You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Let's say we have a bp like this: Worst -> BP A -> BP B
BP A sees these events (only hard criticals are take into account):
4 criticals
[2024-11-15 03:12:54] SERVICE ALERT: B;B;OK;HARD;3;OK
[2024-11-15 03:12:03] SERVICE ALERT: B;B;CRITICAL;HARD;3
[2024-11-15 03:10:39] SERVICE ALERT: B;B;OK;HARD;3
[2024-11-15 03:10:03] SERVICE ALERT: B;B;CRITICAL;HARD;3
[2024-11-15 03:08:07] SERVICE ALERT: B;B;OK;HARD;3
[2024-11-15 03:07:03] SERVICE ALERT: B;B;CRITICAL;HARD;3
[2024-11-15 02:10:03] SERVICE ALERT: B;B;OK;HARD;3
[2024-11-15 02:07:03] SERVICE ALERT: B;B;CRITICAL;HARD;3
So are outages for B:
Start End Duration Type Plugin Output
2024-11-15 03:12:03 2024-11-15 03:12:54 0d 0h 0m 51s SERVICE CRITICAL (HARD) ...
2024-11-15 03:10:03 2024-11-15 03:10:39 0d 0h 0m 36s SERVICE CRITICAL (HARD) ...
2024-11-15 03:07:03 2024-11-15 03:08:07 0d 0h 1m 4s SERVICE CRITICAL (HARD) ...
2024-11-15 02:07:03 2024-11-15 02:09:03 0d 0h 2m 0s SERVICE CRITICAL (HARD) ...
sum = 4 mn 31s
Now for A:
3 criticals
[2024-11-15 03:10:13] SERVICE ALERT: A;A;OK;HARD;3
[2024-11-15 03:10:03] SERVICE ALERT: A;A;CRITICAL;HARD;3
[2024-11-15 03:08:20] SERVICE ALERT: A;A;OK;HARD;3
[2024-11-15 03:08:03] SERVICE ALERT: A;A;CRITICAL;HARD;3
[2024-11-15 03:06:17] SERVICE ALERT: A;A;OK;HARD;3
[2024-11-15 02:11:03] SERVICE ALERT: A;A;OK;HARD;3
[2024-11-15 02:08:03] SERVICE ALERT: A;A;CRITICAL;HARD;3
Si are outages for A
Start End Duration Type Plugin Output
2024-11-15 03:10:03 2024-11-15 03:10:13 0d 0h 0m 10s SERVICE CRITICAL (HARD) ...
2024-11-15 03:08:03 2024-11-15 03:08:20 0d 0h 0m 17s SERVICE CRITICAL (HARD) ...
2024-11-15 02:08:03 2024-11-15 02:10:03 0d 0h 2m 0s SERVICE CRITICAL (HARD) ...
sum = 2 mn 27s
Big difference !
I think it's inherent in the way we calculate so there's a loss of precision from bottom to top of the bp.
If I understand correctly, bp statuses are computed from command thruk bp all in crontab and cron is just 1 mn accurate.
I wonder if it's possible to shorten the time between two calculations?
jfr
The text was updated successfully, but these errors were encountered:
More of a thought than a bug.
Let's say we have a bp like this: Worst -> BP A -> BP B
BP A sees these events (only hard criticals are take into account):
4 criticals
[2024-11-15 03:12:54] SERVICE ALERT: B;B;OK;HARD;3;OK
[2024-11-15 03:12:03] SERVICE ALERT: B;B;CRITICAL;HARD;3
[2024-11-15 03:10:39] SERVICE ALERT: B;B;OK;HARD;3
[2024-11-15 03:10:03] SERVICE ALERT: B;B;CRITICAL;HARD;3
[2024-11-15 03:08:07] SERVICE ALERT: B;B;OK;HARD;3
[2024-11-15 03:07:03] SERVICE ALERT: B;B;CRITICAL;HARD;3
[2024-11-15 02:10:03] SERVICE ALERT: B;B;OK;HARD;3
[2024-11-15 02:07:03] SERVICE ALERT: B;B;CRITICAL;HARD;3
So are outages for B:
Start End Duration Type Plugin Output
2024-11-15 03:12:03 2024-11-15 03:12:54 0d 0h 0m 51s SERVICE CRITICAL (HARD) ...
2024-11-15 03:10:03 2024-11-15 03:10:39 0d 0h 0m 36s SERVICE CRITICAL (HARD) ...
2024-11-15 03:07:03 2024-11-15 03:08:07 0d 0h 1m 4s SERVICE CRITICAL (HARD) ...
2024-11-15 02:07:03 2024-11-15 02:09:03 0d 0h 2m 0s SERVICE CRITICAL (HARD) ...
sum = 4 mn 31s
Now for A:
3 criticals
[2024-11-15 03:10:13] SERVICE ALERT: A;A;OK;HARD;3
[2024-11-15 03:10:03] SERVICE ALERT: A;A;CRITICAL;HARD;3
[2024-11-15 03:08:20] SERVICE ALERT: A;A;OK;HARD;3
[2024-11-15 03:08:03] SERVICE ALERT: A;A;CRITICAL;HARD;3
[2024-11-15 03:06:17] SERVICE ALERT: A;A;OK;HARD;3
[2024-11-15 02:11:03] SERVICE ALERT: A;A;OK;HARD;3
[2024-11-15 02:08:03] SERVICE ALERT: A;A;CRITICAL;HARD;3
Si are outages for A
Start End Duration Type Plugin Output
2024-11-15 03:10:03 2024-11-15 03:10:13 0d 0h 0m 10s SERVICE CRITICAL (HARD) ...
2024-11-15 03:08:03 2024-11-15 03:08:20 0d 0h 0m 17s SERVICE CRITICAL (HARD) ...
2024-11-15 02:08:03 2024-11-15 02:10:03 0d 0h 2m 0s SERVICE CRITICAL (HARD) ...
sum = 2 mn 27s
Big difference !
I think it's inherent in the way we calculate so there's a loss of precision from bottom to top of the bp.
If I understand correctly, bp statuses are computed from command thruk bp all in crontab and cron is just 1 mn accurate.
I wonder if it's possible to shorten the time between two calculations?
jfr
The text was updated successfully, but these errors were encountered: