[IGNITE-7592] Dynamic cache with rebalanceDelay == -1 doesn't trigger late affinity assignment even after explicit rebalance is called on every node

• Type: Bug
• Status: Resolved
• Priority: Major
• Resolution: Fixed
• Affects Version/s: 2.4
• Fix Version/s: 2.7
• Assignee: Maxim Muzafarov
• Reporter: Ilya Lantukh
• Created: 31/Jan/18 15:13
• Updated: 11/Dec/18 12:41
• Resolved: 27/Apr/18 13:09

Description

Reproducer:

startGrids(NODE_COUNT);

IgniteEx ig = grid(0);

ig.cluster().active(true);

awaitPartitionMapExchange();

IgniteCache<Integer, Integer> cache =
    ig.createCache(
        new CacheConfiguration<Integer, Integer>()
            .setName(CACHE_NAME)
            .setCacheMode(PARTITIONED)
            .setBackups(1)
            .setPartitionLossPolicy(READ_ONLY_SAFE)
            .setReadFromBackup(true)
            .setWriteSynchronizationMode(FULL_SYNC)
            .setRebalanceDelay(-1)
    );

for (int i = 0; i < NODE_COUNT; i++)
    grid(i).cache(CACHE_NAME).rebalance().get();

awaitPartitionMapExchange();

Sometimes this code will hang on the last awaitPartitionMapExchange(), though probability that it will happen is rather low (<10%).

Issue Links

• is related to: IGNITE-8086 Flaky test timeouts in Activate/Deactivate Cluster suite
• links to: 1250311
• links to: GitHub Pull Request #3918
• links to: IGNT-CR-589