[
https://issues.apache.org/jira/browse/IGNITE-13323?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17171045#comment-17171045
]
Andrey N. Gura commented on IGNITE-13323:
-----------------------------------------
[~YAMolochkov], you can just make a copy under mutex and log info after mutex
is released.
> GridDhtPartitionsExchangeFuture logging improvements
> ----------------------------------------------------
>
> Key: IGNITE-13323
> URL: https://issues.apache.org/jira/browse/IGNITE-13323
> Project: Ignite
> Issue Type: Improvement
> Affects Versions: 2.8.1
> Reporter: Yaroslav Molochkov
> Assignee: Yaroslav Molochkov
> Priority: Major
>
> Firstly, logging in said class should be improved to address possible cluster
> topology/connectivity issues:
> org.apache.ignite.internal.processors.cache.distributed.dht.preloader.GridDhtPartitionsExchangeFuture#processFullMessage,
> upon entering that method and before critical section begins we should log
> that at info level. This way we won't have to speculate if thread is waiting
> on a lock to
> not.(org.apache.ignite.internal.processors.cache.distributed.dht.preloader.GridDhtPartitionsExchangeFuture#mux)
> Secondly,
> org.apache.ignite.internal.processors.cache.distributed.dht.preloader.GridDhtPartitionsExchangeFuture#sendAllPartitions.
> Logging should reflect the amount of nodes that should receive messages
> (collection that is sent to said method) and the amount of nodes that should
> receive the message (or, in other words, we successfully managed to send the
> message to a particular node, then sum it up).
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
