Graphs for only 1 Device Not Working

[invisimax]

I'm new to the list, so I apologize ahead of time if this is basic information or an old topic.

I have a device that is not graphing. All my other devices are working just fine. I have checked the snmp information and it appears to be fine. When I click on the device it shows that is is down. The Ip address is correct.

I was searching the faq's and turned on the graph debug mode. The RRDTool says: ERROR: opening '/var/www/cacti/rra/finley_cooperstown_58_feed_5min_rate_1329.rrd': No such file or directory

Not sure if this is the cause of the graph not showing or if it's just stating that it can't find the graph. Please let me know if I need to provide more information.

[gandalf]

... When I click on the device it shows that is is down. The Ip address is correct.

I was searching the faq's and turned on the graph debug mode. The RRDTool says: ERROR: opening '/var/www/cacti/rra/finley_cooperstown_58_feed_5min_rate_1329.rrd': No such file or directory

The first issue will _always_ be followed by the second one. If SNMP info is shown correctly for this host, then please check Downed Host Detection (Settings) and change this to "SNMP - Reliable"
Reinhard

[invisimax]

That fix seemed to work, will this affect any of my other working graphs?

I really wouldn't need to ping these devices as I am using Nagios to determine whether those devices are up or down.

[gandalf]

This "fix" doesn't change the graphs behaviour. It is a customization setting for downed host detection. This feature was introduced lately to suppress stupid polling of hosts that are detected as down.
The detection is based on several methods: SNMP will do an SNMP request and assume the host is up when this succeeds.
PING can be configured to ICMP/UDP and will "ping" the host. But firewalls tc. may suppress this so the host is assumed to be down when it really isn't.
First, all tests had to be successfule for the host to be detected as up. In 0.8.6h releases, this was changed so if one test succeeds, the host is assumed to be up. See release notes for more
Reinhard

[TheWitness]

LVM,

With 0.8.6h, I have changed the behavior to allow the host to be up is "EITHER" the snmp or icmp/udp pings respond. I specifically made this change due to the delays in getting 0.9 out the door. Therefore, the preferred method moving forward (until 0.9) will be ICMP/UDP and SNMP pinging.

TheWitness