[
https://issues.apache.org/jira/browse/GEODE-5728?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Dale Emery closed GEODE-5728.
-----------------------------
> Suspect string (NullPointerException) found in
> LocatorDUnitTest.testStartTwoLocators
> ------------------------------------------------------------------------------------
>
> Key: GEODE-5728
> URL: https://issues.apache.org/jira/browse/GEODE-5728
> Project: Geode
> Issue Type: Bug
> Components: membership
> Reporter: Ken Howe
> Assignee: Dale Emery
> Priority: Major
> Labels: pull-request-available, swat
> Fix For: 1.8.0
>
> Time Spent: 20m
> Remaining Estimate: 0h
>
> DistributedTest failure in
> https://concourse.apachegeode-ci.info/teams/main/pipelines/develop/jobs/DistributedTest/builds/379
> {code}
> java.lang.AssertionError: Suspicious strings were written to the log during
> this run.
> Fix the strings or use IgnoredException.addIgnoredException to ignore.
> -----------------------------------------------------------------------
> Found suspect string in log4j at line 1241
> [fatal 2018/09/11 00:04:32.409 UTC <locator request thread[1]> tid=0x676]
> Exception in processing request from 172.17.0.20
> java.lang.NullPointerException
> at
> org.apache.geode.distributed.internal.membership.gms.locator.GMSLocator.processFindCoordinatorRequest(GMSLocator.java:309)
> at
> org.apache.geode.distributed.internal.membership.gms.locator.GMSLocator.processRequest(GMSLocator.java:205)
> at
> org.apache.geode.distributed.internal.InternalLocator$PrimaryHandler.processRequest(InternalLocator.java:1194)
> at
> org.apache.geode.distributed.internal.tcpserver.TcpServer.processOneConnection(TcpServer.java:484)
> at
> org.apache.geode.distributed.internal.tcpserver.TcpServer.lambda$processRequest$0(TcpServer.java:378)
> at
> java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1149)
> at
> java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:624)
> at java.lang.Thread.run(Thread.java:748)
> {code}
> The test doesn't have a history of (flaky) failures that I've discovered, and
> I haven't gotten it to fail locally.
> The NPE comes from a null return from {{v.getCoordinator()}}.
> {code}
> synchronized (registrants) {
> if (isCoordinator) {
> coordinator = localAddress;
> if (v != null && !v.getCoordinator().equals(localAddress)) {
> logger.info("This member is becoming coordinator since view {}", v);
> v = null;
> }
> }
> {code}
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
