Andrey Aleksandrov created IGNITE-9840:
------------------------------------------
Summary: Possible deadlock on transactional future on client node
in case of network problems or long GC pauses
Key: IGNITE-9840
URL: https://issues.apache.org/jira/browse/IGNITE-9840
Project: Ignite
Issue Type: Bug
Components: clients
Affects Versions: 2.6
Reporter: Andrey Aleksandrov
Fix For: 2.8
Steps to reproduce:
1)Start the server node with next timeouts. DefaultTxTimeout should be greater
than other:
{code:java}
<property name="peerClassLoadingEnabled" value="true"/>
<property name="failureDetectionTimeout" value="60000"/>
<property name="clientFailureDetectionTimeout" value="60000"/>
<property name="networkTimeout" value="60000"/>
<property name="gridName" value="name"/>
<!--Transaction timeout setting-->
<property name="transactionConfiguration">
<bean class="org.apache.ignite.configuration.TransactionConfiguration">
<property name="DefaultTxTimeout" value="600000"/>
</bean>
</property>
<property name="idleConnectionTimeout" value="5000"/>
<property name="connectTimeout" value="5000"/>
<property name="ackTimeout" value="20000"/>
{code}
On the server side you should create a cache with next parameters:
{code:java}
<bean class="org.apache.ignite.configuration.CacheConfiguration">
<property name="name" value="CACHE"/>
<property name="cacheMode" value="PARTITIONED"/>
<property name="atomicityMode" value="TRANSACTIONAL"/>
<property name="writeSynchronizationMode" value="FULL_SYNC"/>
<property name="backups" value="1"/>
<property name="statisticsEnabled" value="true"/>{code}
2)After that start the client with the next code:
{code:java}
IgniteCache<String, Object> cache = dr1Hub.getOrCreateCache("LITARGETINFO");
try (Transaction tx = dr1Hub.transactions().txStart()) {
cache.put("Key", new Object());
System.out.println("Stop me");
//here we will get long GC pause on server side
Thread.sleep(10000);
// Commit the transaction.
tx.commitAsync().get();
}
{code}
On step "Stop me" you should suspend all the thread on the server side to
emulate the networking problem or long GC pause on the server side.
Finally, you will face in client node next:
{code:java}
[2018-10-10 16:46:10,157][ERROR][nio-acceptor-tcp-comm-#28%GRIDC1%][root]
Critical system error detected. Will be handled accordingly to configured
handler [hnd=StopNodeOrHaltFailureHandler [tryStop=false, timeout=0,
super=AbstractFailureHandler [ignoredFailureTypes=UnmodifiableSet
[SYSTEM_WORKER_BLOCKED]]], failureCtx=FailureContext
[type=SYSTEM_WORKER_BLOCKED, err=class o.a.i.IgniteException: GridWorker
[name=grid-timeout-worker, igniteInstanceName=GRIDC1, finished=false,
heartbeatTs=1539179057570]]]
{code}
Also, the similar issue could be reproduced in 2.4. In both cases looks like we
have a deadlock during trying to display the TxEntryValueHolder. Looks like
this values are already used by the transaction with long DefaultTxTimeout .
{code:java}
java.lang.Thread.State: WAITING
at sun.misc.Unsafe.park(Unsafe.java:-1)
at java.util.concurrent.locks.LockSupport.park(LockSupport.java:304)
at
org.apache.ignite.internal.util.future.GridFutureAdapter.get0(GridFutureAdapter.java:177)
at
org.apache.ignite.internal.util.future.GridFutureAdapter.get(GridFutureAdapter.java:140)
at
org.apache.ignite.internal.processors.cache.binary.CacheObjectBinaryProcessorImpl.metadata0(CacheObjectBinaryProcessorImpl.java:526)
at
org.apache.ignite.internal.processors.cache.binary.CacheObjectBinaryProcessorImpl.metadata(CacheObjectBinaryProcessorImpl.java:510)
at
org.apache.ignite.internal.processors.cache.binary.CacheObjectBinaryProcessorImpl$2.metadata(CacheObjectBinaryProcessorImpl.java:193)
at
org.apache.ignite.internal.binary.BinaryContext.metadata(BinaryContext.java:1265)
at org.apache.ignite.internal.binary.BinaryUtils.type(BinaryUtils.java:2407)
at
org.apache.ignite.internal.binary.BinaryObjectImpl.rawType(BinaryObjectImpl.java:302)
at
org.apache.ignite.internal.binary.BinaryObjectExImpl.toString(BinaryObjectExImpl.java:205)
at
org.apache.ignite.internal.binary.BinaryObjectExImpl.toString(BinaryObjectExImpl.java:186)
at
org.apache.ignite.internal.binary.BinaryObjectImpl.toString(BinaryObjectImpl.java:919)
at java.lang.String.valueOf(String.java:2994)
at java.lang.StringBuilder.append(StringBuilder.java:131)
at
org.apache.ignite.internal.processors.cache.transactions.TxEntryValueHolder.toString(TxEntryValueHolder.java:161)
...{code}
On the client side, it could be looked like a hanging transaction because we
waiting on:
{code:java}
tx.commitAsync().get();{code}
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
