[jira] [Updated] (IGNITE-11620) GridDhtInvalidPartitionException stops the cluster

["Vyacheslav Koptilin (JIRA)" <ji...@apache.org>]

     [ https://issues.apache.org/jira/browse/IGNITE-11620?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Vyacheslav Koptilin updated IGNITE-11620:
-----------------------------------------
    Ignite Flags:   (was: Docs Required)

> GridDhtInvalidPartitionException stops the cluster
> --------------------------------------------------
>
>                 Key: IGNITE-11620
>                 URL: https://issues.apache.org/jira/browse/IGNITE-11620
>             Project: Ignite
>          Issue Type: Bug
>    Affects Versions: 2.6, 2.7
>            Reporter: Roman Shtykh
>            Priority: Critical
>
> When injecting data and having it expired at the same time rebalancing occurs,
> *GridDhtInvalidPartitionException* triggers *SYSTEM_WORKER_TERMINATION*.
> This can cause cascading failures in the cluster and take the whole cluster down.
> Simple test case:
> {noformat}
> import org.apache.ignite.IgniteCache;
> import org.apache.ignite.configuration.CacheConfiguration;
> import org.apache.ignite.configuration.IgniteConfiguration;
> import org.apache.ignite.failure.StopNodeOrHaltFailureHandler;
> import org.apache.ignite.spi.discovery.tcp.TcpDiscoverySpi;
> import org.apache.ignite.spi.discovery.tcp.ipfinder.TcpDiscoveryIpFinder;
> import org.apache.ignite.spi.discovery.tcp.ipfinder.vm.TcpDiscoveryVmIpFinder;
> import org.apache.ignite.testframework.junits.common.GridCommonAbstractTest;
> import javax.cache.expiry.CreatedExpiryPolicy;
> import javax.cache.expiry.Duration;
> import java.util.concurrent.CountDownLatch;
> import java.util.concurrent.TimeUnit;
> import static org.apache.ignite.cache.CacheAtomicityMode.ATOMIC;
> import static org.apache.ignite.cache.CacheMode.PARTITIONED;
> /**
>  *
>  */
> public class ExpireWhileRebalanceTest extends GridCommonAbstractTest {
>     private static final int ENTRIES = 500000;
>     /**
>      *
>      */
>     protected static final TcpDiscoveryIpFinder IP_FINDER = new TcpDiscoveryVmIpFinder(true);
>     /**
>      * {@inheritDoc}
>      */
>     @Override
>     protected IgniteConfiguration getConfiguration(String gridName) throws Exception {
>         IgniteConfiguration cfg = super.getConfiguration(gridName);
>         ((TcpDiscoverySpi) cfg.getDiscoverySpi()).setIpFinder(IP_FINDER);
>         cfg.setFailureHandler(new StopNodeOrHaltFailureHandler());
>         CacheConfiguration<Object, Object> ccfg = new CacheConfiguration<>(DEFAULT_CACHE_NAME);
>         ccfg.setAtomicityMode(ATOMIC);
>         ccfg.setCacheMode(PARTITIONED);
>         ccfg.setExpiryPolicyFactory(CreatedExpiryPolicy.factoryOf(new Duration(TimeUnit.SECONDS, 1)));
>         cfg.setCacheConfiguration(ccfg);
>         return cfg;
>     }
>     /**
>      * @throws Exception If failed.
>      */
>     public void testExpireWhileRebalancing() throws Exception {
>         startGridsMultiThreaded(4);
>         IgniteCache<Object, Object> cache = ignite(0).cache(DEFAULT_CACHE_NAME);
>         CountDownLatch latch = new CountDownLatch(1);
>         new Thread(() -> {
>             for (int i = 1; i <= ENTRIES; i++) {
>                 cache.put(i, i);
>                 if (i % (ENTRIES / 10) == 0)
>                     System.out.println(">>> Entries put: " + i);
>             }
>             latch.countDown();
>         }).start();
>         // stopping 0 has no effect
>         stopGrid(3);
>         awaitPartitionMapExchange();
>         startGrid(3);
>         latch.await(10, TimeUnit.SECONDS);
>     }
>     /**
>      * {@inheritDoc}
>      */
>     @Override
>     protected void afterTest() throws Exception {
>         stopAllGrids();
>     }
> }
> {noformat}



--
This message was sent by Atlassian JIRA
(v7.6.3#76005)