search filter out not ready cluster

[yanfeng1992]

Signed-off-by: huangyanfeng [email protected]

What type of PR is this?
/kind bug

What this PR does / why we need it:

I have three cluster member1, member2, member3，member1 status is not ready

I did a few steps

step1 create a resourceregistries for all cluster

step 2 kubectl get --raw /apis/search.karmada.io/v1alpha1/search/cache/api/v1/nodes --kubeconfig /etc/karmada/karmada-apiserver.config, but get nothing

error log in karmada-search

I think the not ready cluster should be filtered out

Which issue(s) this PR fixes:
Fixes #

Special notes for your reviewer:

Does this PR introduce a user-facing change?:

karmada-search: filter out not ready clusters

[codecov-commenter]

Codecov Report

Merging #3010 (dd5232c) into master (434b7b6) will decrease coverage by 0.03%.
The diff coverage is 0.00%.

@@            Coverage Diff             @@
##           master    #3010      +/-   ##
==========================================
- Coverage   38.66%   38.62%   -0.04%     
==========================================
  Files         206      206              
  Lines       18815    18818       +3     
==========================================
- Hits         7274     7269       -5     
- Misses      11114    11120       +6     
- Partials      427      429       +2     

Flag	Coverage Δ
unittests	38.62% <0.00%> (-0.04%)	⬇️

Flags with carried forward coverage won't be shown. Click here to find out more.

Impacted Files	Coverage Δ
pkg/search/proxy/controller.go	83.97% <0.00%> (-1.65%)	⬇️
...kg/scheduler/framework/runtime/metrics_recorder.go	82.85% <0.00%> (-5.72%)	⬇️
pkg/util/worker.go	66.66% <0.00%> (-4.77%)	⬇️

Help us with your feedback. Take ten seconds to tell us how you rate us. Have a feature suggestion? Share it here.

[XiShanYongYe-Chang]

/cc @ikaven1024 @liys87x

[ikaven1024]

Thanks @yanfeng1992 , it's LGTM.

While I found a serious problems: Because cluster ready status is updated periodically, before control panel detects a member cluster down, search component will still request to the down cluster and get an error.

[yanfeng1992]

Thanks @yanfeng1992 , it's LGTM.

While I found a serious problems: Because cluster ready status is updated periodically, before control panel detects a member cluster down, search component will still request to the down cluster and get an error.

ok, i squash the commits into one @ikaven1024

[XiShanYongYe-Chang]

While I found a serious problems: Because cluster ready status is updated periodically, before control panel detects a member cluster down, search component will still request to the down cluster and get an error.

One thing I don't understand is why getting the cache of other clusters fails when one cluster fails.

[ikaven1024]

While I found a serious problems: Because cluster ready status is updated periodically, before control panel detects a member cluster down, search component will still request to the down cluster and get an error.

One thing I don't understand is why getting the cache of other clusters fails when one cluster fails.

When client list pods with resourceVersion=, proxy will traverse all the member clusters, fail to connect the down cluster and return error.

karmada/pkg/search/proxy/store/multi_cluster_cache.go

Lines 227 to 231 in 7e24783

	for clusterIdx, cluster := range clusters {
	n, cont, err := listFunc(cluster)
	if err != nil {
	return nil, err
	}

[yanfeng1992]

Is there anything else I need to modify in this PR? @XiShanYongYe-Chang

[XiShanYongYe-Chang]

Is there anything else I need to modify in this PR? @XiShanYongYe-Chang

I thinks it's okay,
Ask @ikaven1024 and @liys87x to confirm again.

[ikaven1024]

/lgtm

[XiShanYongYe-Chang]

Hi @yanfeng1992, can you help update the release note to:

karmada-search: filter out not ready clusters

We may need to cherry-pick this patch to the previous release v1.2-v1.4. Can you help do it?

[yanfeng1992]

Hi @yanfeng1992, can you help update the release note to:

karmada-search: filter out not ready clusters

We may need to cherry-pick this patch to the previous release v1.2-v1.4. Can you help do it?

i have updated @XiShanYongYe-Chang

[XiShanYongYe-Chang]

i have updated @XiShanYongYe-Chang

Thanks~
/approve

[karmada-bot]

[APPROVALNOTIFIER] This PR is APPROVED

This pull-request has been approved by: XiShanYongYe-Chang

The full list of commands accepted by this bot can be found here.

The pull request process is described here

Needs approval from an approver in each of these files:

• pkg/search/OWNERS [XiShanYongYe-Chang]

Approvers can indicate their approval by writing /approve in a comment
Approvers can cancel approval by writing /approve cancel in a comment

[XiShanYongYe-Chang]

Hi @yanfeng1992, can you help cherry-pick this patch to the previous release? You can refer to the https://karmada.io/docs/contributor/cherry-picks/

[yanfeng1992]

Hi @yanfeng1992, can you help cherry-pick this patch to the previous release? You can refer to the https://karmada.io/docs/contributor/cherry-picks/

release1.2 #3018
release1.3 #3019
release1.4 #3016

@XiShanYongYe-Chang

[XiShanYongYe-Chang]

release1.2 #3018 release1.3 #3019 release1.4 #3016

@XiShanYongYe-Chang

Thanks a lot~