You are viewing a plain text version of this content. The canonical link for it is here.

Posted to issues@ignite.apache.org by "Vyacheslav Koptilin (Jira)" <ji...@apache.org> on 2026/03/17 11:18:00 UTC

[jira] [Updated] (IGNITE-28123) ClientPrimaryReplicaTracker: avoid null-partitions failure when primary-replica lookup completes exceptionally

     [ https://issues.apache.org/jira/browse/IGNITE-28123?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Vyacheslav Koptilin updated IGNITE-28123:
-----------------------------------------
    Labels: ignite-3  (was: )

> ClientPrimaryReplicaTracker: avoid null-partitions failure when primary-replica lookup completes exceptionally
> --------------------------------------------------------------------------------------------------------------
>
>                 Key: IGNITE-28123
>                 URL: https://issues.apache.org/jira/browse/IGNITE-28123
>             Project: Ignite
>          Issue Type: Bug
>            Reporter: Anton Laletin
>            Assignee: Anton Laletin
>            Priority: Major
>              Labels: ignite-3
>             Fix For: 3.2
>
>          Time Spent: 0.5h
>  Remaining Estimate: 0h
>
> Under node restarts, partition assignment lookup may complete exceptionally while resolving primary replicas.
> In ClientPrimaryReplicaTracker.primaryReplicasAsyncInternal, the exceptional path was not terminated for non-TableNotFound errors, and execution could continue into result construction with partitions == null, causing a secondary failure (NullPointerException from unboxing).



--
This message was sent by Atlassian Jira
(v8.20.10#820010)