Disabled the "cacti to Device" connectivity, Still showing the device as Up.

[HeshanD]

Hi Guys,

I recently did a cacti setup and now doing the testing. During the testing i found out even the cacti doesn't have the connectivity to the device, in cacti monitoring and in device list section, it still shows as Up.
Need to know why and which misconfiguration that i have done for this outcome.

>>> Device Config

device config1.PNG (63.31 KiB) Viewed 1141 times

device config2.PNG (51.67 KiB) Viewed 1141 times

>>> In Device List, it shows as Up

device list.PNG (88.45 KiB) Viewed 1141 times

>>> In Monitoring it shows as Up

monitoring.PNG (35.78 KiB) Viewed 1141 times

[TheWitness]

Is your poller running? Once your poller finds it down, then it'll go down in Monitor. Avoid the and and or Availability settings. If it's always SNMP, just use SNMP Uptime.

[cargreg]

by changing the ping type as suggested, the status changes but the down email does not arrive

[TheWitness]

Check your mail logs to see if it was even sent. What Cacti, Thold and Monitor versions are you using?

[cargreg]

HI,
I checked the email logs, cacti sends the email, sorry.
Our problem is that we have a false positive problem.
We made a new installation of cacti version 1.2.23 and to overcome the problem of the device being unreachable we set the device detection to Ping or snmp uptime and it worked.
After the upgrade to the version 1.2.25, the problem recurred and as you suggested we changed the device detection to ping only,

but we have a lot of false negative
2023-10-16 19:30:05 - SPINE: Poller[1] PID[2672370] PT[139747594856000] Device[722] Hostname[*.*.*.*] ERROR: HOST EVENT: Device is DOWN Message: ICMP: Ping timed out

but it's not true because the host is reachable
[root@Cacti-rocky9 ~]# ping *.*.*.*
PING *.*.*.* (*.*.*.*) 56(84) bytes of data.
64 bytes from *.*.*.*: icmp_seq=1 ttl=57 time=11.3 ms
64 bytes from *.*.*.*: icmp_seq=2 ttl=57 time=5.21 ms
64 bytes from *.*.*.*: icmp_seq=3 ttl=57 time=1138 ms
64 bytes from *.*.*.*: icmp_seq=4 ttl=57 time=135 ms
64 bytes from *.*.*.*: icmp_seq=5 ttl=57 time=10.1 ms
64 bytes from *.*.*.*: icmp_seq=6 ttl=57 time=8.87 ms
64 bytes from *.*.*.*: icmp_seq=7 ttl=57 time=7.95 ms
64 bytes from *.*.*.*: icmp_seq=8 ttl=57 time=6.75 ms

and olso i have no holes in the graphs

the sistem is Linux Rocky Linux release 9.2 (Blue Onyx) 5.14.0-284.30.1.el9_2.x86_64,
selinux is disabled and those are the spine permission -rwsr-sr-x
Have you any suggestion?

[TheWitness]

If it's an SNMP host, use SNMP. If your point to point is lossy, try setting a higher timeout and retry.