Nagios XI takes too much time to show real status

[morabanc]

Hi,

We have a big Nagios XI implementation with external MySQL Database. Now we have 10000 checks.

The thing is, if we go to Nagios Core to one specific services and we disable notifications for this specific check. This change takes 10 or more minutes to be reflected in the NagiosXI web interface.

Then, what we can see in NagiosXI web interface its not real...

This server it's in production...

[tmcdonald]

Sounds like you might need to implement some of our performance tweaks.

Do you have a ramdisk set up? https://assets.nagios.com/downloads/nag ... giosXI.pdf

Are you using passive checks at all? These can reduce load on the server. What is your load?

How many CPUs/cores do you have? How much RAM?

What sort of checks are you performing? ESX, WMI, and SSH checks can take a lot of resources.

How often are you checking on average? If they are all checking every minute then maybe some can be bumped to every 5 or 10.

[Box293]

In addition to tmdonald's post, you should read through the following guides on improving performance:

http://assets.nagios.com/downloads/nagi ... rmance.pdf

http://assets.nagios.com/downloads/nagi ... ios_XI.pdf

http://assets.nagios.com/downloads/nagi ... Server.pdf

[morabanc]

Hi

We had the data base offloaded, but we returned this one to the server and the problems are solved...

It don't looks like performance, as all system metrics are ok.

[lmiltchev]

We had the data base offloaded, but we returned this one to the server and the problems are solved...

Offloading the mysql database to a remote server improves performance, unless there is a slow connection to the remote server (or intermittent networking issues). Let us know if it is OK to lock this topic. Thanks!

[morabanc]

Hi,

I mean, it's not a server problem or resources problem, the thing it's that NAgios Core + Gearmand + MySQL are working fine, but the XI interface it's very slow. For example:

- After apply new changes/configurations NagiosXI needs more then 5 or 10 minuts to be in ready status.
- If a host goes down, you can see the problem in the Nagios Core but the change is not reflected to the XI interfaces until after 5 or 10 min.
- If someone disable the notifications for one host, this change is not visible in the XI interface until after 5 or 10 min

We moved the Data Base to an external server, but the problem is the same. The XI interface is working very slowly.

I attached some images with Information about System & Nagios Status

Regards

[morabanc]

Find Attached the CPU and Free commands output.

[lmiltchev]

Is mysql currently local or offloaded? Do you have any errors in the mysqld.log? Do you have many users accessing the web UI at the same time? It is possible that you are hitting a mysql connection limit? Can you run the following two commands and show us the output?

Code: Select all

echo "show processlist;"|mysql -pnagiosxi|wc -l
cat /etc/my.cnf|grep max_connections

Also, are you using NAT, Proxy, or SSL on this system? Have you tried tweaking the performance settings (selecting the "Use Unified Tactical Overview", "Use Unified Hostgroup Screens", and "Use Unified Servicegroup Screens" check-boxes)?

Admin->Performance Settings->Pages

[tmcdonald]

Additionally, please run the following command as root from the Nagios server command line and post the results:

Code: Select all

ipcs -q

[morabanc]

Solved,


With local or external DB Server the Solution was add the following lines. The line that solved the problem is "innodb_flush_log_at_trx"

#InnoDB Settings
innodb_buffer_pool_size=3G
innodb_flush_log_at_trx_commit=2

#Other
max_connections=500
max_connect-errors=1000
skip-name-resolve

Now it's working ok with local and external db.

Thanks