On Wed, 22 Aug 2001, Theo Van Dinter wrote:
> I've been trying to update to 0.99.1 and have become very confused by this
> statement. The first thing is that "alertafter" doesn't have a "summary"
> option, it's "alertevery XX summary".
ok, i the docs are inaccurate. i thought i had updated them properly
but i made some changes after i did and didn't re-check myself.
"alertevery" has a new option called "observe_detail". if this is set,
then the old behavior is invoked. the new default is the old "summary".
for example:
watch group1
service fping
interval 1m
...
period _ANYTIME_
alertevery 10m
alert test.alert
if there is a continued failure, this will call test.alert once every
10 minutes unless the summary line changes. it ignores if the detail
changes during successive failures.
watch group1
service fping
interval 1m
...
period _ANYTIME_
alertevery 10m observe_detail
alert test.alert
this is the same as above, but if the detail changes, it will call the
alert more than once every 10 minutes.
let me know how you make out. i'll update the docs when we are all sure
it works the way i described.
also carefully read the "ALERT DECISION LOGIC" of the manual. that mentions
observe_detail.
