[
https://issues.apache.org/jira/browse/IGNITE-21168?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17810279#comment-17810279
]
Vadim Pakhnushev edited comment on IGNITE-21168 at 1/24/24 9:01 AM:
--------------------------------------------------------------------
This is no longer the case as the job executions are cleared automatically
after they are completed.
was (Author: JIRAUSER289102):
This is no more the case as the job executions are cleared automatically after
they are completed.
> Clean up job execution in case sender left
> ------------------------------------------
>
> Key: IGNITE-21168
> URL: https://issues.apache.org/jira/browse/IGNITE-21168
> Project: Ignite
> Issue Type: Improvement
> Components: compute
> Reporter: Vadim Pakhnushev
> Assignee: Vadim Pakhnushev
> Priority: Major
> Labels: ignite-3
>
> When {{ComputeMessaging}} receives execute request, it stores the job
> execution object in order to respond to result or cancel requests later.
> Normally sender requests result immediately after executing and execution
> object will be cleaned as soon as the result response is sent back to sender.
> In case connectivity is lost between execution start and result request we
> will have a dangling reference.
> We should subscribe to topology change events and clean the job execution
> when sender has left the topology.
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
