[
https://issues.apache.org/jira/browse/IGNITE-14439?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17402606#comment-17402606
]
Franco Po commented on IGNITE-14439:
------------------------------------
Hi [~pvinokurov], I reported IGNITE-15343 which is a NPE issue on client start.
The Ignite client in use is GridGain Community Edition 8.8.5, which includes
the fix of [GG-32794 Fix NPE on node join caused by obtaining system
cache|https://github.com/gridgain/gridgain/commit/239cd8b9259a5d755bcee5f3191b4551f40ff434].
I am not sure if the NPE issue I reported is related to IGNITE-11406. In
consideration of minimizing changes in the system, I would like to explore if I
patch GG CE 8.8.5 by patching some class files. Any insight for me as I can see
ClusterProcessor.java has been fixed with below since GG CE 8.8.4?
{code:java}
try {
ctx.cache().awaitStarted();
}
catch (IgniteCheckedException e) {
throw U.convertException(e);
}
{code}
> NPE when accessing clustername before first exchange finished
> -------------------------------------------------------------
>
> Key: IGNITE-14439
> URL: https://issues.apache.org/jira/browse/IGNITE-14439
> Project: Ignite
> Issue Type: Bug
> Components: cache
> Affects Versions: 2.9
> Reporter: Pavel Vinokurov
> Assignee: Pavel Vinokurov
> Priority: Major
> Fix For: 2.11
>
> Time Spent: 20m
> Remaining Estimate: 0h
>
> [IGNITE-11406|https://issues.apache.org/jira/browse/IGNITE-11406] has not
> been fixed properly for two reasons. The first is one is that
> _GridCacheProcessor.utilityCache_ could be accessed before the first exchange
> finished. The second is that it doesn't resolve the original issue, because
> _GridServiceProcessor.onKernelStop_ is followed by
> _GridCacheProcessor.onKernelStop_, so caches should be already initialized.
> Thus that fix should be reverted.
> Revering this fix induces the issue related to accessing the utility cache by
> getting cluster name.
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
