[DEFLAKE] Psync established after rdb load - beyond grace period

[roshkhatri]

Resolves: #2695

We saw the same error once in #2694 where the rdb took a little longer to load and thats where the test failed.

I ran it for 100 time in a workflow here https://github.com/roshkhatri/valkey/actions/runs/18605561661/job/53054114207 where I see max of 5.5 seconds to load the RDB.

Here I have increased the check for total of 6 seconds and IMO we would see it failing again, unless the rdb load get even slower in valgrid.

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 72.65%. Comparing base (b4c93cc) to head (8d7ddb9).
⚠️ Report is 1 commits behind head on unstable.

Additional details and impacted files

@@             Coverage Diff              @@
##           unstable    #2748      +/-   ##
============================================
+ Coverage     72.59%   72.65%   +0.05%     
============================================
  Files           128      128              
  Lines         71301    71300       -1     
============================================
+ Hits          51759    51800      +41     
+ Misses        19542    19500      -42     

see 17 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.

[zuiderkwast]

The syntax of this is proc is

proc wait_for_log_messages {srv_idx patterns from_line maxtries delay}

Reading the file every 5 milliseconds and repeating 2000 times seems like a quite heavy operation. Can we change these to a delay of 100 or 50 like in other wait-for expressions?

Suggested change

	wait_for_log_messages 0 {"*Done loading RDB*"} $loglines 2000 5
	wait_for_log_messages 0 {"*Done loading RDB*"} $loglines 100 100

[roshkhatri]

Yeah, we can so 100 100 and I dont think there should be an issue, we could do the same with the previous test Psync established after rdb load - within grace period

[zuiderkwast]

Yeah, I see there are multiple (seven?) with 2000 1 in test suite, one with 4000 1, one with 20000 1 and eight(?) with 1000 10.

Let's change them all to use a delay of 100 while keeping the same total time?

[sarthakaggarwal97]

@roshkhatri were we able to reproduce the test failure?