You are viewing a plain text version of this content. The canonical link for it is here.
Posted to issues@geode.apache.org by "Bruce Schuchardt (JIRA)" <ji...@apache.org> on 2019/04/30 17:14:00 UTC
[jira] [Created] (GEODE-6724) split brain formed on concurrent
locator startup
Bruce Schuchardt created GEODE-6724:
---------------------------------------
Summary: split brain formed on concurrent locator startup
Key: GEODE-6724
URL: https://issues.apache.org/jira/browse/GEODE-6724
Project: Geode
Issue Type: Bug
Components: membership
Reporter: Bruce Schuchardt
In a test with network-partition-detection disabled four locators were spun up in parallel and they formed two different clusters. Two servers were started and they joined different clusters, ending up with different data. Consistency checks at the end of the test caught the problem.
{noformat}
locatorgemfire_1_2_17088/system.log: [info 2019/04/25 22:00:08.732 PDT <vm_1_thr_1_locator_1_2_host1_17088> tid=0x14] findCoordinator chose rs-FullRegression26040030a0i3large-hydra-client-53(locatorgemfire_1_2_host1_17088:17088:locator)<ec>:41001 out of these possible coordinators: [rs-FullRegression26040030a0i3large-hydra-client-53(locatorgemfire_1_2_host1_17088:17088:locator)<ec>:41001]
locatorgemfire_1_2_17088/system.log: [info 2019/04/25 22:00:08.733 PDT <vm_1_thr_1_locator_1_2_host1_17088> tid=0x14] Discovery state after looking for membership coordinator is SearchState(locatorsContacted=2; findInViewResponses=0; alreadyTried=[]; registrants=[rs-FullRegression26040030a0i3large-hydra-client-53(locatorgemfire_1_2_host1_17088:17088:locator)<ec>:41001]; possibleCoordinator=rs-FullRegression26040030a0i3large-hydra-client-53(locatorgemfire_1_2_host1_17088:17088:locator)<ec>:41001; viewId=-1; hasContactedAJoinedLocator=false; view=null; responses=[])
locatorgemfire_1_2_17088/system.log: [info 2019/04/25 22:00:08.733 PDT <vm_1_thr_1_locator_1_2_host1_17088> tid=0x14] found possible coordinator rs-FullRegression26040030a0i3large-hydra-client-53(locatorgemfire_1_2_host1_17088:17088:locator)<ec>:41001
locatorgemfire_1_2_17088/system.log: [info 2019/04/25 22:00:08.733 PDT <vm_1_thr_1_locator_1_2_host1_17088> tid=0x14] This member is becoming the membership coordinator with address rs-FullRegression26040030a0i3large-hydra-client-53(locatorgemfire_1_2_host1_17088:17088:locator)<ec>:41001
{noformat}
{noformat}
locatorgemfire_1_4_17106/system.log: [info 2019/04/25 22:00:08.762 PDT <vm_3_thr_3_locator_1_4_host1_17106> tid=0x14] findCoordinator chose rs-FullRegression26040030a0i3large-hydra-client-53(locatorgemfire_1_4_host1_17106:17106:locator)<ec>:41000 out of these possible coordinators: [rs-FullRegression26040030a0i3large-hydra-client-53(locatorgemfire_1_4_host1_17106:17106:locator)<ec>:41000]
locatorgemfire_1_4_17106/system.log: [info 2019/04/25 22:00:08.763 PDT <vm_3_thr_3_locator_1_4_host1_17106> tid=0x14] Discovery state after looking for membership coordinator is SearchState(locatorsContacted=3; findInViewResponses=0; alreadyTried=[]; registrants=[rs-FullRegression26040030a0i3large-hydra-client-53(locatorgemfire_1_2_host1_17088:17088:locator)<ec>:41001, rs-FullRegression26040030a0i3large-hydra-client-53(locatorgemfire_1_4_host1_17106:17106:locator)<ec>:41000, rs-FullRegression26040030a0i3large-hydra-client-53(locatorgemfire_1_3_host1_17100:17100:locator)<ec>:41002]; possibleCoordinator=rs-FullRegression26040030a0i3large-hydra-client-53(locatorgemfire_1_4_host1_17106:17106:locator)<ec>:41000; viewId=-1; hasContactedAJoinedLocator=false; view=null; responses=[])
locatorgemfire_1_4_17106/system.log: [info 2019/04/25 22:00:08.763 PDT <vm_3_thr_3_locator_1_4_host1_17106> tid=0x14] found possible coordinator rs-FullRegression26040030a0i3large-hydra-client-53(locatorgemfire_1_4_host1_17106:17106:locator)<ec>:41000
locatorgemfire_1_4_17106/system.log: [info 2019/04/25 22:00:08.763 PDT <vm_3_thr_3_locator_1_4_host1_17106> tid=0x14] This member is becoming the membership coordinator with address rs-FullRegression26040030a0i3large-hydra-client-53(locatorgemfire_1_4_host1_17106:17106:locator)<ec>:41000
{noformat}
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)