[
https://issues.apache.org/jira/browse/IGNITE-14068?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17415160#comment-17415160
]
Ignite TC Bot commented on IGNITE-14068:
----------------------------------------
{panel:title=Branch: [pull/9393/head] Base: [master] : No blockers
found!|borderStyle=dashed|borderColor=#ccc|titleBGColor=#D6F7C1}{panel}
{panel:title=Branch: [pull/9393/head] Base: [master] : New Tests
(3)|borderStyle=dashed|borderColor=#ccc|titleBGColor=#D6F7C1}
{color:#00008b}SPI{color} [[tests
3|https://ci.ignite.apache.org/viewLog.html?buildId=6179706]]
* {color:#013220}IgniteSpiTestSuite:
TcpDiscoverySelfTest.testIncomingConnectionsFailure - PASSED{color}
* {color:#013220}IgniteSpiTestSuite:
TcpDiscoverySslTrustedSelfTest.testIncomingConnectionsFailure - PASSED{color}
* {color:#013220}IgniteSpiTestSuite:
TcpDiscoverySslSelfTest.testIncomingConnectionsFailure - PASSED{color}
{panel}
[TeamCity *--> Run :: All*
Results|https://ci.ignite.apache.org/viewLog.html?buildId=6179777&buildTypeId=IgniteTests24Java8_RunAll]
> Infinite node presence in the ring while outgoing connections are lost
> ----------------------------------------------------------------------
>
> Key: IGNITE-14068
> URL: https://issues.apache.org/jira/browse/IGNITE-14068
> Project: Ignite
> Issue Type: Bug
> Reporter: Vladimir Steshin
> Assignee: Vladimir Steshin
> Priority: Major
> Time Spent: 10h
> Remaining Estimate: 0h
>
> If node looses outgoing connections, it can decide it is alone in the cluster
> and won't fail. Happens on small clusters where failed node attempts to
> connect to every other node before connRecoveryTimeout expires.
> Consider:
> - The cluster n1 -> n2 -> n3 -> n4 -> n1
> - n4 looses all outgoing connections.
> - n3 keeps successful ping to n4.
> - n4 attempts to connect to n1, n2, n3. Fails with each due to outgoing
> network failure.
> - spi.connrecoveryTimeout is not reached. n4 decides it is alone and
> continues working.
> - n3 still sends messages to n4. n4 does not lack incoming connections.
> - ring is actually broken because of n4. n3 cannot determine failure of n4.
> Solution: node could watch its incoming traffic which notyfies of the
> incoming network. If all the outgoing connections are lost but messages are
> received, node must left the grid to prevent ring break.
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
