-
Notifications
You must be signed in to change notification settings - Fork 444
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
[Kubernetes][Dashboard] Fix proxy, controller manager and scheduler dashboards to support TSDB #5992
[Kubernetes][Dashboard] Fix proxy, controller manager and scheduler dashboards to support TSDB #5992
Conversation
Signed-off-by: constanca-m <constanca.manteigas@elastic.co>
Signed-off-by: constanca-m <constanca.manteigas@elastic.co>
Signed-off-by: constanca-m <constanca.manteigas@elastic.co>
Signed-off-by: constanca-m <constanca.manteigas@elastic.co>
Signed-off-by: constanca-m <constanca.manteigas@elastic.co>
🌐 Coverage report
|
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
No problems from my end!
packages/kubernetes/kibana/dashboard/kubernetes-bf9389f0-0c14-11ed-b760-5d1bccb47f56.json
Outdated
Show resolved
Hide resolved
I would suggest to change the labels of "Average Request" to "Max Average Request" . In general you can add the |
…11ed-b760-5d1bccb47f56.json Co-authored-by: Andrew Gizas <andreas.gkizas@elastic.co>
I think that might be confusing. Since it is a counter this |
@constanca-m I would say the opposite. The last value can be diffrent than the max and can lead to totally diffrent results in the calculation. For eg. last value of cpu can be something low but the max could have been a spike for the given period. So users should understand what they see |
I am not sure I understand. Since |
Ah indeed is a counter, was not paying attention. What happens if the counter resets to zero for example? |
That will be a problem @gizas, but I don't see a workaround for that right now. |
Signed-off-by: constanca-m <constanca.manteigas@elastic.co>
…fix-kubernetes-dashboards
@@ -220,7 +220,7 @@ | |||
"suffix": "s" | |||
} | |||
}, | |||
"formula": "last_value(kubernetes.proxy.sync.networkprogramming.duration.us.sum)/(pick_max(last_value(kubernetes.proxy.sync.networkprogramming.duration.us.count),1))/1000000", | |||
"formula": "max(kubernetes.proxy.sync.networkprogramming.duration.us.sum)/(pick_max(max(kubernetes.proxy.sync.networkprogramming.duration.us.count),1))/1000000", |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
if last_value
is already supported, according to elastic/kibana#156168, why it was changed here last_value
-> max
?
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
it is not critical because of this explanation #5992 (comment), trying to understand if there is any other reason
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
When I pushed the changes, last value was still not available @tetianakravchenko, but that should be the correct aggregation. I will check that now.
Signed-off-by: Tetiana Kravchenko <tetiana.kravchenko@elastic.co>
Package kubernetes - 1.37.0 containing this change is available at https://epr.elastic.co/search?package=kubernetes |
What does this PR do?
In proxy, controller manager and scheduler dashboards, there are some visualizations that use
counter
fields with now unsupported formulas -sum
andlast_value
- when TSDB is enabled. This PR fixes those issues so we can have similar visualizations for the same metrics.Checklist
changelog.yml
file.How to test this PR locally
elastic-package build
inside Kubernetes package.elastic-package stack up -v -d --version=8.8.0-SNAPSHOT
.Related issues
Screenshots
These screenshots only include the before/after of the broken visualizations. Others that used a
counter
withlast_value
were switched tomax
, but there was no change in the visualization. This is needed becauselast_value
is also no longer supported forcounter
fields once TSDB is enabled.Proxy dashboard
Before:
After:
Note: "No results found" is shown because there are no documents for the applied filter in these two visualizations. There is nothing wrong with them.
Controller manager dashboard
Before:
After:
Note: Since we can no longer use
sum
we have to breakdown the metrics by the labels. This will likely lead to very confusing visualizations but as of now, there is no workaround.Scheduler dashboard
Before:
After: