-
Notifications
You must be signed in to change notification settings - Fork 1.9k
Add the "Examining cluster metrics" section #16752
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
Conversation
Several fixes and improvements Add a note on running demanding queries
c74d125 to
537db2d
Compare
|
|
||
| The *Metrics* page is located in *Monitoring* -> *Metrics* of the {product-title} web console. | ||
|
|
||
| image::../images/metrics-screen.png[] |
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.
Does it make sense to disambiguate this filename? Some (but not all) of the images, such as for ossm and olm, use a prefix for the filename.
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.
Good idea, implemented across all monitoring images.
| . The metrics table for a query. | ||
| . Color assigned to the graph of the metric. Clicking the square shows or hides the metric's graph. | ||
|
|
||
| Additionally, there is a link to the old Prometheus interface next to the title of the page. |
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.
I'd say: You can access the old Prometheus interface by clicking the Prometheus UI link at the top of the page.
Would someone know it's the old interface? Or is it just an alternate interface?
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.
The perspective of this whole section is "what do we have in this interface", that's why for all items I just call them out and say what they are for. You suggestion is valid for a procedure module.
It's not the best word, true. We are using old to imply that people should move away from that and into the new interface developed by the OCP team. "Alternate" interface doesn't have that implication. As to whether someone would know: maybe not, and that's why we'd better tell them.
|
|
||
| This section shows and explains the contents of the Metrics UI, a Web interface to Prometheus. | ||
|
|
||
| The *Metrics* page is located in *Monitoring* -> *Metrics* of the {product-title} web console. |
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.
I haven't written any UI docs, so I don't know what we're calling these things.
Is it the Monitoring -> Metrics section? Pane? Panel? Page?
The Metrics page is located in Monitoring -> Metrics section of the {product-title} web console.
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.
A valid suggestion from the POV of English. However, I just went through all occurrences of this and specifying this everywhere would add too much noise to the text. When a user reads this, they'll know to just click something saying Monitoring and then something saying Metrics and they won't care what it's called.
Besides, Monitoring -> Metrics here is really a navigation instruction that leads to the Metrics page.
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.
Read your later comment and reconsidered. Implemented this across the whole of monitoring docs. Thanks!
| [id="exploring-the-visualized-metrics_{context}"] | ||
| = Exploring the visualized metrics | ||
|
|
||
| After running the queries, the metrics are displayed on the interactive plot. The X axis of the plot represents time. The Y axis represents the metrics values. Each metric is shown as a colored graph. You can manipulate the plot and explore the metrics. |
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.
I'd probably italicize X and Y, as first usage. From contributing guidelines: https://github.com/openshift/openshift-docs/blob/enterprise-4.1/contributing_to_docs/doc_guidelines.adoc#quick-markup-reference
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's not a term though. It just refers to something on the plot and says what it represents.
And it's also one of the many things I'm describing here, and not particularly important in any way. So I don't see a need to emphasize this. Besides, by that logic, I'd want to italicize "the interactive plot" and "colored graph". I think all that is unnecessary.
It's also used only once, so it's first and only usage.
| .Procedure | ||
|
|
||
| . Initially, all metrics from all enabled queries are shown on the plot. You can select which metrics are shown. | ||
| * To hide all metrics from a query, click the action button of the query and click "Hide all series". |
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.
Based on our guidelines, I'd bold button names, such as action button.
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.
You are right, implemented everywhere.
| * Use the menu in the left upper corner to select the time range. | ||
| -- | ||
| + | ||
| To reset the time range, click "Reset Zoom". |
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.
s/"Reset Zoom"/Reset Zoom/
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.
Implemented.
| -- | ||
| + | ||
| To reset the time range, click "Reset Zoom". | ||
| . To see outputs of all queries at a specific point in time, hold the mouse cursor on the plot at that point. Query outputs will appear in a popup box. |
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.
s/see/display/
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.
That's much better, implemented.
| -- | ||
| + | ||
| To reset the time range, click "Reset Zoom". | ||
| . To see outputs of all queries at a specific point in time, hold the mouse cursor on the plot at that point. Query outputs will appear in a popup box. |
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.
s/Query/The query/
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.
Implemented.
| -- | ||
| + | ||
| To reset the time range, click "Reset Zoom". | ||
| . To see outputs of all queries at a specific point in time, hold the mouse cursor on the plot at that point. Query outputs will appear in a popup box. |
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.
I wonder about box vs window, but I don't think we have any codified style for referring to the UI elements in the Web UI yet.
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.
Calling that small popup rectange a window would be confusing. It think this will be commonly understood much better as a "popup box".
| + | ||
| To reset the time range, click "Reset Zoom". | ||
| . To see outputs of all queries at a specific point in time, hold the mouse cursor on the plot at that point. Query outputs will appear in a popup box. | ||
| . For more detailed information about metrics of a specific query, expand the table of that query. Every metric is shown with its current value. |
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.
How do I expand the query?
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.
Specified: "using the dropdown button".
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.
I asked around, and w're using drop-down instead of dropdown. No one's heard of a drop-down button before though.
| To reset the time range, click "Reset Zoom". | ||
| . To see outputs of all queries at a specific point in time, hold the mouse cursor on the plot at that point. Query outputs will appear in a popup box. | ||
| . For more detailed information about metrics of a specific query, expand the table of that query. Every metric is shown with its current value. | ||
| . To hide the plot, use the "Hide Graph" button. |
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.
s/use the "Hide Graph" button./click Hide Graph./
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.
Implemented.
|
|
||
| .Procedure | ||
|
|
||
| . Open the OpenShift Container Platform web console and navigate to *Monitoring* -> *Metrics*. |
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.
{product-title}
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.
Implemented here and in another place.
|
@openshift/team-documentation So having not done much with UI, I wonder about several things, but particularly when navigating: navigate to Monitoring -> Metrics. Do you navigate to parts of the UI? And what is that part? Is it navigate to the Monitoring > Metrics page/section/pane/panel/area? Or is to Monitoring > Metrics enough without any qualification of what exactly it is? It looks like we have this in our contributor guidelines:
|
|
|
||
| toc::[] | ||
|
|
||
| {product-title} 4 provides a Web interface to Prometheus, which enables you to run Prometheus Query Language (PromQL) queries and examine the metrics visualized on a plot. This provides an extensive overview of the cluster state and enables you to troubleshoot problems. |
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.
Does it make sense to use {product-version} here?
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.
Yep, implemented.
|
|
||
| toc::[] | ||
|
|
||
| {product-title} 4 provides a Web interface to Prometheus, which enables you to run Prometheus Query Language (PromQL) queries and examine the metrics visualized on a plot. This provides an extensive overview of the cluster state and enables you to troubleshoot problems. |
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.
This what? The plot? PromQL? Prometheus?
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.
Specified ("This functionality").
|
|
||
| .Next steps | ||
|
|
||
| xref:../../monitoring/cluster-monitoring/prometheus-alertmanager-and-grafana.adoc#prometheus-alertmanager-and-grafana[Access the Prometheus, Alertmanager, and Grafana.] |
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.
Access the Prometheus, Alertmanager, and Grafana. interfaces? pages? Besides being grouped together, I'm not sure what the relationship is between these?
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.
I was asked (twice) to leave "interfaces" out of this wording due to some technicality (frankly I don't remember what technicality).
These three things is why have the monitoring stack at all. They are how you actually use the stack once you've configured it.
|
|
||
| toc::[] | ||
|
|
||
| {product-title} 4 provides a Web interface to Prometheus, which enables you to run Prometheus Query Language (PromQL) queries and examine the metrics visualized on a plot. This provides an extensive overview of the cluster state and enables you to troubleshoot problems. |
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.
Per IBM:
web adj, n
Write as shown - lowercase “w.”
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.
Implemented everywhere.
| + | ||
| [NOTE] | ||
| ==== | ||
| Queries that operate on large amounts of data might timeout or overload the browser when drawing timeseries graphs. To avoid this, you might prefer to hide the graph and calibrate your query using only the metrics table. Then, after finding a feasible query, enable the plot to draw the graphs. |
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.
s/you might prefer to hide/hide/
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.
Implemented.
| . Open the OpenShift Container Platform web console and navigate to *Monitoring* -> *Metrics*. | ||
|
|
||
| . In the query field, enter your PromQL query. | ||
| * To show all available metrics and PromQL functions, use the "Insert Metric at Cursor" button. |
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.
s/use the "Insert Metric at Cursor" button/click Insert Metric at Cursor/
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.
Implemented.
|
|
||
| . In the query field, enter your PromQL query. | ||
| * To show all available metrics and PromQL functions, use the "Insert Metric at Cursor" button. | ||
| . For multiple queries, use the "Add Query" button. |
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.
s/use the "Add Query" button./click Add Query./
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.
Implemented.
| . In the query field, enter your PromQL query. | ||
| * To show all available metrics and PromQL functions, use the "Insert Metric at Cursor" button. | ||
| . For multiple queries, use the "Add Query" button. | ||
| . For deleting queries, click the action button for the query and select "Delete query". |
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.
Is 12 earlier the action button? We had some discussion about how to describe the ... button.
We actually have an attribute for it:
:kebab: image:kebab.png[title="Options menu"]
So you could replace instances of these with:
click {kebab} for the query, and then select Delete query.
s/"Delete query"/Delete query/
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.
Nice! I never liked my wording. Implemented for both occurrences.
| * To show all available metrics and PromQL functions, use the "Insert Metric at Cursor" button. | ||
| . For multiple queries, use the "Add Query" button. | ||
| . For deleting queries, click the action button for the query and select "Delete query". | ||
| . For keeping but not running a query, click the "Disable query" button. |
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.
s/the "Disable query" button/Disable query/
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.
Implemented.
| . For multiple queries, use the "Add Query" button. | ||
| . For deleting queries, click the action button for the query and select "Delete query". | ||
| . For keeping but not running a query, click the "Disable query" button. | ||
| . Once you finish creating queries, click the "Run Queries" button. The metrics from the queries are visualized on the plot. If a query is invalid, the UI shows an error message. |
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.
s/click the "Run Queries" button/click Run Queries/
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.
Implemented.
|
@rh-max, I completed the review. |
|
The preview will be available shortly at: |
Implemented everywhere in monitoring docs. |
Thanks, very thorough! I've implemented much of the feedback. If you are fine with the current state, please merge this and cherrypick to enterprise-4.2. Thank you. |
|
/cherrypick enterprise-4.2 |
|
@jboxman: new pull request created: #16778 DetailsIn response to this:
Instructions for interacting with me using PR comments are available here. If you have questions or suggestions related to my behavior, please file an issue against the kubernetes/test-infra repository. |
|
|
||
| . Actions. | ||
| * Add query. | ||
| * Expand all query tables. |
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 one result table is collapse, the menu is "Collapse all query tables", if all tables are collapsed, the menu is "Expand all query tables"
I think
Expand/Collapse all query tables(Based on the result table is collapsed or expanded) is better.
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.
I changed to "Expand or collapse", but didn't use "(Based on the result table is collapsed or expanded)", it is too long and unnecessary.
Implemented in #16978.
| . Clear query. | ||
| . Disable query. | ||
| . Actions for a specific query. | ||
| * Disable query. |
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 the query is disabled, the menu is "Enable query",
if the query is enabled, the menu is "Disable query"
I think
Enable /Disable query(Based on the query is disabled or enabled ) is better.
It is the same for "Hide all series"
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.
I changed to "Enable or disable", but didn't use "(Based on the query is disabled or enabled )", it is too long and unnecessary.
Implemented in #16978.
| -- | ||
| + | ||
| To reset the time range, click *Reset Zoom*. | ||
| . To display outputs of all queries at a specific point in time, hold the mouse cursor on the plot at that point. The query outputs will appear in a pop-up box. |
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.
hold the mouse cursor on the plot at that point
what about changing to "move the mouse cursor over the line"?
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's enough to just be on the right point in the X axis (time), you can be anywhere in the Y axis (metric value). It's good if the users know that, that's why I used the wording "on the plot at that point [in time]". You don't need to put cursor directly over the graph ("the line").
Not implemented.
| = Accessing Prometheus, Alerting UI, and Grafana using the web console | ||
|
|
||
| You can access Prometheus, Alerting UI, and Grafana web UIs using a Web browser through the {product-title} Web console. | ||
| You can access Prometheus, Alerting UI, and Grafana web UIs using a web browser through the {product-title} web console. |
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.
Prometheus, Alerting UI, and Grafana web UIs
=>
Prometheus, Alerting, and Grafana web UIs
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.
Implemented in #16978.
| . In the {product-title} web console, navigate to the *Operators* -> *OperatorHub* page and install the Prometheus Operator in the namespace where your application is. | ||
|
|
||
| . Navigate to *Catalog* -> *Developer Catalog* and install Prometheus, Alertmanager, Prometheus Rule, and Service Monitor in the same namespace. | ||
| . Navigate to the *Catalog* -> *Developer Catalog* page and install Prometheus, Alertmanager, Prometheus Rule, and Service Monitor in the same namespace. |
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.
Navigate to the Catalog -> Developer Catalog page
It is better change to
Navigate to the Operators -> Installed Operators page
reason: there is not link in the left navigation area links to /catalog page
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.
Implemented in #16978.
| The main three pages of the Alerting UI are the *Alerts*, the *Silences*, and the *YAML* pages. | ||
|
|
||
| The *Alerts* page is located in *Monitoring* -> *Alerts* of the {product-title} web console. | ||
| The *Alerts* page is accessible by clicking *Monitoring* -> *Alerts* in the {product-title} web console. |
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.
The Alerts page is accessible by clicking Monitoring -> Alerting -> Alerts
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.
Implemented in #16978.
| . Actions you can do with the alert. | ||
|
|
||
| The *Silences* page is located in *Monitoring* -> *Silences* of the {product-title} web console. | ||
| The *Silences* page is accessible by clicking *Monitoring* -> *Silences* in the {product-title} web console. |
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.
The Silences page is accessible by clicking Monitoring -> Alerting -> Silences
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.
Implemented in #16978.
| . Open the {product-title} web console and navigate to *Monitoring* -> *Alerts*. | ||
| . Open the {product-title} web console and navigate to the *Monitoring* -> *Alerting* -> *Alerts* page. | ||
|
|
||
| . Optional: Filter the alerts by name using the *Filter alerts by name* field. |
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.
Filter alerts by name
=>
Filter Alerts by name
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.
Implemented in #16978.
@jboxman Let's use this PR (opened against master). Feel free to ignore the original #16495 . Upon merging, it is to be cherrypicked to enterprise-4.2.
Once you finish the review, please indicate it somehow (say, with a comment).
To address your comments in the original PR: