[ https://issues.apache.org/jira/browse/HBASE-9305?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Andrew Purtell updated HBASE-9305: ---------------------------------- Attachment: 9305.patch Triage by not checking hit counts at the problematic point in the test. May just move the problem. Testing. > [0.92] TestFromClientSide.testCacheOnWriteEvictOnClose fails occasionally > ------------------------------------------------------------------------- > > Key: HBASE-9305 > URL: https://issues.apache.org/jira/browse/HBASE-9305 > Project: HBase > Issue Type: Bug > Affects Versions: 0.92.3 > Reporter: Andrew Purtell > Assignee: Andrew Purtell > Priority: Minor > Attachments: 9305.patch > > > The assertion failures are like this: > {noformat} > java.lang.AssertionError: expected:<2089> but was:<2109> > at org.junit.Assert.fail(Assert.java:93) > at org.junit.Assert.failNotEquals(Assert.java:647) > at org.junit.Assert.assertEquals(Assert.java:128) > at org.junit.Assert.assertEquals(Assert.java:472) > at org.junit.Assert.assertEquals(Assert.java:456) > at > org.apache.hadoop.hbase.client.TestFromClientSide.testCacheOnWriteEvictOnClose(TestFromClientSide.java:4248) > {noformat} > Also: > {noformat} > expected:<2067> but was:<2087> > {noformat} > {noformat} > expected:<2070> but was:<2090> > {noformat} > The test saves off the current block cache stats - block count and hits and > misses - then puts a value and gets it back: > {code} > 4242: Put put = new Put(ROW); > 4243: put.add(FAMILY, QUALIFIER, data); > 4244: table.put(put); > 4245: assertTrue(Bytes.equals(table.get(new Get(ROW)).value(), data)); > {code} > then we have these asserts: > {code} > 4246: //data was in memstore so don't expect any changes > 4247: assertEquals(startBlockCount, cache.getBlockCount()); > 4248: assertEquals(startBlockHits, cache.getStats().getHitCount()); > 4249: assertEquals(startBlockMiss, cache.getStats().getMissCount()); > {code} > There are exactly 20 more hits than expected every time. In the log looks > like there's a meta scan happening around the same time. -- This message is automatically generated by JIRA. If you think it was sent incorrectly, please contact your JIRA administrators For more information on JIRA, see: http://www.atlassian.com/software/jira