5.5.0 upgrades resuting in more 'unknowns'

[JGCG]

Out of curiosity, do you know what version you upgraded from?

Also, do you know if you have every needed to edit this plugin in previous versions?

We upgraded from 5.4.13 to 5.5.0 on the 9th and then to 5.5.1 yesterday, and have not needed to edit this script.
We are still in the process of migrating from Nagios 3 to XI (unable to go live until it's stable) and the older versions which are monitoring the same service in the same way have no issues (I can't ever recall them going unknown).

They are set at different check intervals to elimate the checks potentially executing at the same time, but I can't see this being an issue any way.

[hbouma]

I have been able to duplicate the issue with SNMP after the upgrade from 5.4.13 to 5.5.0 then 5.5.1.

Running Red Hat 7 64bit VM's. Both systems fully updated via yum with the same repositories.

check_xi_service_snmp_linux_storage! -C $COMMUNITYSTRING --v2c -m "^/boot$" -w 80 -c 95 -f works on multiple servers running 5.4.13
check_xi_service_snmp_linux_storage! -C $COMMUNITYSTRING --v2c -m "^/boot$" -w 80 -c 95 -f fails on the server running 5.5.1. Error output of ERROR: Alarm signal (Nagios time-out)

I have verified that both hosts are allowed in our hosts.allow file and verified that the hosts are allowed in the snmpd.conf file.

When attempting to re-run the configuration wizard, it works fine from the 5.4.13 servers, but takes forever on the 5.5.1 server. The 5.5.1 server eventually giving the error:

"The wizard detected that this server does not have snmpwalk permission on the target host. This will prevent auto population of processes and prevent them from running successfully, but you can continue with the wizard manually. To troubleshoot this ensure that these OIDs are available on the target host: "HOST-RESOURCES-MIB::hrSWRunName" and "HOST-RESOURCES-MIB::hrFSMountPoint"."

Running "snmpwalk -v 2c -c COMMUNITYSTRING XXXXXXXXX" works fine on the servers running 5.4.13, but fails on the server running 5.5.1.

Could the update have some of the SNMP files on the server?

[scottwilkerson]

Could you send me the check_snmp_storage.pl plugin from both servers specifying which is which?

[hbouma]

Could you send me the check_snmp_storage.pl plugin from both servers specifying which is which?

Sent. When I compare them, they are identical.

net-snmp-utils are both the same version as well. 5.7.2-33

[scottwilkerson]

Well that rules out plugin problems.

Can you also verify the check_xi_service_snmp_linux_storage commands in the CCM on both servers and see if they match

If they do lets lengthen the timeout on the new server by changing the command to the following:

Code: Select all

$USER1$/check_snmp_storage_wizard.pl -t 30 -H $HOSTADDRESS$ $ARG1$

Thanks

[hbouma]

Nagios XI 5.4.13:
$USER1$/check_snmp_process_wizard.pl -H $HOSTADDRESS$ $ARG1$

Nagios XI 5.5.1:
$USER1$/check_snmp_storage_wizard.pl -H $HOSTADDRESS$ $ARG1$

Change made to add the timeout to the command (-t 30), and it is still timing out.

When run from the command line, I get:
[nagios@XXXXXXXX~]$ /usr/local/nagios/libexec/check_snmp_storage_wizard.pl -H XXXXXXXXX-t 30 -v -C COMMUNITYSTRING --v2c -m "^/boot$" -w 80 -c 95 -f
Alarm at 15
SNMP v2c login
ERROR: General time-out (Alarm signal)

[scottwilkerson]

Nagios XI 5.4.13:
$USER1$/check_snmp_process_wizard.pl -H $HOSTADDRESS$ $ARG1$

Nagios XI 5.5.1:
$USER1$/check_snmp_storage_wizard.pl -H $HOSTADDRESS$ $ARG1$

These are 2 different plugins. Are you sure you compared the same command?

[scottwilkerson]

Also, if you cannot run the snmpwalk or the plugin from the command line, are you sure you don't have something blocking the UDP packets on the 5.5.1 system?

[hbouma]

Ok, silly me. There was a typo in my hosts.allow file.

This fixes my issue, but may be different from JGCG's issue.

[scottwilkerson]

Ok, silly me. There was a typo in my hosts.allow file.

This fixes my issue, but may be different from JGCG's issue.

I am glad you got that resolved!

I'll leave the thread open for the OP.