Does TheDude support flipflop with regards to ping monitors?
If it does, how can you change the number of failed attempts before it considers a site to be down?
I’m asking as recently we’ve been getting packet loss and the dude is giving false reports that a site is down when infact it’s simply a case of the packets are getting dropped along the way.
Thanks all.
Any response to this i am having a similar problem
The ping command is configurable just open the probes panel or double click on the ping service. You can modify the retry count and the interval. I watched it work in wireshark. A single ping is sent every probe interval (probe intervals can be set per device, per map and globally) and if there is no response it will attempt to retry based on the ping Retry Count. If ping Retry Count is reached then the failed probe count is incremented. The device will not be “down” until the Probe Down Count is exceeded.
So if you have ping Retry Count set to 3, a ping Retry Interval of 5000ms, a Probe Interval of 30 and Probe Down Count set to 3 when a device fails it will take 15 seconds after the first ping for the ping probe to register one Probe Down Count and the device will show down at 105 seconds.
I hope you don’t have that much packet loss! I do not know what flipflop means so if you would explain that would be good. I have found that 30 second probe intervals can generate false positives although I do have about 1000 probes.
And finally Probes based on SNMP by default have a 300 second negative cache time meaning if they fail they stay failed for 5 minutes. This “feature” needs to be configured by modifying each probe or function that uses SNMP.
HTH,
Lebowski