Nagios XI Migration to New Server

[kfanselow]

Hi Azreen,

Thank you for that information. The dashlet is pulling data from the nagiosxi database and we're interested in looking at some of the processes that update it. Specifically we would like to see the output from the script sysstat.php, your nagiosxi crontab file, as well as the status of the nagios user account on the system. It appears that my message from yesterday might have been delayed due to some system issues, my apologies; we would like to see the following:

Could you run the following as the user nagios and PM me a copy of the sysstat.log file along with your nagios crontab (/etc/cron.d/nagiosxi) ?

Code: Select all

 /usr/bin/php  /usr/local/nagiosxi/cron/sysstat.php >> /usr/local/nagiosxi/var/sysstat.log 2>&1 

Also could you post the output from the following command (as root):

Code: Select all

 chage -l nagios 

Thanks and Best Regards,
Keith

[azreenariff]

Hi Keith,

Unfortunately, as this is the weekend I couldn't access the server for now. Actually it is on our client's side, and we can only get to it via Anydesk access from their PC. We are the ones tasked to perform their Nagios upgrade.

By the way, I have actually asked them to reinstall the OS so I can reattempt to perform the upgrade again fresh as I suspected something was wrong with the DB during the previous attempt.

I will update you again soon.

Thanks.

[azreenariff]

Hi Keith,

After this second upgrade attempt on a clean OS, Nagios seems to be good now.

systemstatus3.png

However, I found that the vmware check have a problem that it's taking too long with 'plugin timed out (timeout 30s)' error

vmware1.png

The VMware vSphere SDK is version 7, as this is on a CentOS 8 OS

sdk1.png

Any suggestion on how to resolve this vmware plugin problem?

Thanks.

[kfanselow]

Good Morning,

We're glad to hear you were able to resolve the problem with fresh new install. Regarding the VMware plugin problem let's have you run the following commands as the user nagios, with the correct values inserted for USERNAME, VMWAREPASS, and IPADDRESS:

Code: Select all

/usr/bin/perl -w /usr/local/nagios/libexec/check_vmware_api.pl -u USERNAME -p VMWAREPASS -H IPADDRESS  -t 60  -l CPU

Code: Select all

/usr/bin/perl -w /usr/local/nagios/libexec/check_vmware_api.pl -u USERNAME -p VMWAREPASS -H IPADDRESS  -t 60  -l MEM

Code: Select all

/usr/bin/perl -w /usr/local/nagios/libexec/check_vmware_api.pl -u USERNAME -p VMWAREPASS -H IPADDRESS  -t 60 -l NET

If the commands time out you can try adjusting the -t argument to a value higher than 60 ( 30 is the default) such as 120 or 180 and see if it completes in that window.


Thanks and Best Regards,
Keith

[azreenariff]

Hi Keith,

Thanks for the suggestion. I ended up having to increase the timeout to 900 to have most of the checks returning results, although still there are a few randomly still times out. Also for services of one server they error out with the 'server version unavailable at...' error.

Have you any experience on the alternative plugin like box_293? Would that be better and resolve this you think?

Thanks.

[kfanselow]

Hi Azreen,

Unfortunately check_vmware_api can be slow in large, or complex, VMware environments. We're working to improve it however 900 seconds is a long delay and it may be worth exploring something like box_293. I personally do not have any experience with box_293 but others may have better feedback. My recommendation would be to start a new forum thread on box_293 -- I have a hunch others will know more or have thoughts on using it.


Thanks and Best Regards,
Keith

[azreenariff]

Hi Keith,

Sure. Thank you so much for your kind support.


Thanks.

[kfanselow]

Thank you - it's our pleasure to help. Locking it.