eth0 interface down

Question

We have a few Ubuntu servers in our environment that are monitored by Orion. I have setup SNMP on them using the standard Ubuntu SNMP install package. After building a custom yet simple snmpd.conf file, I restarted the SNMP service. Orion is able to report on all the resources (disk, memory, cache memory, etc) fine except eth0. On selected servers, eth0 is found to be DOWN and yet I can see all the statistics for the node. SNMP installation and snmpd.conf files are identical on servers which report eth0 correctly vs servers that don't. The only caveat here is that I did not build the servers. Why would eth0 be reported as down on some servers and not on others? This has been like this since day 1 of monitoring of the these servers. How do I troubleshoot this issue?

rharland2012 · Answer

Sorry - did mean eth0.

As far as how you can troubleshoot this:

I'd check the following:

look at your firewall config again and compare with a known-good.

what are your routes on the problem box? Any differences between known-good?

Just a short start, but it's a weird problem.

solarnewbie · Answer

Did you mean eth0? There is nothing connected on eth1. I usually test every connection from the Orion server (instead of my computer). So, yes, I can ping eth0 from the Orion server. Matter of fact, I use the same IP for SSH as well.

rharland2012 · Answer

Can you ping the IP of eth1 from your comp?