Missing performance graph data on service after restore

[dshearon]

We recently ran into issues with our CentOS 6.6 installation and took the chance to upgrade to CentOS 7. In order to accomplish this we created a clean OS install on a new VM. Installed the 2014R2.7 version of XI and upgraded it to the latest. I then transferred the backup from our old installation and restored it on the new box. Everything including performance data transferred fine and it appears to be working great. I am running into an issue where the performance graph on one of services appears but is blank after the time of the restore. I have confirmed that the service check is working and getting proper results. Any ideas?

[dshearon]

To add some more information this may be a symptom of a larger issue. The service in question is a single service with multiple hosts added. The performance graphs for the other hosts are working fine. To see if it would help I removed the problem host from the service and applied configuration which had no errors. I waited a minute and then re-added the host and applied configuration again with not issues. The problem now is that service is not even showing up for that host in the service group.

[lmiltchev]

Can you show us the configs for the service, host, servicegroup, and any relevant templates that they are using? Thanks!

[dshearon]

I actually failed back over to our CentOS 6.6 installation as I was finding other small issues with the new install. Is there a proper migration path to move Nagios from a CentOS 6 server to a CentOS 7 server?

[jolson]

We don't have a foolproof migration path - the backup + restore that you mentioned is the exact procedure you should be using, keeping in mind that you should be backing up and restoring to the same versions.

When you backed up your installation on CentOS 6.6, what version of Nagios XI were you running at the time? It should be the same version that you're restoring to.

[dshearon]

Both Nagios installations were 5.2.2. I did run into an issue on the first go around as I installed 5.2.2 directly and found out that since I upgraded from 2011 on the previous box I needed to do the same on the new box for the MySQL issues. My process was to install 2011, upgrade to 5.2.2 on the new server, backup the CentOS 6.6 box and restore to the CentOS 7 box. I did upgrade Nagios on the new box to 5.2.3 after the restore was complete. I will try it again with another clean install and document the issues. Most of them appeared to be related to the DB, for instance I would remove a host from a service and it wold still show on the summary screen even after applying the configuration successfully. In one instance I removed a host and it was no longer on the summary screen but adding it back did not return it.

[ssax]

Was it CentOS 6.6 32bit to CentOS 7 64bit or CentOS 6.6 64bit to CentOS 7 64bit?

[dshearon]

It was CentOS 6.6 x64 to CentOS 7 x64.

[lmiltchev]

I will try it again with another clean install and document the issues.

Please, do that. Meanwhile, show us the config of the *problem* service, a screenshot of the "Service Status Detail" page ("Advanced" tab), and permissions of the RRD, XML file, and the containing directory:

Code: Select all

ll -d /usr/local/nagios/share/perfdata/<hostname>
ll /usr/local/nagios/share/perfdata/<hostname>