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

Posted to issues@ignite.apache.org by "Roman Puchkovskiy (Jira)" <ji...@apache.org> on 2024/09/25 14:17:00 UTC

[jira] [Updated] (IGNITE-23273) Shared ConversionService causes troubles after a node gets stopped

     [ https://issues.apache.org/jira/browse/IGNITE-23273?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Roman Puchkovskiy updated IGNITE-23273:
---------------------------------------
    Description: 
By default, Micronaut uses a shared instance of ConversionService. This means that all Ignite nodes on the same JVM (which is what happens in our tests) share the same ConversionService instance. When one of them is stopped, the service gets reset, non-default converters get removed from it, including JsonNode->Object, which causes all REST calls where data is passed as JSON in request body to start failing.

We need to get rid of using the shared approach and create a ConversionService instance per Ignite node.

> Shared ConversionService causes troubles after a node gets stopped
> ------------------------------------------------------------------
>
>                 Key: IGNITE-23273
>                 URL: https://issues.apache.org/jira/browse/IGNITE-23273
>             Project: Ignite
>          Issue Type: Bug
>            Reporter: Roman Puchkovskiy
>            Priority: Major
>              Labels: ignite-3
>
> By default, Micronaut uses a shared instance of ConversionService. This means that all Ignite nodes on the same JVM (which is what happens in our tests) share the same ConversionService instance. When one of them is stopped, the service gets reset, non-default converters get removed from it, including JsonNode->Object, which causes all REST calls where data is passed as JSON in request body to start failing.
> We need to get rid of using the shared approach and create a ConversionService instance per Ignite node.



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