False: Host is down

[neurotox]

The problem:
I have add 7 servers(win2k3) to my cacti but cacti see all the 7 host down. Cacti can see SNMP information of those host so it can reach all of the 7 host.

Network information:
-Cacti and the 7 host are in 2 different network (logical and physical network)
-The 7 host is behind a gateway and share the same public IP
-The 7 host use a non-standard snmp port (yes I have configured the right port in "device")
-The 7 host use different port
-I have one device (switch) on this network how work properly and it use the standard port (161/udp)

Cacti information:
-My "downed host detection" is set to "SNMP - Reliable"
-Cacti 0.8.6h under Debian (unstable), Installed from package.

What I think:
The "downed host detection" "SNMP - Reliable" work only with port 161????

Thanks in advance.

Francois Rousseau, Neurotox

[neurotox]

Any suggestion?

Someone have already use "SNMP - Reliable" with a non-standard port?

Thanks

Francois Rousseau, Neurotox

[BSOD2600]

Turn the logging level to debug for one cycle and take a look at when cacti determines each host is up/down. Post the log.

[neurotox]

Hum thanks, I will post it as soon as I find this part of the log

[neurotox]

Here a part of the log:

05/05/2006 05:30:06 PM - CMDPHP: Poller[0] DEBUG: SQL Assoc: "select poller_output.output, poller_output.time, poller_output.local_data_id, poller_item.rrd_path, poller_item.rrd_name, poller_item.rrd_num from (poller_output,poller_item) where (poller_output.local_data_id=poller_item.local_data_id and poller_output.rrd_name=poller_item.rrd_name)"
05/05/2006 05:30:06 PM - CMDPHP: Poller[0] DEBUG: SQL Assoc: "select poller_id,end_time from poller_time where poller_id = 0"
05/05/2006 05:30:06 PM - CACTID: Poller[0] DEBUG: The Value of Active Threads is 1
05/05/2006 05:30:06 PM - CACTID: Poller[0] Host[34] DEBUG: HOST COMPLETE: About to Exit Host Polling Thread Function
05/05/2006 05:30:06 PM - CACTID: Poller[0] DEBUG: MySQL Query ID '72': OK
05/05/2006 05:30:06 PM - CACTID: Poller[0] DEBUG: MySQL Query ID '72': 'SELECT action,hostname,snmp_community,snmp_version,snmp_username,snmp_password,rrd_name,rrd_path,arg1,arg2,arg3,local_data_id,rrd_num,snmp_port,snmp_timeout FROM poller_item WHERE host_id=34 ORDER BY arg1'
05/05/2006 05:30:06 PM - CACTID: Poller[0] DEBUG: MySQL Insert ID '43': OK
05/05/2006 05:30:06 PM - CACTID: Poller[0] DEBUG: MySQL Insert ID '43': 'update host set status='1',status_event_count='228', status_fail_date='0000-00-00 00:00:00',status_rec_date='0000-00-00 00:00:00',status_last_error='Host did not respond to SNMP',min_time='9.999990',max_time='0.000000',cur_time='0.000000',avg_time='0.000000',total_polls='12',failed_polls='12',availability='0.0000' where id='34''
05/05/2006 05:30:06 PM - CACTID: Poller[0] Host[34] SNMP Result: Host did not respond to SNMP
05/05/2006 05:30:05 PM - CACTID: Poller[0] DEBUG: The Value of Active Threads is 2
05/05/2006 05:30:05 PM - CACTID: Poller[0] Host[33] DEBUG: HOST COMPLETE: About to Exit Host Polling Thread Function
05/05/2006 05:30:05 PM - CACTID: Poller[0] DEBUG: MySQL Query ID '71': OK
05/05/2006 05:30:05 PM - CACTID: Poller[0] DEBUG: MySQL Query ID '71': 'SELECT action,hostname,snmp_community,snmp_version,snmp_username,snmp_password,rrd_name,rrd_path,arg1,arg2,arg3,local_data_id,rrd_num,snmp_port,snmp_timeout FROM poller_item WHERE host_id=33 ORDER BY arg1'
05/05/2006 05:30:05 PM - CACTID: Poller[0] DEBUG: MySQL Insert ID '42': OK
05/05/2006 05:30:05 PM - CACTID: Poller[0] DEBUG: MySQL Insert ID '42': 'update host set status='1',status_event_count='228', status_fail_date='0000-00-00 00:00:00',status_rec_date='0000-00-00 00:00:00',status_last_error='Host did not respond to SNMP',min_time='9.999990',max_time='0.000000',cur_time='0.000000',avg_time='0.000000',total_polls='12',failed_polls='12',availability='0.0000' where id='33''
05/05/2006 05:30:05 PM - CACTID: Poller[0] Host[33] SNMP Result: Host did not respond to SNMP
05/05/2006 05:30:05 PM - CACTID: Poller[0] DEBUG: The Value of Active Threads is 3
05/05/2006 05:30:05 PM - CACTID: Poller[0] Host[25] DEBUG: HOST COMPLETE: About to Exit Host Polling Thread Function
05/05/2006 05:30:05 PM - CACTID: Poller[0] DEBUG: MySQL Insert ID '41': OK


Well I'm not sure if it is what you asking for. For me, it does not seem to be really helpful.

[neurotox]

I have continue my search...

It seem to be impossible to use a different port with the current snmpwalk version. It is possible?

I use snmpwalk version 5.2.2

[neurotox]

In case of someone can help me... here some new information:

With cmd.php, it's work fine
with cactid, it didn't work for any smnp over a different port number of the standard port (161/udp)


Any suggestion?

[neurotox]

I have found some information about a similar bug in the bug section:
http://bugs.cacti.net/view.php?id=691

But this bug concern the version 0.8.6g and is supposed to be resolv in the version 0.8.6h. I use the version 0.8.6h so I supposed is a miss configuration in my installation of cacti?

[BSOD2600]

Hmm, sounds like maybe the bug wasn't fixed yet...

You can try compiling the latest cactid from the svn repository and see if that fixes the problem. If not, give TheWitness a shout.

[neurotox]

Perfect, I will do it tonight.

[neurotox]

Well maybe tomorrow But I will do it and I will post the result of the test.

[tiggermanh]

I am experiencing the same problem. On any device where I am running snmp on any port besides 161, the status goes to down.

I am using the cactid poller, because I have 978 different data sources and cmd.php is too slow.

Some of my remote sites are running nat, with several snmp enabled devices behind the nat device. So I NAT say, 162,163,164 to the internal devices on port 161.

A shortcut I found, was to switch to downed host detection of ICMP only, but I fear this wont be as reliable.

[BSOD2600]

tiggermanh: the same suggestion applies to you. Download and compile the svn version of cactid to see if it fixes your problem.

[jga]

I suspect this is a bug
if you check with snmpwalk on command line and the machine responds correctly, (obiously try with a correct version and community)
snmpwalk -v 2c -c "public" 10.10.10.1 (ip and community yours)

The bug possibly can be fixed fastly, this is a problem when you not have any data source associated with host (with snmp), try checking this graphing anything on the host with problems