Blog on "What's new in Network Observability 1.11"#31
Blog on "What's new in Network Observability 1.11"#31stleerh wants to merge 9 commits intonetobserv:mainfrom
Conversation
|
🚀 PR Preview for netobserv.io has been successfully deployed! |
| name: websvc | ||
| ``` | ||
|
|
||
| The FlowCollectorSlice instances won't take effect unless they are enabled in the FlowCollector. In OpenShift web console, configure this in the **Processor configuration** in **Slices configuration**. On the terminal, enter `oc edit flowcollector`, and configure the following under the **processor** section: |
There was a problem hiding this comment.
| The FlowCollectorSlice instances won't take effect unless they are enabled in the FlowCollector. In OpenShift web console, configure this in the **Processor configuration** in **Slices configuration**. On the terminal, enter `oc edit flowcollector`, and configure the following under the **processor** section: | |
| The FlowCollectorSlice instances won't take effect unless they are enabled in the FlowCollector. In OpenShift web console, configure this in the **Slices configuration** under **Processor configuration**. On the terminal, enter `oc edit flowcollector`, and configure the following under the **processor** section: |
There was a problem hiding this comment.
I reworded this so it's clearer. I didn't want to say "Slices configuration" first because you don't see this unless you click "Processor configuration" to open this up.
leandroberetta
left a comment
leandroberetta
left a comment
There was a problem hiding this comment.
LGTM with just a minor comment (optional). Thanks @stleerh, very nice blog.
| <br> | ||
| Figure 6. Configure recording rules in the healthRules section. | ||
|
|
||
| Both the recording rules and alerts fall under the umbrella of **healthRules**. When you select a template for a health rule, you will also need to include the corresponding metric. That is done in the **include list** section. Don't worry; a warning will appear if you didn't include the necessary metric. For the DNS-related health rule, you also need to enable the `DNSTracking` feature in eBPF. |
There was a problem hiding this comment.
I think clarifying the inclusion of metrics is good advice, although it’s not always needed. Also, I’m not sure if it’s worth mentioning, but when you include a metric, you overwrite the metrics list, so the user will probably lose other metrics (this is something we’ve discussed improving in the near future).
There was a problem hiding this comment.
Yes, this is what I experienced. I wasn't sure I wanted to mention it here, rather than to file an issue to improve this.
|
/lgtm |
| name: websvc | ||
| ``` | ||
|
|
||
| The FlowCollectorSlice instances won't take effect unless they are enabled in the FlowCollector. In OpenShift web console, click **Processor configuration** and then **Slices configuration** to configure this. Alternatively, on the terminal, enter `oc edit flowcollector`, and configure the following under the **processor** section: |
There was a problem hiding this comment.
To demonstrate the intended use-case, we should probably note:
- rbacs that cluster-admin should add non-admins so that they can create FlowCollectorSlice for their tenant.
- Also, that FlowCollector CR is for cluster-admins only.
There was a problem hiding this comment.
Added a sentence on this:
Unlike the FlowCollector CRD, which is configured by the cluster-admin, a typical use case for FlowCollectorSlice is to give RBAC access to a non-admin user, such as a project admin, so they can configure a FlowCollectorSlice instance for their namespaces.
|
@stleerh could you have a word about the performance improvement, and the fact that users upgrading have to apply the config change manually (https://docs.redhat.com/en/documentation/openshift_container_platform/4.21/html/network_observability/network-observability-operator-release-notes#:~:text=Improved%20performance%20for%20new%20Network%20Observability%20installations) |
7 participants
No description provided.