Skip to content

Update eck pdb docs #2361

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

Open
wants to merge 2 commits into
base: main
Choose a base branch
from
Open

Conversation

naemono
Copy link
Contributor

@naemono naemono commented Aug 1, 2025

The following elastic/cloud-on-k8s#8780 is planned to be released with the ECK 3.2/Stack 9.2 release. How do we tag this specifically for that?

naemono added 2 commits August 1, 2025 10:24
Signed-off-by: Michael Montgomery <[email protected]>
Copy link

github-actions bot commented Aug 1, 2025

🔍 Preview links for changed docs

Copy link
Collaborator

@shainaraskas shainaraskas left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

provided a suggestion with the info at hand. might need some adjustment depending on the impact of the stack version on this process.

@@ -12,7 +12,28 @@ products:

A [Pod Disruption Budget](https://kubernetes.io/docs/tasks/run-application/configure-pdb/) (PDB) allows you to limit the disruption to your application when its pods need to be rescheduled for some reason such as upgrades or routine maintenance work on the Kubernetes nodes.

ECK manages a default PDB per {{es}} resource. It allows one {{es}} Pod to be taken down, as long as the cluster has a `green` health. Single-node clusters are not considered highly available and can always be disrupted.
ECK manages either a single default PDB, or multiple PDBs per {{es}} resource according to the license available.
Copy link
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Does the user have to be on a stack 9.2 cluster as well as use ECK 3.2?

Consider this approach. More tagging / clarification might be necessary if the stack version is also a factor.

We have a couple of other ways we could do this with heading tagging or tabs, but this is probably a nice lightweight way to do it

Suggested change
ECK manages either a single default PDB, or multiple PDBs per {{es}} resource according to the license available.
ECK manages either a single default PDB, or multiple PDBs per {{es}} resource according to the license available.
:::{note}
In ECK 3.1 and earlier, all clusters follow the [non-enterprise behavior](#non-enterprise-licensed-customers), regardless of license type.
:::

ECK manages a default PDB per {{es}} resource. It allows one {{es}} Pod to be taken down, as long as the cluster has a `green` health. Single-node clusters are not considered highly available and can always be disrupted.
ECK manages either a single default PDB, or multiple PDBs per {{es}} resource according to the license available.

## Enterprise licensed customers
Copy link
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Suggested change
## Enterprise licensed customers
## Enterprise licensed customers
```{applies_to}
deployment:
eck: ga 3.2
```


Single-node clusters are not considered highly available and can always be disrupted.

## Non-enterprise licensed customers
Copy link
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Suggested change
## Non-enterprise licensed customers
## Non-enterprise licensed customers
:::{note}
In ECK 3.1 and earlier, all clusters follow this behavior regardless of license type.
:::


A separate PDB is created for each type of nodeSet defined in the manifest allowing upgrade or maintenance operations to be more quickly executed. The PDBs allow one {{es}} Pod per nodeSet to simultaneously be taken down as long as the cluster has the health defined in the following table:

| Role | Cluster Health Required | Notes |
Copy link
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Suggested change
| Role | Cluster Health Required | Notes |
| Role | Cluster health required | Notes |

| ML | Yellow | |
| Coordinating | Yellow | |
| Transform | Yellow | |
| Remote Cluster Client | Yellow | |
Copy link
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Suggested change
| Remote Cluster Client | Yellow | |
| Remote cluster client | Yellow | |


## Non-enterprise licensed customers

It allows one {{es}} Pod to be taken down, as long as the cluster has a `green` health. Single-node clusters are not considered highly available and can always be disrupted.

In the {{es}} specification, you can change the default behavior as follows:
Copy link
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

do the instructions below this line only apply to non-enterprise? if not, you might want to change the headings a little:

## Default behavior
... 
### Enterprise licensed customers
...
### Non-enterprise licensed customers
...
## Override the default behavior
...
## Pod disruption budget ...

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

2 participants