Skip to content
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

[BUG] Filter Steve WebSocket errors. #40627

Open
KevinJoiner opened this issue Feb 21, 2023 · 1 comment
Open

[BUG] Filter Steve WebSocket errors. #40627

KevinJoiner opened this issue Feb 21, 2023 · 1 comment
Labels
area/scalability 10k or bust kind/bug Issues that are defects reported by users or that we know have reached a real release priority/1 team/frameworks the team that is responsible for frameworks, projects, webhook, and other various areas team/scaling
Milestone

Comments

@KevinJoiner
Copy link
Contributor

Rancher Server Setup

  • Rancher version: v2.7
  • Installation option (Docker install/Helm Chart): Any

Information about the Cluster

  • Kubernetes version: NA
  • Cluster Type (Local/Downstream): N/A

User Information

  • What is the role of the user logged in? (Admin/Cluster Owner/Cluster Member/Project Owner/Project Member/Custom)
    • If custom, define the set of permissions: N/A

Describe the bug
Steve was updated to send errors encountered when creating events to the frontend via resource.error messages.
Some of these errors are expected and harmless and thus should be suppressed and not sent to the frontend.

  • clusters.provisioning.cattle.io "<resource_name" not found.
    These errors are caused by steve attempting to stream information about related resources that may or may not still exist.
    If the resource does not exist the message should be suppressed. Steve Code

  • unable to decode an event from the watch stream: http2: response body closed
    These messages are sent when Steve closes the WebSocket via context cancelation. Resulting in the underlying k8s decoder for the watch command failing to read a valid object. apimachinery already checks for some of these types of error messages here, but does not handle this one since it is specific to the watch command
    Related to: [BUG] Creating/deleting management resources cause resources to be unsubscribed #40558

To Reproduce

  1. Start a new rancher server
  2. Inspect the WebSocket traffic for v1/subscribe
  3. Create a new project and then delete the project

Result
Users observe the above error messages.

Expected Result
users observe new resource stop and start messages

Screenshots
image

Additional context

@KevinJoiner KevinJoiner added kind/bug Issues that are defects reported by users or that we know have reached a real release area/scalability 10k or bust team/area1 labels Feb 21, 2023
@KevinJoiner KevinJoiner added this to the 2023-Q2-v2.7x milestone Feb 21, 2023
@samjustus samjustus modified the milestones: 2023-Q2-v2.7x, v2.7.x Feb 27, 2023
@richard-cox
Copy link
Member

richard-cox commented Jun 15, 2023

Users often quote these when providing logs / reporting issues and can really muddy the water. Hiding or downgrading them would help a lot.

Edit: We've had a few more cases where customers have supplied these thinking it's the root cause of their issue, which makes it harder for us to get the required part log section

@MbolotSuse MbolotSuse modified the milestones: v2.7.x, v2.9-Next2 Apr 23, 2024
@MbolotSuse MbolotSuse added team/frameworks the team that is responsible for frameworks, projects, webhook, and other various areas and removed team/area3 labels Apr 23, 2024
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
area/scalability 10k or bust kind/bug Issues that are defects reported by users or that we know have reached a real release priority/1 team/frameworks the team that is responsible for frameworks, projects, webhook, and other various areas team/scaling
Projects
None yet
Development

No branches or pull requests

6 participants