You are here

Opsview Graphing Import

7 posts / 0 new
Last post
stuart.laws
stuart.laws's picture
Opsview Graphing Import

Hi,

We've upgraded to 5.0.1 and since then have had a fair bit of trouble with Opsview Graphing Import becoming critical atleast once a day with the below 

Additional Information:

LOGS CRITICAL - Oldest log file is 3707 seconds old, 371 files backlogged

The only way i've found of being able to sort it is to restart opsview using /opt/opsview/watchdog/bin/opsview-monit

Thanks in advance

Duncan Ferguson
dferguson's picture
If this happens again can you

If this happens again can you please provide the output from running the following as root:

/opt/opsview/watchdog/bin/opsview-monit summary

and make sure all 'Process' lines show 'Running'. 

If 'Running' is not seen on one or more lines please run:

/opt/opsview/watchdog/bin/opsview-monit validate

  Duncs

stuart.laws
stuart.laws's picture
I've printed out exactly what

I've printed out exactly what i see when running those commands

/opt/opsview/watchdog/bin/opsview-monit summary
Status not available -- the monit daemon is not running

/opt/opsview/watchdog/bin/opsview-monit validate
'import_perfdatarrd' process is not running
'import_perfdatarrd' trying to restart
'import_perfdatarrd' start: /usr/bin/sudo
'opsviewnfd' process is not running
'opsviewnfd' trying to restart
'opsviewnfd' start: /usr/bin/sudo

I've tried running
pkill -u opsview
pkill -u nagios
service opsview-watchdog start
/opt/opsview/watchdog/bin/opsview-monit start all

and got this:

service opsview-watchdog start
Redirecting to /bin/systemctl start opsview-watchdog.service
Failed to issue method call: Unit opsview-watchdog.service failed to load: No such file or directory.

So the watchdog is still not running.

Thanks

Duncan Ferguson
dferguson's picture
Did you restart the opsview

Did you restart the opsview-monit service as the root user?

$ su - root
# service opsview-watchdog start
# /opt/opsview/watchdog/bin/opsview-monit summary

Can you confirm what OS you are on? If on RHEL or CentOS and that last command failed, please run as root:

rpm -V opsview-watchdog

to make sure the package is installed correctly

Duncs

stuart.laws
stuart.laws's picture
Sorry i'm running CentOS i

Sorry i'm running CentOS i tried to restart the monit service but get this

[root@* ~]# service opsview-watchdog start
Redirecting to /bin/systemctl start opsview-watchdog.service
Failed to issue method call: Unit opsview-watchdog.service failed to load: No such file or directory.
[root@*~]# /opt/opsview/watchdog/bin/opsview-monit summary Status not available -- the monit daemon is not running
[root@* ~]# rpm -V opsview-watchdog
[root@* ~]# service opsview-watchdog start
Redirecting to /bin/systemctl start opsview-watchdog.service
Failed to issue method call: Unit opsview-watchdog.service failed to load: No such file or directory.

stuart.laws
stuart.laws's picture
So after playing around for a

So after playing around for a little while today (sorry it's the first time I've had to look in to this properly) I've seemed to have gotten a little further then before. The install was done a fair while ago on your quick install but as said above since the update it hadn't worked.

now it seems that there is a problem with in the upgrade process that causes the watchdog not to correctly install (or atleast on my system) so i ran the below:

yum reinstall opsview-watchdog

I'm now able to see the watchdog monit daemon with a PID so hopefully it will fix but i will not know for sure for a little while.

Duncan Ferguson
dferguson's picture
Since you are running CentOS,

Since you are running CentOS, can you please ensure 'Defaults requiretty'  is disabled in your sudo configuration?  Having this enabled has also proven to break the watchdog.

  Duncs