You are viewing a plain text version of this content. The canonical link for it is here.
Posted to issues@geode.apache.org by "Jinmei Liao (Jira)" <ji...@apache.org> on 2021/10/11 17:47:00 UTC

[jira] [Commented] (GEODE-9613) Investigate the effect of putting re-auth in ClientCacheUpdater in a different thread

    [ https://issues.apache.org/jira/browse/GEODE-9613?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17427278#comment-17427278 ] 

Jinmei Liao commented on GEODE-9613:
------------------------------------

closing this because re-auth needs to be blocking. We won't/shouldn't be receiving any more messages from the server to this client until re-auth succeeded anyway. And if re-auth fails, we should close updater thread anyway.

> Investigate the effect of putting re-auth in ClientCacheUpdater in a different thread
> -------------------------------------------------------------------------------------
>
>                 Key: GEODE-9613
>                 URL: https://issues.apache.org/jira/browse/GEODE-9613
>             Project: Geode
>          Issue Type: Sub-task
>          Components: core
>            Reporter: Jinmei Liao
>            Assignee: Jinmei Liao
>            Priority: Major
>              Labels: GeodeOperationAPI
>
> Currently, when client receives a "re-authenticate" message, it's doing re-authentication in the updater thread, the updater is blocked until the re-auth returned. It seems ok since theoretically the updater won't receive any further message from the server until the re-auth succeeds anyway. But investigate what's the benefit of putting re-auth in another thread.



--
This message was sent by Atlassian Jira
(v8.3.4#803005)