The point is that service_id have regexp matchers, so it will
match all data instead of null.
On Thu, Apr 15, 2021 at 12:23:24 PM +0800, chuanjia xing wrote:
Ping for this. Anyone can give some pointers for this? Thanks.
On Tuesday, April 13, 2021 at 4:41:55 PM UTC-7 chuanjia xing
wrote:
Hi there,
I hit following issue when using Prometheus + Grafana,
and want to
check if this is a bug in Prometheus query.
Scenario:
I am using Prometheus to collect CPUUtilization for aws
instances. In
one dashboard, I want to show the metrics for services which
are
underutilized. To do this, I first use Grafana Variable
"underutilized_services" to find out the service names; then I
use this
Variable in my Prometheus query to show the actual
CPUUtilization data for
these services.
I am using following Prometheus query:
"*avg by
(service_id)($Metric{service_id=~"$underutilized_services"})*"
and $underutilized_services" is a list of service names, like
service1,
service2...
It is working well when the $underutilized_services is not
empty,
like folllwing (first graph)
But if the $underutilized_services is empty, then my
understanding is
that it should now "No Data", but somehow it is showing
something wired,
like following (second graph).
If you look at the legend of the second graph, it shows:
"*avg by
(service_id)(aws_ec2_cpuutilization_p50{service_id=~"()"})*"
It looks like above query actually returns some value,
which in my
opinion it shouldn't.
This is showing some confusing data for my
customer. Ideally it
should show "No Data" in the second case.
Is there a bug in Prometheus? Or I am not using the
query correctly?
--
You received this message because you are subscribed to the Google Groups
"Prometheus Users" group.
To unsubscribe from this group and stop receiving emails from it, send an email
to [email protected].
To view this discussion on the web visit
https://groups.google.com/d/msgid/prometheus-users/60794399.1c69fb81.8f79e.51caSMTPIN_ADDED_BROKEN%40gmr-mx.google.com.
