Ilya Kasnacheev created IGNITE-14466:
----------------------------------------
Summary: Changing cache configuration once cache is created
confuses PME on node join
Key: IGNITE-14466
URL: https://issues.apache.org/jira/browse/IGNITE-14466
Project: Ignite
Issue Type: Bug
Affects Versions: 2.7.5
Reporter: Ilya Kasnacheev
The following code triggers the issue:
{code:java}
inputCache = ignite.cache(inputCacheName);
CacheConfiguration configuration =
inputCache.getConfiguration(CacheConfiguration.class);
outputCache = ignite.getOrCreateCache(configuration.setName("cache_" +
ctx.name()));
{code}
It is possible for user code to accidentally reuse the same cache configuration
instance when creating caches (see linked example). This causes the following
hard to debug NPE:
{code:java}
Mar 30, 2021 11:53:25 AM org.apache.ignite.logger.java.JavaLogger error
SEVERE: Exception in discovery notyfier worker thread.
java.lang.NullPointerException
at
org.apache.ignite.internal.managers.discovery.GridDiscoveryManager.addClientNode(GridDiscoveryManager.java:445)
at
org.apache.ignite.internal.processors.cache.ClusterCachesInfo.processCacheChangeRequests(ClusterCachesInfo.java:596)
at
org.apache.ignite.internal.processors.cache.ClusterCachesInfo.onCacheChangeRequested(ClusterCachesInfo.java:430)
at
org.apache.ignite.internal.processors.cache.GridCacheProcessor.onCustomEvent(GridCacheProcessor.java:3827)
at
org.apache.ignite.internal.managers.discovery.GridDiscoveryManager$4.onDiscovery0(GridDiscoveryManager.java:697)
at
org.apache.ignite.internal.managers.discovery.GridDiscoveryManager$4.lambda$onDiscovery$0(GridDiscoveryManager.java:604)
at
org.apache.ignite.internal.managers.discovery.GridDiscoveryManager$DiscoveryMessageNotifierWorker.body0(GridDiscoveryManager.java:2667)
at
org.apache.ignite.internal.managers.discovery.GridDiscoveryManager$DiscoveryMessageNotifierWorker.body(GridDiscoveryManager.java:2705)
at
org.apache.ignite.internal.util.worker.GridWorker.run(GridWorker.java:120)
at java.lang.Thread.run(Thread.java:748)
{code}
We should, maybe, clone a cache configuration before putting it to our internal
data structures? Or better, serialize-deserialize as all of the remaining nodes
in the cluster do.
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
