Bahedile, Look at your NMIS Runtime Graph - what is the average and maximum Collect Time? From there check the Collection Time for the device; if it exceeds 23s this is usually a sign that the device is timing out. Finally, I would go to the command line on the NMIS server and run an SNMP walk on the device. This should give you an indication of the device's response ability. Your final check would be to SSH directly to the device and check it's CPU and memory availability.
The last update column shows when NMIS was last able to complete a polling cycle on that individual device. There are several reasons a polling cycle may not complete, these can include loss of connection to the device, SNMP down or bad SNMP Community string, or the device may be overloaded and not responding within the maximum polling time.
The Device Details screen should give you an indication if it's a ping or SNMP related issue. To diagnose slow response times first check the NMIS Runtime Graph (System->Host Diagnostics->Runtime Graph); your Collect Time must be shorter than your polling interval (i.e. 300 seconds). You should also check the Collection Time (Reports->Current->Collect/Update Time); once this opens sort on the Collect Time column and look for devices with excessively long collection times.
SNMP string is fine. i suspect its related to devices overloaded. how do i see if the device is overloaded?
NMIS Runtime Graph- average collect time 603.14 secs and Maximum collect time 912.26 secs. i have only one with 25.66secs collect time. all seems to be ok besides graph collect time averaging 600secs. what causes this ERROR: the improbable has happened, a UUID conflict has been found for 4a385a1d-b0c3-11e6-bd66-969d98e3eb88 and Can't use string ("HASH(0x3659e20)") as a HASH ref while "strict refs" in use at /home/nmis/bin/nmis.pl line 2097.
Behedile, Your average collection time is over 10 minutes suggesting either you have a large number of devices not responding properly or the server is overloaded. How many devices are you monitoring and what is your server configuration (i.e. CPU, RAM, etc).
true, i've a lot of devices not responding properly. before i upgraded NMIS to 8.6.0 CPU and Ram on the VM machine were increased. number of devices monitored is around 700 nodes.
Intel Xenon CPU x5675 3.0Ghz, 20GB Ram, used : 7,2Gb and Free 12,8.
Mark, how do i see if the server is overloaded. is there anything i can do to resolve it besides all i mentioned above?
Bahedile, Let's coordinate a quick screen share and get to the bottom of this. Please send an email to me at email@example.com and we can arrange a date and time.
will drop you a mail. thanks Mark