagios? I had the same problem and that was caused by the fact that all our systems in Nagios are in uppercase and snmptt sends them in lowercase to Nagios. Because Nagios is case sensitive there is a mismatch and you will not see anything in Nagios. I had to manually change something in the script to make everything uppercase (I did not wanted to reconfigure every host in Nagios). <o:p></o:p>I am not able to look up and send the exact things I changed right now but if needed I can look it up tomorrow.<o:p></o:p><o:p> </o:p><div>Met vriendelijke groet/with kind regards, <o:p></o:p>Arnold H�lzel<o:p></o:p></div><o:p> </o:p><div><div style='border:none;border-top:solid #B5C4DF 1.0pt;padding:3.0pt 0cm 0cm 0cm'>From: Marco Borsani <a href="mailto:[mailto:m.borsani@it.net]">[mailto:m.borsani@it.net]</a> Sent: dinsdag 11 september 2012 17:01 To: 'Nagios Users List' Subject: [Nagios-users] R: configure receiving snmp traps<o:p></o:p></div></div><o:p> </o:p>Hi Mike<o:p></o:p><o:p> </o:p>Actually , I have :<o:p></o:p><![if !supportLists]>- <![endif]> installed snmptt (in daemon mode)<o:p></o:p><![if !supportLists]>- <![endif]>Configured a generic passive service in nagios<o:p></o:p><![if !supportLists]>- <![endif]>Run a snmptrap command from a client to test the configuration<o:p></o:p><o:p> </o:p><o:p> </o:p>What’s happen ?<o:p></o:p><![if !supportLists]>1) <![endif]>The TRAP has be sent from the client (HOST B)<o:p></o:p><![if !supportLists]>2) <![endif]>The TRAP has be received from Nagios server (HOST A)<o:p></o:p><![if !supportLists]>3) <![endif]>Nothing arrived on Nagios <o:p></o:p><o:p> </o:p>When I run manually submit_trap command, Nagios show me the message on the web gui<o:p></o:p><o:p> </o:p>I loose something between the TRAP received and Nagios service….what could be ?<o:p></o:p><o:p> </o:p>Moreover in the syslog a read many and messages like: snmptt-sys[501]: Unable to delete trap file #snmptt-trap-1347373658632581 from spool dir<o:p></o:p><o:p> </o:p>Regards<o:p></o:p>Marco<o:p></o:p><o:p> </o:p><div><div style='border:none;border-top:solid #B5C4DF 1.0pt;padding:3.0pt 0cm 0cm 0cm'>Da: Mike Lindsey <a href="mailto:[mailto:mike-nagios@5dninja.net]">[mailto:mike-nagios@5dninja.net]</a> Inviato: venerd� 7 settembre 2012 22:41 A: Nagios Users List Oggetto: Re: [Nagios-users] configure receiving snmp traps<o:p></o:p></div></div><o:p> </o:p> You'll need to ensure that snmptrapd is enabled on your Nagios poller, and the typical route from there to get snmp traps submitted into Nagios is to install SNMPTT. <a href="http://snmptt.sourceforge.net/">http://snmptt.sourceforge.net/</a> I recommend reading the docs for these, but, a very basic snmptrapd.conf would be: ###### snmptrapd.conf snmpTrapdAddr <a href="udp:localhost,udp:YOUR_IP_HERE,tcp:YOUR_IP_HERE">udp:localhost,udp:YOUR_IP_HERE,tcp:YOUR_IP_HERE</a> authCommunity log,execute public logOption f/var/log/snmptrapd.log traphandle default /usr/sbin/snmptt -i /usr/local/share/snmp/snmptt.ini ###### And then in the TrapFiles section of snmptt.ini you might have: ###### [TrapFiles] snmptt_conf_files = <<END /usr/local/share/snmp/snmptt/asyncos.conf END ###### In the included config file you map trap oids to script executions, like so: ###### asyncos.conf # snmptt.conf file for AsyncOS traps. # # All of these are stateless so the handler script needs to set and clear the service. # The service entry must have 0 retries set and be volatile. # # .1.3.6.1.4.1.15497 # # powerSupplyStatusChange # Status: .1.3.6.1.4.1.15497.1.1.1.8.1.2 EVENT powerSupplyStatusChange .1.3.6.1.4.1.15497.1.1.2.0.2 "asyncos" Critical FORMAT $N trap from $r EXEC /usr/local/nagios/customplugins/submit_trap $r AsyncOS-Trap_Alert $s 0 "$N: $*" # ##### Your submit_trap script takes that, and hands it off to Nagios. You can submit through NSCA, or you can create a result file in the checkresult directory, or you can submit through the external command pipe. I do it through NSCA: ##### submit_trap #!/usr/local/bin/bash PATH=/bin:/usr/bin:/usr/local/bin:/usr/local/nagios/customplugins:/usr/local/nagios/bin CONFIG=/usr/local/nagios/etc/send_nsca.cfg NSCA=`hostname` HOST=$1 SERVICE=$2 STATUS=$3 STATEFUL=$4 MESSAGE=$5 case $STATUS in "Critical") CODE=2 ;; "Warning") CODE=1 ;; "Normal") CODE=0 ;; *) CODE=3 ;; esac printf "%s\t%s\t%s\t%s\n" "$HOST" "$SERVICE" $CODE "$MESSAGE" | send_nsca -H $NSCA -c $CONFIG if [[ "$STATEFUL" == "0" ]] && [[ "$STATUS" != "0" ]] then # Clear Nagios via delayed at now that the volatile ticket's gone through. echo "/usr/local/nagios/customplugins/clear.sh $HOST \"$SERVICE\" \"$MESSAGE\"" | at now + 15 minutes fi ##### ... and clear.sh for clearing stateless alerts. ##### #!/usr/local/bin/bash PATH=/bin:/usr/bin:/usr/local/bin:/usr/local/nagios/bin:/usr/local/ironport/nagios/bin HOST=$1 SVC=$2 OUT=$3 if [[ "$HOST" == "" ]] || [[ "$SVC" == "" ]] then echo "Need host, service, optional message." exit 3 fi # Clear it printf "%b" "$HOST\t$SVC\t0\tWas:$OUT\n" | send_nsca -H `hostname` -c /usr/local/nagios/etc/send_nsca.cfg fi ##### If you're using the auto-clear bits, your Nagios user will need to be able to add items to the at queue, you'll need to look at your distribution's documentation on how that's managed. This is just one way of getting snmp traps working. Unfortunately none of them that I know of overly straightforward. Even if this doesn't work for you, it should give enough of an insight so that you've got a better idea on what to google for. Good luck.<o:p></o:p><pre>-- <o:p></o:p></pre><pre>Mike Lindsey<o:p></o:p></pre></div></body></html>