You can use "within" option with "trigger". For example if your rpm executed once in 1 min , so you need to check if rpm was succesful 15 times
We configured this some time ago
services {
rpm {
probe script {
test src1 {
probe-type icmp-ping;
target address 10.7.0.1;
probe-count 1;
probe-interval 1;
test-interval 1;
source-address 10.7.0.2;
}
}
event-options {
policy ping-fail-src1 {
events ping_probe_failed;
within 32 {
trigger on 15;
}
attributes-match {
ping_probe_failed.test-owner matches .*script.*;
ping_probe_failed.test-name matches .*src1.*;
}
then {
event-script change-nexthop.slax {
arguments {
nexthop-address 10.7.0.1;
metric 20;
}
}
}
}
policy ping-success-src1 {
events ping_test_completed;
within 31 {
trigger on 15;
events ping_test_completed;
}
attributes-match {
ping_test_completed.test-owner matches .*script.*;
ping_test_completed.test-name matches .*src1.*;
}
then {
event-script change-nexthop.slax {
arguments {
nexthop-address 10.7.0.1;
metric 10;
}
}
}
}
}