Uploaded image for project: 'Infinispan'
  1. Infinispan
  2. ISPN-1386

Nodes stopping can be merged without a reduced view being installed

    XMLWordPrintable

Details

    Description

      The attached log is showing a test failure because a view is not stablished with the two remaining nodes:

      2011-09-14 13:01:15,124 ERROR (testng-TopologyInfoBroadcastTest) [org.infinispan.test.fwk.UnitTestTestNGListener] Method 
testNodeLeaves(org.infinispan.distribution.topologyaware.TopologyInfoBroadcastTest) threw an exception
java.lang.RuntimeException: Timed out before caches had complete views.  Expected 2 members in each view.  Views are as follows: 
[[TopologyInfoBroadcastTest-NodeA-20211(s0), TopologyInfoBroadcastTest-NodeB-32215(s1), TopologyInfoBroadcastTest-NodeC-36650(s2)], 
[TopologyInfoBroadcastTest-NodeA-20211(s0), TopologyInfoBroadcastTest-NodeB-32215(s1), TopologyInfoBroadcastTest-NodeC-36650(s2)]]

      Having glanced some TRACE logs captured by Dan, it would appear that as NodeB is shutting down, TEST_PING finds it as a sole cluster member and finds another cluster formed of NodeA,NodeC, leading to a merge. This is somehow stopping a correct 2-node view being installed.

      Needs further investigation

      Attachments

        Issue Links

          is blocked by

          Enhancement - An enhancement to or refactoring of existing functionality that is not configurable by an end user (typically a change made by an internal team that affects users) ISPN-1229 Use ProtocolStackConfigurator to edit the JGroups configuration instead of regular expressions

          • Minor - To be worked after urgent, high, and medium priorities are resolved. This term is chosen when the severity of the issue is low, or the complexity or effort to correct it may be higher, relatively speaking. For low priority issues, known workarounds exist or are not needed due to the trivial effort needed to address the issue.
          • Closed
          is related to

          Bug - A problem that impairs or prevents the functions of the product. JGRP-1336 GMS: if a member leaving takes a long time, it can get merged back into the cluster

          • Minor - To be worked after urgent, high, and medium priorities are resolved. This term is chosen when the severity of the issue is low, or the complexity or effort to correct it may be higher, relatively speaking. For low priority issues, known workarounds exist or are not needed due to the trivial effort needed to address the issue.
          • Resolved
          relates to

          Enhancement - An enhancement to or refactoring of existing functionality that is not configurable by an end user (typically a change made by an internal team that affects users) ISPN-1395 Remove TEST_PING workaround that avoids replying to MERGE ping when stopped

          • Major - To be worked after higher priority work items are addressed. When the severity is high and the effort to change it is low to moderate, this term is chosen. Issues in this category likely have an existing workaround but implementation or execution may be non-trivial.
          • Resolved

          Activity

            People

              rh-ee-galder Galder Zamarreño
              rh-ee-galder Galder Zamarreño
              Votes:
              0 Vote for this issue
              Watchers:
              2 Start watching this issue

              Dates

                Created:
                Updated:
                Resolved: