Unable to parse response

[vAJ]

Fusion 2014r1.0

Main dashboard has three tactical overview dashlets (2 XI instances, 1 Core). One of the XI instances occasionally will go red, drop out the message: "Unable to parse response"

unabletoparse.JPG

All authentication for the XI instances are session.

Broken image references:

Code: Select all

http://:/nagiosfusion/images/critical_small.png

A few refreshes of the page and it's back to good. Logs in /usr/local/nagiosfusion/var/components don't show anything suspect.

Thoughts?

[tmcdonald]

Is it always the same XI instance?

How big are the XI instances in terms of hosts + services?

You might be hitting a timeout.

[vAJ]

The one it's having a problem with is 1150+ hosts, 21000+ services.

[lmiltchev]

Do you see anything unusual in the poller.log?

Code: Select all

tail -100 /usr/local/nagiosfusion/var/poller.log

What is the "active" and "passive" polling intervals set at?

Configure->System Configuration

[tmcdonald]

Yea, almost certainly a timeout then. That's a lot just for the XI server without having to push it all across the network.

Let's check your php settings:

Code: Select all

grep "memory_limit" /etc/php.ini
grep "max_execution_time" /etc/php.ini
grep "max_input_time" /etc/php.ini
grep "default_socket_timeout" /etc/php.ini

[vAJ]

Code: Select all

Polling Top Alert Producers...
TRUNCATING topalertproducers table
Saved top alert producers for AustinDF
Saved top alert producers for VDC
Saved top alert producers for AmazonEC2
3 callbacks run
***GET DATA!***
RUNTIME: 73 INTERVAL: 73 POLL INTERVAL: 10
Saving Tac Data to DB: AustinDF
DONE

[vAJ]

Code: Select all

memory_limit = 256M
max_execution_time = 60
max_input_time = 60
default_socket_timeout = 60

[vAJ]

What is the "active" and "passive" polling intervals set at?

Configure->System Configuration

Active = 10
Passive = 270

[lmiltchev]

Try increasing the values below (in the "/etc/php.ini" file):

Code: Select all

memory_limit = 256M
max_execution_time = 60
max_input_time = 60
default_socket_timeout = 60

and restart apache:

Code: Select all

service httpd restart

See if this is going to fix your problem.

If you are still having issue, please open a new ticket in our email ticketing system. We may need to schedule a remote session to further troubleshoot this issue.

[vAJ]

Updated the values to 512MB and 90 seconds for all the others... still get the error.

Do you think 10 sec is too soon for the active polling interval?