You are viewing a plain text version of this content. The canonical link for it is here.
Posted to issues@ignite.apache.org by "Evgeny Stanilovsky (Jira)" <ji...@apache.org> on 2024/07/29 14:03:00 UTC
[jira] [Updated] (IGNITE-22855) Sql. Missing ClusterNode can be involved into tx enlist
[ https://issues.apache.org/jira/browse/IGNITE-22855?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Evgeny Stanilovsky updated IGNITE-22855:
----------------------------------------
Epic Link: IGNITE-22815
> Sql. Missing ClusterNode can be involved into tx enlist
> -------------------------------------------------------
>
> Key: IGNITE-22855
> URL: https://issues.apache.org/jira/browse/IGNITE-22855
> Project: Ignite
> Issue Type: Bug
> Components: sql
> Affects Versions: 3.0.0-beta1
> Reporter: Evgeny Stanilovsky
> Priority: Major
> Labels: ignite-3
>
> DistributedQueryManager#sendFragments enlists partitions:
> {noformat}
> if (!tx.isReadOnly()) {
> for (MappedFragment mappedFragment : mappedFragments) {
> enlistPartitions(mappedFragment, tx);
> }
> }
> {noformat}
> and obtain ClusterNode from topologyService, here:
> IgniteRelShuttle#enlist
> but node can leave topology at that moment and *null* will return in such a case.
> {noformat}
> @Nullable ClusterNode getByConsistentId(String consistentId);
> {noformat}
> need to cover such a case and re run existing query
--
This message was sent by Atlassian Jira
(v8.20.10#820010)