Great. Other than that, we see escalation as the responsibility of a system
after Alertmanager in the chain. For example, PagerDuty (one of the
built-in notifier mechanisms in AM) allows setting up both a schedule and
an escalation policy for notifications.
On Fri, Jan 22, 2021 at 8:51 PM steve wa
I was looking for a way to escalate the alert after it was firing for a
period of time. Looks like defining the alert twice with different for and
label will do the trick.
Thanks for you assistance Julius.
On Friday, January 22, 2021 at 1:16:17 PM UTC-5 juliu...@promlabs.com wrote:
> If your
If your goal is to send an alert notification only after it has been firing
for some time, the usual way of doing that in Prometheus would be to use
the "for" duration field in the alerting rule to define how long the alert
has to be active before it transitions from "pending" into "firing" state
(
That makes sense which leads to a couple of other questions:
1. Can an alert rule annotation definition contain a promql query to
define?
2. if 1. is true, can the annotation label be used in alertmanager config
route matching?
Thanks,
Steve
On Thursday, January 21, 2021 at 12:04:52 PM UTC-5 S
On 21/01/2021 16:58, steve waldron wrote:
Is it possible to add a label with a dynamic value? for example - the
amount of time an alert has been firing? if so, any examples of
PROMQL queries to accomplish this?
I'm trying to determine if routes can be created using a label defined
by length
Is it possible to add a label with a dynamic value? for example - the
amount of time an alert has been firing? if so, any examples of PROMQL
queries to accomplish this?
I'm trying to determine if routes can be created using a label defined by
length of time an alert has been firing.
Thanks ,
6 matches
Mail list logo
