DOC-2653 Change Impact Analysis Quickstart - Harness feature flag as change source. #788
Conversation
…d change impact analisys quick start
|
@shankarh7924 , I have made the following changes:
|
|
Preview environment: https://ci.pr.harness.io/pr-788 |
|
Preview environment: https://hdh.pr.harness.io/pr-788 |
SudheendraKatte
added
Content - enhancement
|
Please review and provide comments. Here are the major changes:
|
| --- | ||
|
|
||
|
|
||
| You can add Harness Feature Flag as a change source. This helps you monitor the impact of feature flag on service health. The following steps explain how to add feature flag as a change source. |
There was a problem hiding this comment.
Feature flag is by default added as a change source since it is an internal change source.
What need to explain is how changes of a "given" feature flag can be seen in SRM against the service health.
There was a problem hiding this comment.
Thanks @deepak-harness . I have updated the topic. Please review and let me know.
SudheendraKatte
changed the title
|
Preview environment: https://hdh.pr.harness.io/pr-788 |
| --- | ||
|
|
||
|
|
||
| The Changes dashboard provides a quick overview of change events from multiple change sources. A change source is a combination of a Harness Service and associated environment that monitors changes such as deployments, infrastructure changes, and incidents. Change events are grouped in one location, enabling you to see how a certain change affects the rest of the system. |
There was a problem hiding this comment.
"change source is a combination"
Change source is a system which generate the change events when ever it records any change.
There was a problem hiding this comment.
Replaced with the following:
A change source is a system that generates change events whenever a change is recorded.
|
|
||
| The Changes dashboard provides a quick overview of change events from multiple change sources. A change source is a combination of a Harness Service and associated environment that monitors changes such as deployments, infrastructure changes, and incidents. Change events are grouped in one location, enabling you to see how a certain change affects the rest of the system. | ||
|
|
||
| The Changes dashboard helps you understand when the change occurred, why it took place, and what kind of effect it had. |
There was a problem hiding this comment.
what kind of effect it had on the service health of the monitored service.
|
|
||
| These changes might include deployments, infrastructure changes, incidents, and feature flags. Harness Change Impact Analysis helps you: | ||
|
|
||
| - Correlate recent changes in your service and their broader impact on your service. |
|
|
||
| To create a new health source, do the following: | ||
|
|
||
| 1. Select **+ Add New Health Source**. |
There was a problem hiding this comment.
It should be seperate section. Just say add a new health source.
Now, Health source section(module) can talk about how to configure each type of health source
There was a problem hiding this comment.
same in health source. just say add a connection and divert the user to add connector which can have seperate doc for each type of connector.
There was a problem hiding this comment.
Noted. As discussed, these issues will be addressed while restructuring the documentation.
|
Hi @deepak-harness , I have incorporated the second set of comments. Please review and approve. |
|
Preview environment: https://hdh.pr.harness.io/pr-788 |
|
|
||
| The monitored service that you selected appears in the **Services** section. | ||
|
|
||
| Whenever the feature flag is turned off or on, a change event is recorded in the Health Source page of the monitored service. |
There was a problem hiding this comment.
Change event is recorded on health score page not health source page of monitored service
There was a problem hiding this comment.
Updated as:
Whenever the feature flag is turned off or on, a change event is recorded in the Service Health page of the monitored service.
|
Preview environment: https://hdh.pr.harness.io/pr-788 |
| --- | ||
| title: Changes dashboard | ||
| description: Get a quick overview of all the change events. | ||
| sidebar_position: 30 |
There was a problem hiding this comment.
Overview of all the change events.
There was a problem hiding this comment.
Changed to "Get an overview of all change events"
| The Changes dashboard helps you understand when the change occurred, why it took place, and what impact it had on the health of the monitored service. | ||
|
|
||
| Perform the following step to go to the Changes dashboard: | ||
|
|
There was a problem hiding this comment.
Outlined below is the step to navigate to the Changes dashboard.
There was a problem hiding this comment.
I would prefer the existing text :)
|
|
||
|
|
||
| #### Changes Timeline | ||
|
|
|
|
||
| Select a change event to see its details. | ||
| The change event details screen appears. It has the following details: | ||
|
|
There was a problem hiding this comment.
It has the following details.
There was a problem hiding this comment.
Changed to "It displays the following details"
|  | ||
|
|
||
| To view the change details for a specific time slot, select the time slot on the Changes Timeline. A time window appears with the details of all the changes that occurred in that specific time slot. You can increase or decrease the time slot using the handles. | ||
|
|
There was a problem hiding this comment.
You can modify the time slot using the handles.
There was a problem hiding this comment.
Good one! Re writing it as "You can change the time slot using the handles"
| #### Time Window | ||
|
|
||
| The Overall Health Score timeline has a **Time Window** which displays granular details such as health score, incident count, metrics, and logs for a specific timeline on the graph. You can move the Time Window along the graph to choose a specific timeline. You can increase or decrease the timeline using the handles. | ||
|
|
There was a problem hiding this comment.
You can modify the timeline using the handles.
There was a problem hiding this comment.
Updated text:
You can change the time period by using the handles.
| ### Changes list | ||
|
|
||
| The **Changes** list provides a detailed breakdown of all the changes occurred in deployment, infrastructure, and feature flags during the selected time period. It also displays the incidents from the incident management tool during the selected time period. You can see the following information in the **Chages** list: | ||
|
|
There was a problem hiding this comment.
The Changes list provides a detailed breakdown of all the changes that occurred in deployment, infrastructure, and feature flags during the selected time period. It also displays the incidents from the incident management tool during the selected time period. You can see the following information in the Changes list:
|
|
||
| - **All Metrics**: Displays all the metrics captured by the health sources. | ||
| - **Anomalous Metrics**: Displays the anomalous metrics captured by the health sources. | ||
|
|
There was a problem hiding this comment.
All metrics
Anomalous metrics
There was a problem hiding this comment.
They are UI elements.
| ### Metrics, logs, and errors | ||
|
|
||
| The **Metrics**, **Logs**, and **Errors** tabs displays the metrics, log, and error data collected at regular intervals. | ||
|
|
There was a problem hiding this comment.
The Metrics, Logs, and Errors tabs display the metrics, logs, and error data collected at regular intervals, respectively.
| ### Service dependencies | ||
|
|
||
| The **Service Dependency** section displays a graphical representation of the dependencies between applications, infrastructure, and data. | ||
|
|
There was a problem hiding this comment.
The Service Dependency section depicts the dependencies between applications, infrastructure, and data graphically.
There was a problem hiding this comment.
Updated text:
The Service Dependence section graphically displays the dependencies between applications, infrastructure, and data.
There was a problem hiding this comment.
Hi, Sudhi. I know much of the content in this PR is existing content that you had to bring into line somewhat. However, check if you can incorporate these comments. I've used suggestions as much as possible to make your task easier. Let me know how it goes. We should meet to discuss the format of the quickstart, though.
I did not reach the end of the PR in this round, but I'll take another look tomorrow. Thanks.
I'm hoping Kim or someone else familiar with FF can look at the FF part of this PR.
| @@ -57,9 +57,9 @@ The following quickstart guides are available for Feature Flags: | |||
|
|
|||
| The following quickstart guides are available for Service Reliability Management: | |||
|
|
|||
| * [Change Impact Analysis Quickstart](../service-reliability-management/howtos-service-reliability-management/change-impact-analysis-quickstart.md) describes how you can correlate change events and understand their impact on a Monitored Service. | |||
| * [Change Impact Analysis Quickstart](../service-reliability-management/howtos-service-reliability-management/change-impact-analysis/change-impact-analysis-quickstart.md) describes how you can correlate change events and understand their impact on a Monitored Service. | |||
There was a problem hiding this comment.
| * [Change Impact Analysis Quickstart](../service-reliability-management/howtos-service-reliability-management/change-impact-analysis/change-impact-analysis-quickstart.md) describes how you can correlate change events and understand their impact on a Monitored Service. | |
| * [Change Impact Analysis Quickstart](../service-reliability-management/howtos-service-reliability-management/change-impact-analysis/change-impact-analysis-quickstart.md) describes how you can correlate change events and understand their impact on a monitored service. |
(Since we seem to have accepted lowercase for this term--I see lowercase in other topics--we should apply that here as well. Thanks. Good to check globally, within this PR.)
| @@ -0,0 +1,35 @@ | |||
| --- | |||
| title: Connect Harness Feature Flag to a monitored service | |||
There was a problem hiding this comment.
Are we referring to the module here? If yes, change to:
| title: Connect Harness Feature Flag to a monitored service | |
| title: Connect Harness Feature Flags to a monitored service |
If it's the Harness module we are referring to, is "connect" the right choice of word? Please review the title with someone who is familiar with FF (Kim, for example). Thanks.
There was a problem hiding this comment.
We are not referring to the entire FF module here. It is just an FF within the module. In layman's terms, this topic describes how to tell a monitored service that it should display any changes that happen in a particular FF. So the term "connect' is valid in this context.
Also, it is not "Harness Feature Flags" but just a "Harness feature flag" as we are talking about a single/one ff that exists within the FF module. To avoid the confusion, changing the title to "Connect a feature flag to a monitored service".
| @@ -0,0 +1,35 @@ | |||
| --- | |||
| title: Connect Harness Feature Flag to a monitored service | |||
| description: Learn how to connect Harness Feature Flag to a monitored service. | |||
There was a problem hiding this comment.
Same as previous. Consider the previous comment as a global one.
There was a problem hiding this comment.
Changed to "...connect a feature flag..."
| --- | ||
|
|
||
|
|
||
| You can connect Harness Feature Flag to a monitored service. This helps you monitor the impact of feature flag status change on the service health. The following steps explain how to connect a feature flag to a monitored service. |
There was a problem hiding this comment.
The first sentence is not required unless it gives the user more information than the title does. Can we change to:
| You can connect Harness Feature Flag to a monitored service. This helps you monitor the impact of feature flag status change on the service health. The following steps explain how to connect a feature flag to a monitored service. | |
| Connecting Harness Feature Flags to a monitored service helps you monitor the impact of a change in the status of a feature flag on service health. The following steps explain how to connect a feature flag to a monitored service. |
(Also rewording "feature flag status change" to improve readability. Does the proposal read better?)
|
|
||
| You can connect Harness Feature Flag to a monitored service. This helps you monitor the impact of feature flag status change on the service health. The following steps explain how to connect a feature flag to a monitored service. | ||
|
|
||
| 1. In your Harness project, navigate to the **Feature Flag** module, and then select **Feature Flags**. |
There was a problem hiding this comment.
| 1. In your Harness project, navigate to the **Feature Flag** module, and then select **Feature Flags**. | |
| 1. In your Harness project, navigate to the **Feature Flags** module, and then select **Feature Flags**. |
|
|
||
| 1. In your Harness project, navigate to the **Service Reliability Management** module, and then select **Monitored Services**. | ||
| A list of all the monitored services appears. | ||
| 2. Choose the monitored service for which you want to see the health data. |
There was a problem hiding this comment.
| 2. Choose the monitored service for which you want to see the health data. | |
| 2. Select the monitored service for which you want to see health data. |
|
|
||
| ### Change summary | ||
|
|
||
| Change summary displays the number of changes in deployments, infrastructure, and incidents for a specific time period. You can set the time period based on the period of length for which you want to see the data. The available options are: |
There was a problem hiding this comment.
| Change summary displays the number of changes in deployments, infrastructure, and incidents for a specific time period. You can set the time period based on the period of length for which you want to see the data. The available options are: | |
| The change summary displays the number of changes in deployments, infrastructure, and incidents for a specific time period. You can specify the time period for which you want to see data. The available options are: |
Not sure what "You can set the time period based on the period of length for which you want to see the data" means but we would have to ask the original author. We could hazard a guess. Not sure whether "specify" is the right word but feel free to replace.
There was a problem hiding this comment.
"You can specify the time period for which you want to see the data" is accurate.
|
|
||
| ### Overall Health Score timeline | ||
|
|
||
| The overall health score timeline provides a graphical representation of the high level overview of the service health for the time period that you have selected. The graph is divided into time slots of equal length based on the time period setting. The graph also displays the changes in deployment, infrastructure, and feature flag. It also displays the incidents. This helps you correlate the data over a time period and analyze the impact of change on the reliability of the service. |
There was a problem hiding this comment.
| The overall health score timeline provides a graphical representation of the high level overview of the service health for the time period that you have selected. The graph is divided into time slots of equal length based on the time period setting. The graph also displays the changes in deployment, infrastructure, and feature flag. It also displays the incidents. This helps you correlate the data over a time period and analyze the impact of change on the reliability of the service. | |
| The overall health score timeline is a graphical overview of the health of a service over a specified time period. The graph is divided into time slots of equal length. The span of each division depends on the time period setting. The graph displays changes in deployment, infrastructure, and feature flag. It also displays incidents. This helps you correlate data over a time period and analyze the impact of change on the reliability of the service. |
- Not sure what "graphical representation of the high level overview", but I proposed a change. Let me know whether it works.
- Also not sure what "based on the time period setting" refers to, but I took a leap.
- Either "displays changes in the deployment, infrastructure, and feature flag" or "displays changes in deployments, infrastructures, and feature flags". Which one?
- What does the "this" that begins the last sentence refer to? The incidents in the previous sentence or the timeline/graph overall?
There was a problem hiding this comment.
Points taken. Here is the updated text:
The overall health score timeline is a graphical overview of the health of a service over a specified time period. The graph is divided into time slots of equal length. The span of each division depends on the time period setting. The graph displays change in deployments, infrastructures, and feature flags. It also displays incidents. The graph helps you correlate data over a time period and analyze the impact of change on the reliability of the service.
|
|
||
| The overall health score timeline provides a graphical representation of the high level overview of the service health for the time period that you have selected. The graph is divided into time slots of equal length based on the time period setting. The graph also displays the changes in deployment, infrastructure, and feature flag. It also displays the incidents. This helps you correlate the data over a time period and analyze the impact of change on the reliability of the service. | ||
|
|
||
| The overall health score is computed based on the change events such as deployments, infrastructure changes, and incidents. The health score ranges between 0 and 100. |
There was a problem hiding this comment.
| The overall health score is computed based on the change events such as deployments, infrastructure changes, and incidents. The health score ranges between 0 and 100. | |
| The overall health score is computed based on change events such as deployments, infrastructure changes, and incidents. The health score ranges between 0 and 100. |
- Can we review this sentence from a technical standpoint? Are deployments considered change events here? Ignore if accurate. Also, "change events such as ... infrastructure changes" sounds odd. Would help to reword.
- Also, when we use "between" with a numeric range, the end points are not included. Can a health score not be 0 or 100?
There was a problem hiding this comment.
Updated text:
The total health score is calculated based on deployment and infrastructure modification change events, as well as incidents. The total health score is calculated based on deployment and infrastructure modification change events, as well as incidents. The health score ranges from 0 and 100.
|
|
||
| #### Time Window | ||
|
|
||
| The Overall Health Score timeline has a **Time Window** which displays granular details such as health score, incident count, metrics, and logs for a specific timeline on the graph. You can move the Time Window along the graph to choose a specific timeline. You can increase or decrease the timeline using the handles. |
There was a problem hiding this comment.
| The Overall Health Score timeline has a **Time Window** which displays granular details such as health score, incident count, metrics, and logs for a specific timeline on the graph. You can move the Time Window along the graph to choose a specific timeline. You can increase or decrease the timeline using the handles. | |
| The Overall Health Score timeline has a time window that displays granular details such as health score, incident count, metrics, and logs for a specific time period on the graph. You can move the time window along the graph to choose a specific time period. You can increase or decrease the time window by using the handles. |
|
@deepakbolangady I'm not an expert in FF, but I can take a look. I'll tag someone from the FF team if I see anything that needs follow-up from an SME. |
|
@deepakbolangady and @KimberlyFields , the scope of FF topic here is limited to selecting a monitored service from within the FF. From SRM context, the user is not expected to do anything more here. |
SudheendraKatte
dismissed stale reviews from SmritiSatya and deepak-harness
via
2098674
|
Preview environment: https://hdh.pr.harness.io/pr-788 |
|
Preview environment: https://hdh.pr.harness.io/pr-788 |
Change Impact Analysis Quickstart
Harness Developer Pull Request
Thanks for helping us make the Developer Hub better. The PR will be looked at
by the maintainers.
What Type of PR is This?
If tied to an Issue, list the Issue(s) here:
House Keeping
Some items to keep track of. Screen shots of changes are optional but would help the maintainers review quicker.