Set repl-timeout for slotmigrations tests to prevent disconnections

[enjoy-binbin]

This test will insert keys, and as can be seen from the logs, the insertion
somehow is very slow in this CI, eventually causing a repl-timeout disconnection.

50655:M 14 May 2026 00:44:57.812 - DB 0: 5 keys (0 volatile) in 7 slots HT.
50655:M 14 May 2026 00:45:02.884 - DB 0: 6 keys (0 volatile) in 7 slots HT.
50655:M 14 May 2026 00:45:06.014 # Timing out slot migration xxx after not receiving ack for too long

Set repl-timeout for these tests to prevent the timeout disconnections.
Also the tests does not actually require inserting distinct keys, we only
need to fill the replication buffer, so there is no need for the different
keyname, this can save the CI some memory.

Closes #3702.

[coderabbitai]

No actionable comments were generated in the recent review. 🎉

ℹ️ Recent review info

⚙️ Run configuration

Configuration used: Repository UI

Review profile: CHILL

Plan: Pro Plus

Run ID: e05fe207-a6af-47c1-8691-07d2abd1c281

📥 Commits

Reviewing files that changed from the base of the PR and between fdf13ca and 2204760.

📒 Files selected for processing (1)

• tests/unit/cluster/cluster-migrateslots.tcl

---

📝 Walkthrough

Walkthrough

Two test cases in the cluster migration test suite are updated: each receives an increased replication timeout override and a modification to how test data is populated. The data generation now writes the same key repeatedly instead of creating unique keys per loop iteration.

Changes

Cluster migration test reliability adjustments

Layer / File(s)	Summary
AOF/consistency test replication timeout and data population tests/unit/cluster/cluster-migrateslots.tcl	Adds repl-timeout 3600 override to the start_cluster call in the AOF/consistency test section and changes the data population loop to write to a single fixed key "$16383_slot_tag:key" instead of generating unique keys with :$j suffix.
Slot migration test replication timeout and data population tests/unit/cluster/cluster-migrateslots.tcl	Adds repl-timeout 3600 override to the start_cluster call in the slot-migration-max-failover-repl-bytes test section and changes the data population loop to write to a single fixed key "$16383_slot_tag:key" instead of generating unique keys with :$j suffix.

Estimated code review effort

🎯 2 (Simple) | ⏱️ ~8 minutes

Possibly related issues

• [TEST-FAILURE] slot-migration-max-failover-repl-bytes -1 disables repl bytes limit in tests/unit/cluster/cluster-migrateslots.tcl #3702: The changes to the slot-migration-max-failover-repl-bytes test (increased replication timeout and modified data population pattern) directly address the test failure in that scenario by adjusting both the timeout duration and the volume of data generated during migration, which affects the remaining_repl_size assertion.

🚥 Pre-merge checks | ✅ 5

✅ Passed checks (5 passed)

Check name	Status	Explanation
Title check	✅ Passed	The title clearly and specifically describes the main change: setting repl-timeout for slot migration tests to prevent disconnections.
Description check	✅ Passed	The description is clearly related to the changeset, explaining the slow key insertion causing repl-timeout issues and the solution implemented.
Linked Issues check	✅ Passed	The PR adequately addresses issue #3702 by implementing repl-timeout override (3600 seconds) to prevent replication timeout disconnections that were causing test failures.
Out of Scope Changes check	✅ Passed	All changes are scoped to addressing the repl-timeout issue: adding repl-timeout override and simplifying key generation to reduce memory usage, both directly addressing the test failure.
Docstring Coverage	✅ Passed	No functions found in the changed files to evaluate docstring coverage. Skipping docstring coverage check.

_{✏️ Tip: You can configure your own custom pre-merge checks in the settings.}

---

_{Comment @coderabbitai help to get the list of available commands and usage tips.}

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 76.69%. Comparing base (ca9dee3) to head (2204760).
⚠️ Report is 7 commits behind head on unstable.

Additional details and impacted files

@@             Coverage Diff              @@
##           unstable    #3703      +/-   ##
============================================
- Coverage     76.71%   76.69%   -0.03%     
============================================
  Files           162      162              
  Lines         80656    80662       +6     
============================================
- Hits          61872    61860      -12     
- Misses        18784    18802      +18     

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.

[rainsupreme]

The change looks good to me. Did you verify that this fixes the flakiness? It's weird that it would be so slow in the CI workflow 🤔

[zuiderkwast]

OK, if you saw this failure...

Replicating 1MB key takes more than one minute?

[zuiderkwast]

Ah, I see the ASM code uses repl-timeout for this:

            if (last_interaction &&
                (server.unixtime - last_interaction > server.repl_timeout)) {
                serverLog(LL_WARNING,
                          "Timing out slot migration %s "
                          "after not receiving ack for too long",
                          job->description);

Then, I'm convinced it will fix the failure where this warning is logged. 👍

[enjoy-binbin]

Did you verify that this fixes the flakiness? It's weird that it would be so slow in the CI workflow

I did not bother to trigger the test in slow CI... I think the change is good enough to merge.