Nagios XI Negative Durations

[Smark]

Hello,

As you can see in the below image I am seeing negative durations for some service checks.



I've browsed the forums here and can provide the following information:

Admin > System Profile:

Code: Select all

System:

Nagios XI Version : 2012R2.9
[redacted] 2.6.32-431.11.2.el6.x86_64 x86_64
CentOS release 6.5 (Final)
Gnome is not installed
Apache Information

PHP Version: 5.3.3
Agent: Mozilla/5.0 (Windows NT 6.1; WOW64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/35.0.1916.99 Safari/537.36
Server Name: [redacted]: [redacted]
Server Port: 80
Date/Time

PHP Timezone: America/Phoenix 
PHP Time: Thu, 08 May 2014 12:48:18 -0700
System Time: Thu, 08 May 2014 12:48:18 -0700

/etc/php.ini:

Code: Select all

[Date]
; Defines the default timezone used by the date functions
; http://www.php.net/manual/en/datetime.configuration.php#ini.date.timezone
date.timezone = America/Phoenix

/etc/localtime:

Code: Select all

[root@xxxnagios01:~] ls -alh /etc/localtime
lrwxrwxrwx 1 root root 35 Apr 28 08:27 /etc/localtime -> /usr/share/zoneinfo/America/Phoenix

NTP Status:

Code: Select all

[root@xxxnagios01:~] service ntpd status
ntpd (pid  10412) is running...

NTP Config:

Code: Select all

server xxxntp01.[redacted]
server xxxntp02.[redacted]
server yyyntp01.[redacted]

The server is a VM running on VMware ESXi 5.5 Update 1 with the NTP service running pointed to those same three time servers.

Any idea how I can correct this?

Thank you for your time,
Smark

[abrist]

Maybe there is uncorrected clock skew:

Code: Select all

date
ntpdate xxxntp01.[redacted]

Is your browsers OS date correct?

[Smark]

Code: Select all

[root@xxxnagios01:~] service ntpd stop
Shutting down ntpd:                                        [  OK  ]
[root@xxxnagios01:~] date && ntpdate xxxntp01.[redacted]
Fri May  9 10:02:37 MST 2014
 9 May 10:02:37 ntpdate[11673]: adjust time server xxx.xxx.xxx.xxx offset -0.016526 sec
[root@xxxnagios01:~] service ntpd start
Starting ntpd:                                             [  OK  ]

The time above is within a second of matching my desktop's time. I assume the browser time is the same.

[sreinhardt]

You might be running into one of several potential services being (re)started prior to an ntp update that would correct time. Such that when php\nagios\javascript gets the time from your nagios system and displays it in your browser, due to the applications time being off it is displayed strangely within your browser. Let's try restarting some of those key services and see how it reacts.

Code: Select all

service nagios stop
service ndo2db stop
service mysqld stop
service httpd restart
service mysqld start
service ndo2db start
service nagios start

Give it a few moments to schedule things normally, and check to see if the time is still skewed.

[Smark]

Looks like it's fixed. I'm not getting any more negative durations and ndo2db was not running when I tried to stop it.

Thanks for all your help!

[Smark]

I check back and am still experiencing the same issues that the sequenced restart fixed before. The breakage seems to be aligned with my upgrade to 2014R0.

Any ideas? I will try to post back tomorrow with additional details.

[Smark]

It appears that the 2014R1.0 upgrade reverted the /etc/timezone link.

I'll make a note to change that back after an upgrade. It is currently fixed.

Can we leave this thread open until I verify that this is still functioning correctly after a week? I'll post back on June 5th.