Windows Server Not Showing Memory Usage in Nagios Xi Metrics

[sarfarosh]

Hello Team,
We was getting wrong memory usage for windows server while using check_nt. So we did required changes to NSC.ini in windows and used check_nrpe for getting physical memory. After doing this we are getting correct values of memory usage. But now we are not able to see that windows server under metrics memory usage of naigos xi. We tried the same with 4 windows server and all 4 servers are now not showing in metrics memory usage.

Please find the attached .docx and NSC.ini for the brief details.

[mcapra]

There may be older performance data from the check_nt checks that is being referenced. Can you share the output of the following command executed from the CLI of your Nagios XI machine (replace <host> with the Nagios XI host name):

Code: Select all

ls -al /usr/local/nagios/share/perfdata/<host>/

If i'm right, we should see 4 unique files representing the memory usage of this host. What I suspect is that the check_nt data is referenced (which is likely incredibly stale now) rather than the check_nrpe data.

[sarfarosh]

Hello mcapra,
Please find the output of command ls -al /usr/local/nagios/share/perfdata/pclpadkm.mypiramal.com/

[root@nagiosxi ~]# ls -al /usr/local/nagios/share/perfdata/pclpadkm.mypiramal.com/
total 6076
drwxrwxr-x 2 nagios nagios 4096 Mar 29 10:31 .
drwxrwxr-x. 552 nagios nagios 20480 Mar 17 13:52 ..
-rwxrwxr-x 1 nagios nagios 384952 Mar 29 10:28 CPU_Usage.rrd
-rw-rw-r-- 1 nagios nagios 2222 Mar 29 10:28 CPU_Usage.xml
-rwxrwxr-x 1 nagios nagios 384952 Mar 29 10:29 Drive_C__Disk_Usage.rrd
-rw-rw-r-- 1 nagios nagios 2392 Mar 29 10:29 Drive_C__Disk_Usage.xml
-rwxrwxr-x 1 nagios nagios 384952 Sep 28 09:03 Drive_D__Disk_Usage.rrd
-rwxrwxr-x 1 nagios nagios 2014 Sep 28 09:03 Drive_D__Disk_Usage.xml
-rwxrwxr-x 1 nagios nagios 1534768 Mar 29 10:31 _HOST_.rrd
-rw-rw-r-- 1 nagios nagios 4067 Mar 29 10:31 _HOST_.xml
-rwxrwxr-x 1 nagios nagios 384952 Jul 27 2016 Logon_Errors.rrd
-rwxrwxr-x 1 nagios nagios 2109 Jul 27 2016 Logon_Errors.xml
-rwxrwxr-x 1 nagios nagios 384952 Mar 29 10:30 Memory_Usage.rrd
-rw-rw-r-- 1 nagios nagios 3055 Mar 29 10:30 Memory_Usage.xml
-rwxrwxr-x 1 nagios nagios 384952 Mar 29 10:29 Page_File_Usage.rrd
-rw-rw-r-- 1 nagios nagios 2499 Mar 29 10:29 Page_File_Usage.xml
-rwxrwxr-x 1 nagios nagios 1534768 Mar 29 10:31 Ping.rrd
-rw-rw-r-- 1 nagios nagios 4180 Mar 29 10:31 Ping.xml
-rwxrwxr-x 1 nagios nagios 384952 Mar 29 10:28 Server_Work_Queues.rrd
-rw-rw-r-- 1 nagios nagios 2750 Mar 29 10:28 Server_Work_Queues.xml
-rwxrwxr-x 1 nagios nagios 384952 Mar 29 10:29 Uptime.rrd
-rw-rw-r-- 1 nagios nagios 2145 Mar 29 10:29 Uptime.xml

[mcapra]

My next thought would be to move these files out of the main perfdata path:

Code: Select all

-rwxrwxr-x 1 nagios nagios 384952 Mar 29 10:30 Memory_Usage.rrd
-rw-rw-r-- 1 nagios nagios 3055 Mar 29 10:30 Memory_Usage.xml

And put them in some sort of backup path for the time being. Fresh files will regenerate over the course of a day or so, then I would think this host would show back up in the metrics appropriately.

If there is good data in there, we can attempt to merge it into the newly generated file later on.

[sarfarosh]

Hi mcapra,
I moved the existing Memory_Usage.rrd and Memory_Usage.xml files yesterday and checked today. Files are created but still not showing memory usage for the host(pclpadkm.mypiramal.com) in metrics.

[root@nagiosxi pclpadkm.mypiramal.com]# ls -l
total 6428
-rwxrwxr-x 1 nagios nagios 384952 Mar 31 14:33 CPU_Usage.rrd
-rw-rw-r-- 1 nagios nagios 2222 Mar 31 14:33 CPU_Usage.xml
-rwxrwxr-x 1 nagios nagios 384952 Mar 31 14:33 Drive_C__Disk_Usage.rrd
-rw-rw-r-- 1 nagios nagios 2392 Mar 31 14:33 Drive_C__Disk_Usage.xml
-rwxrwxr-x 1 nagios nagios 384952 Sep 28 2016 Drive_D__Disk_Usage.rrd
-rwxrwxr-x 1 nagios nagios 2014 Sep 28 2016 Drive_D__Disk_Usage.xml
-rwxrwxr-x 1 nagios nagios 1534768 Mar 31 14:31 _HOST_.rrd
-rw-rw-r-- 1 nagios nagios 4067 Mar 31 14:31 _HOST_.xml
-rwxrwxr-x 1 nagios nagios 384952 Jul 27 2016 Logon_Errors.rrd
-rwxrwxr-x 1 nagios nagios 2109 Jul 27 2016 Logon_Errors.xml
-rwxrwxr-x 1 nagios nagios 768224 Mar 31 14:35 Memory_Usage.rrd
-rw-rw-r-- 1 nagios nagios 3055 Mar 31 14:35 Memory_Usage.xml
-rwxrwxr-x 1 nagios nagios 384952 Mar 31 14:33 Page_File_Usage.rrd
-rw-rw-r-- 1 nagios nagios 2499 Mar 31 14:33 Page_File_Usage.xml
-rwxrwxr-x 1 nagios nagios 1534768 Mar 31 14:31 Ping.rrd
-rw-rw-r-- 1 nagios nagios 4180 Mar 31 14:31 Ping.xml
-rwxrwxr-x 1 nagios nagios 384952 Mar 31 14:30 Server_Work_Queues.rrd
-rw-rw-r-- 1 nagios nagios 2750 Mar 31 14:30 Server_Work_Queues.xml
-rwxrwxr-x 1 nagios nagios 384952 Mar 31 14:30 Uptime.rrd
-rw-rw-r-- 1 nagios nagios 2145 Mar 31 14:30 Uptime.xml

[mcapra]

@lmiltchev actually did some investigation into this, and it's a bug within the Metrics component that we're going to need to fix. A big re-work of that component to better include 3rd party plugins is currently on our to-do list.

I do apologize for the inconvenience. Let us know if you have additional questions regarding this.

[sarfarosh]

Hello mcapra,
Please go through the link https://support.nagios.com/forum/viewto ... 16&t=42210. I want to know the workaround that has been performed from your side to resolve it. The issue of incorrect value in %Utilization is resolved, but i can see the Yellow(warining) and Red(critical) color even when the threshold is not reached for linux servers.

Also please let me know when the subjected issue can be resolved. As we are facing this since long time and need to do lot of manual work before providing the data to client, which is time consuming.

[mcapra]

I talked with @ssax about this, and the changes he made were a one-off deal and not really in the same wheelhouse as the full on Metrics Component re-work our devs are working on.

but i can see the Yellow(warining) and Red(critical) color even when the threshold is not reached for linux servers.

Can you send us a sampling of your Nagios logs from the past few days? They should be located in /usr/local/nagios/var/archives.

Also please let me know when the subjected issue can be resolved.

I do not have a timeline for the Metrics Component re-work at this time.

[sarfarosh]

Hello mcapra,
Please find the sampling of nagios logs of last 4 days in attachment.

[mcapra]

The issue of incorrect value in %Utilization is resolved, but i can see the Yellow(warining) and Red(critical) color even when the threshold is not reached for linux servers.

Can you tell me which hosts are producing these false alerts? Everything I see in the logs seems to be good math, though the numbers the plugin is handling may not be correct.