On 25/10/2007, Graham Bloice <[EMAIL PROTECTED]> wrote: > > > > On 25/10/2007, dperalta <[EMAIL PROTECTED]> wrote: > > > > Hi all, > > > > Is there any way of monitoring ping responses time? The idea is to have > > a graph representing the Round-trip time for specific IP addresses. > > > > Thanks in advance > > > > > > > Using 2.1 : > > > 1. Navigate to Devices/Ping/Templates/Device. > 2. Edit the Description to suit, e.g. "Uses Nagios check_ping plugin > to monitor an IP device", and click Save. > 3. From the Data Sources drop down select "Add DataSource...". > 4. Enter a name for the data source, e.g. "ipcheck", and set the > type to "COMMAND". > 5. Set the Component, Event Class, Event Key, Severity and Cycle > Time fields as required. > 6. Enter the following into the Command field: "check_ping -H > $devname -w 180,100% -c 300,100% | sed -e 's# - Packet > loss#|loss#;s#%,##;s#ms##;s# = #=#g'", note spaces are critical, but drop > the outer quotes. > 7. Click Save. > 8. From the DataPoints drop down select "Add DataPoint...". > 9. Add a data point for the round trip averages, e.g. "RTA". > 10. Click Save, and from the breadcrumb trail at the top go back to > the DataSource, e.g. "ipcheck". > 11. Add another data point for the packet loss, e.g. "Loss". > 12. Click Save, and from the breadcrumb trail at the top go back to > the DataSource, e.g. "ipcheck". > 13. From the breadcrumb trail at the top, go back one level to the > template "Device". > 14. From the Thresholds drop down select "Add Threshold...". > 15. Enter a name for the threshold, e.g. RTA Limit. > 16. Select the data source for the threshold, e.g. "ipcheck_RTA". > 17. Set Event Class, Severity, Escalate Count as required, then > click Save. > 18. From the breadcrumb trail at the top, go back one level to the > template "Device". > 19. From the Graph Definitions drop down select "Add Graph...". > 20. Enter a name for the graph, e.g. "RTA". > 21. From the Graph Points drop down, select "Add DataPoint...". > 22. Select the correct data point, e.g. "ipcheck_RTA", and click OK. > 23. Edit the Units field to read "mS", and click Save. > 24. From the breadcrumb trail at the top, go back one level to the > template "Device". > 25. Repeat 19 - 24 for the Loss graph, setting the Units field to > "Counts". > > The template should now be set up, add a device with the "Ping" Device > Class Path, setting the Discovery Protocol to None. Wait a while (5 mins or > so depending on the Cycle Time you specified) and then select the Perf tab > for the new device and you should see some graphs. > > Notes. > > 1. The parameters to the checkPing command may need to be adjusted > for your requirements. Use check_ping --help to find out more. > 2. For some reason, on my system (Ubuntu 7.04, zenoss installed from > a tarball), the default zproperty "zCommandPath" was set to > "opt/zenoss/libexec", where it really needed to be > "/usr/local/zenoss/libexec". This may be a bug in the build, I'm not sure. > I fixed this by changing the zProperty at the top level "Device" template. > > A slight typo in the above, the data point name in point 11 should be "loss". It has to match the element name in the command response.
-- Regards, Graham Bloice
_______________________________________________ zenoss-users mailing list [email protected] http://lists.zenoss.org/mailman/listinfo/zenoss-users
