Pavel Voronkin created IGNITE-9793:
--------------------------------------
Summary: Deactivation, segmentation of one node, activation may
lead to hang activation forever
Key: IGNITE-9793
URL: https://issues.apache.org/jira/browse/IGNITE-9793
Project: Ignite
Issue Type: Bug
Affects Versions: 2.5
Reporter: Pavel Voronkin
There is coordinator and ring of nodes
coordinator -> 1 -> 2 - > 3 -> 4
coordinator deactivated:
2018-09-24 15:09:01.609 [INFO
][exchange-worker-#153%DPL_GRID%DplGridNodeName%][o.a.i.i.p.c.d.d.p.GridDhtPartitionsExchangeFuture]
Successfully deactivated data structures, services and caches
[nodeId=e002e011-8d1c-4353-a0f3-b71264c5b0f4, client=false,
topVer=AffinityTopologyVersion [topVer=183, minorTopVer=1]]
2018-09-24 15:09:01.620
[DEBUG][exchange-worker-#153%DPL_GRID%DplGridNodeName%][o.a.i.i.p.c.d.d.p.l.ExchangeLatchManager]
Server latch is created [latch=CompletableLatchUid\{id='exchange',
topVer=AffinityTopologyVersion [topVer=183, minorTopVer=1]},
participantsSize=160]
2018-09-24 15:09:01.621 [INFO ][exchange-worker-#153%DPL_GRID%DplGridNodeName%]
nodes 1, 2, 3, 4 were deactivated:
2018-09-24 15:09:01.609 [INFO
][exchange-worker-#153%DPL_GRID%DplGridNodeName%][o.a.i.i.p.c.d.d.p.GridDhtPartitionsExchangeFuture]
Successfully deactivated data structures, services and caches
[nodeId=e002e011-8d1c-4353-a0f3-b71264c5b0f4, client=false,
topVer=AffinityTopologyVersion [topVer=183, minorTopVer=1]]
2018-09-24 15:09:03.328 [INFO
][exchange-worker-#153%DPL_GRID%DplGridNodeName%][o.a.i.i.p.c.d.d.p.GridDhtPartitionsExchangeFuture]
Successfully deactivated data structures, services and caches
[nodeId=22a58223-47b5-43c2-897b-e70e8e50edf7, client=false,
topVer=AffinityTopologyVersion [topVer=183, minorTopVer=1]]
2018-09-24 15:09:03.334 [INFO
][exchange-worker-#153%DPL_GRID%DplGridNodeName%][o.a.i.i.p.c.d.d.p.GridDhtPartitionsExchangeFuture]
Successfully deactivated data structures, services and caches
[nodeId=973eb8ce-3b8c-463d-a6ab-00ac66d93f13, client=false,
topVer=AffinityTopologyVersion [topVer=183, minorTopVer=1]]
2018-09-24 15:09:03.332 [INFO
][exchange-worker-#153%DPL_GRID%DplGridNodeName%][o.a.i.i.p.c.d.d.p.GridDhtPartitionsExchangeFuture]
Successfully deactivated data structures, services and caches
[nodeId=a904bac4-aaed-4f69-90f3-c13bc4d331d1, client=false,
topVer=AffinityTopologyVersion [topVer=183, minorTopVer=1]]
Node 2 SEGMENTED
2018-09-24 15:17:50.068 [WARN
][tcp-disco-msg-worker-#2%DPL_GRID%DplGridNodeName%][o.a.i.s.d.tcp.TcpDiscoverySpi]
Node is out of topology (probably, due to short-time network problems).
2018-09-24 15:17:50.069 [WARN
][disco-event-worker-#152%DPL_GRID%DplGridNodeName%][o.a.i.i.m.d.GridDiscoveryManager]
Local node SEGMENTED: TcpDiscoveryNode
[id=a904bac4-aaed-4f69-90f3-c13bc4d331d1, addrs=ArrayList [10.116.206.98],
sockAddrs=HashSet [grid724.domain/10.116.206.98:47500], discPort=47500,
order=110, intOrder=110, lastExchangeTime=1537791470063, loc=true,
ver=2.5.1#20180906-sha1:ebde6c79, isClient=false]
Coordinator started activation on topology without node2
2018-09-24 15:19:48.686 [INFO
][exchange-worker-#153%DPL_GRID%DplGridNodeName%][o.a.i.i.p.c.d.d.p.GridDhtPartitionsExchangeFuture]
Start activation process [nodeId=e002e011-8d1c-4353-a0f3-b71264c5b0f4,
client=false, topVer=AffinityTopologyVersion [topVer=188, minorTopVer=1]]
But node 3 which is next to node 2 haven't received activation message.
Coordinator sent activation to all except
2018-09-24 15:24:25.911 [INFO
][sys-#28144%DPL_GRID%DplGridNodeName%][o.a.i.i.p.c.d.d.p.GridDhtPartitionsExchangeFuture]
Coordinator received single message [ver=AffinityTopologyVersion [topVer=188,
minorTopVer=1], node=073f1598-6b70-49df-8f45-126735611775, allReceived=false]
GridDhtPartitionsExchangeFuture hangs forever.
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)