Page 1 of 1
Nagios Reports too slow
Posted: Tue Nov 12, 2019 5:05 am
by iam33
HI!
Thanks in advance.
For some time, we are having problems with Nagios reports, such as the Alert History of a service, which takes several minutes to show.
When this process is executed, the CGI process takes 100% CPU.
I have seen that this problem occurred in previous versions, but it was supposed to be resolved.
What could be the problem?
How can I expedite the creation of reports?
Nagios Core 4.3.4
Total Services: 3191
Total Hosts: 151
Mem: 3924192
2 CPUSs.
Thanks
Re: Nagios Reports too slow
Posted: Tue Nov 12, 2019 5:51 pm
by benjaminsmith
Hello
@iam33,
Have you checked the nagios log and Apache logs to make sure the server is running normally without errors?
1. To check the Apache Logs
Code: Select all
tail -n 50 /var/log/httpd/*error_log
2. To check the Nagios log
Code: Select all
tail -n 50 /usr/local/nagios/var/nagios.log
3. What time period are you running the report on?
4. Can you post the output of the top command while running the report?
5. Is this a physical server or a VM? Thanks.
Re: Nagios Reports too slow
Posted: Wed Nov 13, 2019 8:36 am
by iam33
Hi!
1º- There are not errors in Apache Logs
2º- There are not errors in Nagios Logs
3º There is not time period defined for report, it´s on demand
4º- The output of top:
Code: Select all
top - 14:28:43 up 114 days, 3:38, 4 users, load average: 1.10, 1.74, 1.84
Tasks: 247 total, 3 running, 244 sleeping, 0 stopped, 0 zombie
Cpu(s): 52.6%us, 16.9%sy, 0.0%ni, 30.3%id, 0.0%wa, 0.0%hi, 0.2%si, 0.0%st
Mem: 3924192k total, 3727676k used, 196516k free, 540724k buffers
Swap: 4194300k total, 900564k used, 3293736k free, 1798796k cached
PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ COMMAND
18740 apache 20 0 659m 561m 840 R 99.3 14.6 0:18.85 history.cgi
5º - It´s an VM Server.
thanks!!
Re: Nagios Reports too slow
Posted: Wed Nov 13, 2019 3:59 pm
by benjaminsmith
Hello
@iam33,
Thanks for answering those questions and sending over the top command results. Actually, that looks pretty good. The Io wait is 0 and the overall CPU load average is normal. Also, It's normal to see high CPU load with running the report.
My recommendation would be to increase the number of virtual CPUs for this VM. Additionally, we've made improvements since 4.3.4 and it would be a good idea to upgrade to the latest version.