[DEFLAKE] Fix/cluster slot migration flaky test

[hanxizh9910]

Address: #2693

I add an additional check to ensure that all nodes recognize the full cluster (4 nodes) after a new node is added. This helps prevent the test from failing due to cluster gossip not being fully propagated.

I ran the test 100 times for both the unmodified and modified versions, and all runs passed:

Original (unmodified): CI run

Modified (with fix): CI run

For reference, this test previously failed once in https://github.com/valkey-io/valkey/actions/runs/18266295942/job/52001092528

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 72.46%. Comparing base (b4c93cc) to head (8df0ded).
⚠️ Report is 3 commits behind head on unstable.

Additional details and impacted files

@@             Coverage Diff              @@
##           unstable    #2756      +/-   ##
============================================
- Coverage     72.59%   72.46%   -0.13%     
============================================
  Files           128      128              
  Lines         71301    71303       +2     
============================================
- Hits          51759    51671      -88     
- Misses        19542    19632      +90     

see 21 files with indirect coverage changes

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.
• 📦 JS Bundle Analysis: Save yourself from yourself by tracking and limiting bundle sizes in JS merges.

[enjoy-binbin]

we have wait_for_cluster_size 4 in line 276, it is indedd checking the cluster_known_nodes

# Check if cluster size is consistent.
proc cluster_size_consistent {cluster_size} {
    for {set j 0} {$j < $cluster_size} {incr j} {
        if {[CI $j cluster_known_nodes] ne $cluster_size} {
            return 0
        }
    }
    return 1
}

[hanxizh9910]

You are right! i will remove this.

[zuiderkwast]

As binbin noted, this change is not necessary, so there is no need to review this right now. I changed it to draft to avoid maintainers attention. I don't know the plans for this PR, to update it or close it later?

[hanxizh9910]

As binbin noted, this change is not necessary, so there is no need to review this right now. I changed it to draft to avoid maintainers attention. I don't know the plans for this PR, to update it or close it later?

Yes, thank you for doing this! I’m trying to think through other possible causes. It seems that the code inserts the node into cluster->nodes on each server, which increases the size to 4, but some internal registration steps might not have completed yet. I’ll look into it further.