Flaky test_close_renamed in Windows

[jsoriano]

Flaky Test

• Test Name: test_close_renamed
• Link:

  beats/filebeat/tests/system/test_harvester.py

  Lines 69 to 72 in af4007e

  	# Make sure new file was picked up. As it has the same file name,
  	# one entry for the new and one for the old should exist
  	self.wait_until(
  	lambda: len(self.get_registry()) == 2, max_timeout=10)

• Branch: 7.x
• Notes: Related to recent registry refactors?

Stack Trace

self = <test_harvester.Test testMethod=test_close_renamed>
    def test_close_renamed(self):
        """
        Checks that a file is closed when its renamed / rotated
        """
    
        self.render_config_template(
            path=os.path.abspath(self.working_dir) + "/log/test.log",
            close_renamed="true",
            clean_removed="false",
            scan_frequency="0.1s"
        )
        os.mkdir(self.working_dir + "/log/")
    
        testfile1 = self.working_dir + "/log/test.log"
        testfile2 = self.working_dir + "/log/test.log.rotated"
        file = open(testfile1, 'w')
    
        iterations1 = 5
        for n in range(0, iterations1):
            file.write("rotation file")
            file.write("\n")
    
        file.close()
    
        filebeat = self.start_beat()
    
        # Let it read the file
        self.wait_until(
            lambda: self.output_has(lines=iterations1), max_timeout=10)
    
        os.rename(testfile1, testfile2)
    
        with open(testfile1, 'w') as f:
            f.write("Hello World\n")
    
        # Wait until error shows up
        self.wait_until(
            lambda: self.log_contains(
                "Closing because close_renamed is enabled"),
            max_timeout=15)
    
        # Let it read the file
        self.wait_until(
            lambda: self.output_has(lines=iterations1 + 1), max_timeout=10)
    
        # Wait until registry file is created
        self.wait_until(
            lambda: self.log_contains_count("Registry file updated") > 1)
    
        # Make sure new file was picked up. As it has the same file name,
        # one entry for the new and one for the old should exist
>       self.wait_until(
            lambda: len(self.get_registry()) == 2, max_timeout=10)
tests\system\test_harvester.py:71: 
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
self = <test_harvester.Test testMethod=test_close_renamed>
cond = <function Test.test_close_renamed.<locals>.<lambda> at 0x0000000EB2D58CA0>
max_timeout = 10, poll_interval = 0.1, name = 'cond'
    def wait_until(self, cond, max_timeout=10, poll_interval=0.1, name="cond"):
        """
        Waits until the cond function returns true,
        or until the max_timeout is reached. Calls the cond
        function every poll_interval seconds.
    
        If the max_timeout is reached before cond() returns
        true, an exception is raised.
        """
        start = datetime.now()
        while not cond():
            if datetime.now() - start > timedelta(seconds=max_timeout):
>               raise TimeoutError("Timeout waiting for '{}' to be true. ".format(name) +
                                   "Waited {} seconds.".format(max_timeout))
E               beat.beat.TimeoutError: Timeout waiting for 'cond' to be true. Waited 10 seconds.
..\libbeat\tests\system\beat\beat.py:362: TimeoutError

[elasticmachine]

Pinging @elastic/integrations-services (Team:Services)

[botelastic]

This issue has been automatically marked as stale because it has not had recent activity. It will be closed if no further activity occurs. Thank you for your contributions.

[jlind23]

Backlog grooming: Closing for now.

[elasticmachine]

Pinging @elastic/elastic-agent-data-plane (Team:Elastic-Agent-Data-Plane)

[jlind23]

@cmacknz What should we do with this one? Fixing it or removing it?

[cmacknz]

It looks like it is only skipped for Windows, so the test can stay for the other platforms.

Fixing this is probably fairly low value, we could just close this issue and leave the test as is. The issue link will stay in the code where it is marked as skipped.

The test has been disabled for over a year without issue.