Hello Community.
Can you please tell me how to fix this, I don't understand why this is happening. I have explored various topics but have not been able to find a solution.
I have an application which is configured by Splunk_TA_nix on remote servers.
But not all servers are getting the CPU=all field
I first encountered this when a team with their dashboard contacted me. They had 2 lonely servers. On one of them the CPU field was extracted and the dashboard worked. On the other one it didn't work anymore.
I have set up a new server to forward the logs. But there was no CPU field on that one either.
I even installed the sysstat utility. But I can't figure it out yet. Thus I am asking for help. Regards to everyone
Hi @igor04653,
in addition to the question of @richgalloway, could you check if the version of installed Splunk_TA_nix is the same of the other servers?
Ciao.
Giuseppe
Thank you for responding to my question. I checked the Splunk_TA_nix version. On the working server and on the new one where it doesn't work - they are the same.
In the inputs.conf settings the same parameters are included.
But I noticed a difference in the versions of the Universal Forwarder agent
On the working server, version 7.3.3 is used
I installed 8.2.6
There is also a difference in operating system versions.
The working one uses Oracle Linux. And I do the setup on Ubuntu server 20.04.
Hi @igor04653,
after UF upgrade is the problem still present?
The operative system version shouldn't be so relevant, to be sure, try to manually run scripts in that UF.
Ciao.
Giuseppe
Is the TA installed on all servers and is the cpu_metric.sh input enabled on them all?
Thank you for responding to my question. I checked the parameters of the inputs.conf file
They are configured like this on both servers
# Shows stats per CPU (useful for SMP machines)
[script://./bin/cpu.sh]
sourcecetype = cpu
source = cpu
interval = 300
index = os
disabled = 0