snmp timesout getting CPU usage

[brdr]

Hi,

We use XI 5.2.2.

This is not a Nagios issue. More a snmp question, and wondering if you have seen this before.

We run snmp against all our Linux servers (about 200). For 3 like servers we are getting snmp timeouts from command line. I have narrowed it down a problem reading the hrDeviceTable. This table holds snmp info for Processors and Network Devices.

The command is:
snmpwalk -v2c -c Ronly1 server_1 hrDeviceType

The output
.
.
HOST-RESOURCES-MIB::hrDeviceType.196639 = OID: HOST-RESOURCES-TYPES::hrDeviceProcessor
HOST-RESOURCES-MIB::hrDeviceType.262145 = OID: HOST-RESOURCES-TYPES::hrDeviceNetwork
Timeout: No Response from server_1

I have concluded that hrDeviceType.262145 is interface lo, and the next one 262145 is interface eth0

I can run successful the command below against eth0
./check_snmp_int.pl -H bed-400-143 -C Ronly1 -2 -n eth0 -f -w 80,80 -c 90,90
eth0:DOWN: 1 int NOK : CRITICAL

My question is: Have you seen this before? That snmp stops reading DeviceType because it timesout on eth0?

FYI... eth0 is down by design.

[Box293]

This article seems to be similar to what you are talking about:

https://sourceforge.net/p/net-snmp/mail ... /10732498/

Perhaps you could increase the timeout in snmpwalk to something like -t 20

[brdr]

Thanks Box293! This gives me enough. Putting a higher timeout works.

I did search a lot for answers on web and in this forum before reaching out to support.

As always much appreciated. Please close.

[rkennedy]

No problem! We're glad to help.

As requested, I'll mark this as resolved and close it out. Feel free to open a new thread if you need assistance in the future.