Cacti wrong downtime shown after device recovered

[Mbaye]

Dear All

I have installed cacti Version 0.8.8b to monitor my network et my customers link. Up to now everything is working fine but recently I have configured automatic Email alerte. I have noticed when one device went down and recovered, the downtime sent in the Email alert doesn't match with exactly the real downtime duration. The recovery alerte takes time to proceed as well.


Link XXXX down notification

System Error : XXXX is DOWN
Reason: Host did not respond to SNMP

Average system response : 1.65 ms
System availability: 99.8 %
Total Checks Since Clear: 6929
Total Failed Checks: 14
Last Date Checked DOWN : 2020-02-15 12:16:00 Host Previously UP for: 16h 40m 10s

Link XXXX recovered notification

System XXXX status: UP

Current ping response: 1.38 ms
Average system response : 1.72 ms
System availability: 99.8 %
Total Checks Since Clear: 6932
Total Failed Checks: 14
Last Date Checked UP: 2020-02-15 12:16:00 Host Previously DOWN for: 20m 4s

I need help to correct this bug.

Thank you in advance.

Best Regards.
Mbaye

[cigamit]

First of all, 0.8.8b is extremely old, so you will not receive much help for it.

The downtime displayed is how long it was down in Cacti, not the actual downtime, not much that can be done for that since Cacti doesn't know exactly when it went down or came back up. Cacti only knows that it was down when it tried to access it, and later it was back up when it tried to access it. You also have to note, that the "Host Recovery" setting in Settings also affects this.

If the recovery email is taking a long time, it could be because your doing 5 minute polling and you have a Host Recovery time set to 2 or higher.

[Mbaye]

Dear Cigamit

Many thanks for your fast feedback, I agree, 0.8.8b is too old, I plan to upgrade to last cacti version.

For the poller setting, you can find in attachment my current configuration. If you can suggest the best values to use.

Best Regards.
Mbaye