You are viewing a plain text version of this content. The canonical link for it is here.
Posted to issues@ignite.apache.org by "Atri Sharma (JIRA)" <ji...@apache.org> on 2015/06/03 07:42:49 UTC

[jira] [Commented] (IGNITE-979) Revisit timeout logic in Ignite.

    [ https://issues.apache.org/jira/browse/IGNITE-979?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14570306#comment-14570306 ] 

Atri Sharma commented on IGNITE-979:
------------------------------------

For locks, can we have something like exponential backoff?

> Revisit timeout logic in Ignite.
> --------------------------------
>
>                 Key: IGNITE-979
>                 URL: https://issues.apache.org/jira/browse/IGNITE-979
>             Project: Ignite
>          Issue Type: Task
>          Components: general
>    Affects Versions: sprint-4
>            Reporter: Vladimir Ozerov
>            Priority: Critical
>             Fix For: sprint-6
>
>
> Migrated from GG private JIRA (GG-9487).
> "Also think of removing registration of events in timeout processor, but have several threads scanning the corresponding futures collection and cancel operations if necessary.
> List to check:
> - tasks
> - cache ops (atomic and tx modes):
> - get
> - put
> - query
> - lock
> - service deployment
> - continuos queries
> - message listeners deployment
> - ?"



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)