Hi all, I found misbehavior of prioritized triggered alerts. It is useful that rising high priority alert masks low priority ones. But later, then high-prio alert releases, node status becomes undefined since low-prio alerts that can be still active are never exposed.
My proposition is as follow: then high-prio alert is clearing, check
for nearest active low-prio alert and expose It.
I realized a prototype of that algorithm, It does not reports clearing
alerts and belives that some alert with lowest priority is always
active. It produce the messages as follow:
time1: alert no-loss (init).
time2: alert medium-loss (rising).
time3: alert high-loss (rising).
time4: alert medium-loss (resuming).
time5: alert no-loss (resuming).
The code is attached as path to Smokeping.pm of version 2.4.2.
I hope that the current misbehavior will be corrected with the above
manner or something like this.
It would nice to hear the comments.
--
Kind regards,
Veniamin
smokeping-2.4.2.diff
Description: Binary data
_______________________________________________ smokeping-users mailing list [email protected] https://lists.oetiker.ch/cgi-bin/listinfo/smokeping-users
