[
https://issues.apache.org/jira/browse/IGNITE-18451?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Roman Puchkovskiy updated IGNITE-18451:
---------------------------------------
Description:
# Start and init a cluster of 3 nodes
# Via node 0, create a table with 3 replicas
# Do some stuff
# Stop the cluster
The cluster cannot be stoppled, stop of node 2 seems to hang forever. There is
a thread:
"%itrst_tsimwkpn_2%JRaft-FSMCaller-Disruptor-_stripe_3-0" #253053 daemon prio=5
os_prio=0 cpu=0,59ms elapsed=1169,99s tid=0x00007fb4867e3000 nid=0xa1668
waiting on condition [0x00007fb4832f0000]
java.lang.Thread.State: WAITING (parking)
at jdk.internal.misc.Unsafe.park([email protected]/Native Method)
- parking to wait for <0x0000000707d9a508> (a
java.util.concurrent.CompletableFuture$Signaller)
at
java.util.concurrent.locks.LockSupport.park([email protected]/LockSupport.java:194)
at
java.util.concurrent.CompletableFuture$Signaller.block([email protected]/CompletableFuture.java:1796)
at
java.util.concurrent.ForkJoinPool.managedBlock([email protected]/ForkJoinPool.java:3128)
at
java.util.concurrent.CompletableFuture.waitingGet([email protected]/CompletableFuture.java:1823)
at
java.util.concurrent.CompletableFuture.join([email protected]/CompletableFuture.java:2043)
at org.apache.ignite.internal.table.TableImpl.pkId(TableImpl.java:126)
at
org.apache.ignite.internal.table.TableImpl.awaitIndexes(TableImpl.java:347)
at
org.apache.ignite.internal.table.TableImpl.lambda$indexStorageAdapters$0(TableImpl.java:237)
at
org.apache.ignite.internal.table.TableImpl$$Lambda$1842/0x0000000800a2fc40.get(Unknown
Source)
at
org.apache.ignite.internal.table.distributed.raft.PartitionListener.addToIndexes(PartitionListener.java:435)
at
org.apache.ignite.internal.table.distributed.raft.PartitionListener.lambda$handleUpdateAllCommand$5(PartitionListener.java:234)
at
org.apache.ignite.internal.table.distributed.raft.PartitionListener$$Lambda$2551/0x0000000800ceb440.execute(Unknown
Source)
at
org.apache.ignite.internal.storage.rocksdb.RocksDbMvPartitionStorage.runConsistently(RocksDbMvPartitionStorage.java:249)
at
org.apache.ignite.internal.table.distributed.raft.snapshot.outgoing.SnapshotAwarePartitionDataStorage.runConsistently(SnapshotAwarePartitionDataStorage.java:59)
at
org.apache.ignite.internal.table.distributed.raft.PartitionListener.handleUpdateAllCommand(PartitionListener.java:219)
at
org.apache.ignite.internal.table.distributed.raft.PartitionListener.lambda$onWrite$1(PartitionListener.java:153)
at
org.apache.ignite.internal.table.distributed.raft.PartitionListener$$Lambda$1944/0x0000000800a7a440.accept(Unknown
Source)
at java.util.Iterator.forEachRemaining([email protected]/Iterator.java:133)
at
org.apache.ignite.internal.table.distributed.raft.PartitionListener.onWrite(PartitionListener.java:123)
at
org.apache.ignite.internal.raft.server.impl.JraftServerImpl$DelegatingStateMachine.onApply(JraftServerImpl.java:563)
at
org.apache.ignite.raft.jraft.core.FSMCallerImpl.doApplyTasks(FSMCallerImpl.java:565)
at
org.apache.ignite.raft.jraft.core.FSMCallerImpl.doCommitted(FSMCallerImpl.java:530)
at
org.apache.ignite.raft.jraft.core.FSMCallerImpl.runApplyTask(FSMCallerImpl.java:449)
at
org.apache.ignite.raft.jraft.core.FSMCallerImpl$ApplyTaskHandler.onEvent(FSMCallerImpl.java:137)
at
org.apache.ignite.raft.jraft.core.FSMCallerImpl$ApplyTaskHandler.onEvent(FSMCallerImpl.java:131)
at
org.apache.ignite.raft.jraft.disruptor.StripedDisruptor$StripeEntryHandler.onEvent(StripedDisruptor.java:217)
at
org.apache.ignite.raft.jraft.disruptor.StripedDisruptor$StripeEntryHandler.onEvent(StripedDisruptor.java:181)
at com.lmax.disruptor.BatchEventProcessor.run(BatchEventProcessor.java:137)
at java.lang.Thread.run([email protected]/Thread.java:834)
Probably the node stop was initiated before the indices were created for the
table, then index manager was stopped, so the indices never get created. But
only index creation completes the future that is being avaited by the shown
thread.
This might have something to do with IGNITE-18203.
was:TBD
> Node stop may hang if a Raft service is not started yet
> -------------------------------------------------------
>
> Key: IGNITE-18451
> URL: https://issues.apache.org/jira/browse/IGNITE-18451
> Project: Ignite
> Issue Type: Bug
> Reporter: Roman Puchkovskiy
> Priority: Major
> Labels: ignite-3
> Fix For: 3.0.0-beta2
>
>
> # Start and init a cluster of 3 nodes
> # Via node 0, create a table with 3 replicas
> # Do some stuff
> # Stop the cluster
> The cluster cannot be stoppled, stop of node 2 seems to hang forever. There
> is a thread:
> "%itrst_tsimwkpn_2%JRaft-FSMCaller-Disruptor-_stripe_3-0" #253053 daemon
> prio=5 os_prio=0 cpu=0,59ms elapsed=1169,99s tid=0x00007fb4867e3000
> nid=0xa1668 waiting on condition [0x00007fb4832f0000]
> java.lang.Thread.State: WAITING (parking)
> at jdk.internal.misc.Unsafe.park([email protected]/Native Method)
> - parking to wait for <0x0000000707d9a508> (a
> java.util.concurrent.CompletableFuture$Signaller)
> at
> java.util.concurrent.locks.LockSupport.park([email protected]/LockSupport.java:194)
> at
> java.util.concurrent.CompletableFuture$Signaller.block([email protected]/CompletableFuture.java:1796)
> at
> java.util.concurrent.ForkJoinPool.managedBlock([email protected]/ForkJoinPool.java:3128)
> at
> java.util.concurrent.CompletableFuture.waitingGet([email protected]/CompletableFuture.java:1823)
> at
> java.util.concurrent.CompletableFuture.join([email protected]/CompletableFuture.java:2043)
> at org.apache.ignite.internal.table.TableImpl.pkId(TableImpl.java:126)
> at
> org.apache.ignite.internal.table.TableImpl.awaitIndexes(TableImpl.java:347)
> at
> org.apache.ignite.internal.table.TableImpl.lambda$indexStorageAdapters$0(TableImpl.java:237)
> at
> org.apache.ignite.internal.table.TableImpl$$Lambda$1842/0x0000000800a2fc40.get(Unknown
> Source)
> at
> org.apache.ignite.internal.table.distributed.raft.PartitionListener.addToIndexes(PartitionListener.java:435)
> at
> org.apache.ignite.internal.table.distributed.raft.PartitionListener.lambda$handleUpdateAllCommand$5(PartitionListener.java:234)
> at
> org.apache.ignite.internal.table.distributed.raft.PartitionListener$$Lambda$2551/0x0000000800ceb440.execute(Unknown
> Source)
> at
> org.apache.ignite.internal.storage.rocksdb.RocksDbMvPartitionStorage.runConsistently(RocksDbMvPartitionStorage.java:249)
> at
> org.apache.ignite.internal.table.distributed.raft.snapshot.outgoing.SnapshotAwarePartitionDataStorage.runConsistently(SnapshotAwarePartitionDataStorage.java:59)
> at
> org.apache.ignite.internal.table.distributed.raft.PartitionListener.handleUpdateAllCommand(PartitionListener.java:219)
> at
> org.apache.ignite.internal.table.distributed.raft.PartitionListener.lambda$onWrite$1(PartitionListener.java:153)
> at
> org.apache.ignite.internal.table.distributed.raft.PartitionListener$$Lambda$1944/0x0000000800a7a440.accept(Unknown
> Source)
> at java.util.Iterator.forEachRemaining([email protected]/Iterator.java:133)
> at
> org.apache.ignite.internal.table.distributed.raft.PartitionListener.onWrite(PartitionListener.java:123)
> at
> org.apache.ignite.internal.raft.server.impl.JraftServerImpl$DelegatingStateMachine.onApply(JraftServerImpl.java:563)
> at
> org.apache.ignite.raft.jraft.core.FSMCallerImpl.doApplyTasks(FSMCallerImpl.java:565)
> at
> org.apache.ignite.raft.jraft.core.FSMCallerImpl.doCommitted(FSMCallerImpl.java:530)
> at
> org.apache.ignite.raft.jraft.core.FSMCallerImpl.runApplyTask(FSMCallerImpl.java:449)
> at
> org.apache.ignite.raft.jraft.core.FSMCallerImpl$ApplyTaskHandler.onEvent(FSMCallerImpl.java:137)
> at
> org.apache.ignite.raft.jraft.core.FSMCallerImpl$ApplyTaskHandler.onEvent(FSMCallerImpl.java:131)
> at
> org.apache.ignite.raft.jraft.disruptor.StripedDisruptor$StripeEntryHandler.onEvent(StripedDisruptor.java:217)
> at
> org.apache.ignite.raft.jraft.disruptor.StripedDisruptor$StripeEntryHandler.onEvent(StripedDisruptor.java:181)
> at
> com.lmax.disruptor.BatchEventProcessor.run(BatchEventProcessor.java:137)
> at java.lang.Thread.run([email protected]/Thread.java:834)
> Probably the node stop was initiated before the indices were created for the
> table, then index manager was stopped, so the indices never get created. But
> only index creation completes the future that is being avaited by the shown
> thread.
> This might have something to do with IGNITE-18203.
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
