[
https://issues.apache.org/jira/browse/CASSANDRA-18710?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17786446#comment-17786446
]
Jacek Lewandowski commented on CASSANDRA-18710:
-----------------------------------------------
The problem with small flush affecting the test case in question can be easily
reliably reproduced by adding the following delays:
in DiskSizeMetricsTest:
{code:java}
private void insertN(String keyspace, ColumnFamilyStore cfs, int n, long base)
throws Throwable
{
for (int i = 0; i < n; i++)
{
Thread.sleep(1); // <------------ HERE
executeFormattedQuery(formatQuery(keyspace, "INSERT INTO %s (pk) VALUES
(?)"), base + i);
}
// flush to write the sstable
cfs.forceBlockingFlush();
}
{code}
in CQLTester:
{code:java}
@After
public void afterTest() throws Throwable
{
dropPerTestKeyspace();
...
// We want to clean up after the test, but dropping a table is rather long
so just do that asynchronously
ScheduledExecutors.optionalTasks.execute(new Runnable()
{
public void run()
{
try
{
Thread.sleep(500); // <----------------- HERE
for (int i = tablesToDrop.size() - 1; i >= 0; i--)
schemaChange(String.format("DROP TABLE IF EXISTS %s.%s",
KEYSPACE, tablesToDrop.get(i)));
{code}
Then run the whole test class (not just the test case)
> Test failure:
> org.apache.cassandra.io.DiskSpaceMetricsTest.testFlushSize-.jdk17 (from
> org.apache.cassandra.io.DiskSpaceMetricsTest-.jdk17)
> ------------------------------------------------------------------------------------------------------------------------------------------
>
> Key: CASSANDRA-18710
> URL: https://issues.apache.org/jira/browse/CASSANDRA-18710
> Project: Cassandra
> Issue Type: Bug
> Components: Test/unit
> Reporter: Ekaterina Dimitrova
> Assignee: Jacek Lewandowski
> Priority: Normal
> Fix For: 5.0-beta, 5.0.x, 5.x
>
> Attachments: org.apache.cassandra.io.DiskSpaceMetricsTest.txt
>
>
> Seen here:
> [https://ci-cassandra.apache.org/job/Cassandra-trunk/1644/testReport/org.apache.cassandra.io/DiskSpaceMetricsTest/testFlushSize__jdk17/]
> h3.
> {code:java}
> Error Message
> expected:<7200.0> but was:<1367.83970468544>
> Stacktrace
> junit.framework.AssertionFailedError: expected:<7200.0> but
> was:<1367.83970468544> at
> org.apache.cassandra.io.DiskSpaceMetricsTest.testFlushSize(DiskSpaceMetricsTest.java:119)
> at java.base/jdk.internal.reflect.NativeMethodAccessorImpl.invoke0(Native
> Method) at
> java.base/jdk.internal.reflect.NativeMethodAccessorImpl.invoke(NativeMethodAccessorImpl.java:77)
> at
> java.base/jdk.internal.reflect.DelegatingMethodAccessorImpl.invoke(DelegatingMethodAccessorImpl.java:43)
> {code}
>
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
---------------------------------------------------------------------
To unsubscribe, e-mail: commits-unsubscr...@cassandra.apache.org
For additional commands, e-mail: commits-h...@cassandra.apache.org
