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

Posted to issues@ignite.apache.org by "Maxim Muzafarov (Jira)" <ji...@apache.org> on 2019/10/03 10:03:05 UTC

[jira] [Updated] (IGNITE-10625) Do first checkpoint on node start before join to topology

     [ https://issues.apache.org/jira/browse/IGNITE-10625?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Maxim Muzafarov updated IGNITE-10625:
-------------------------------------
    Fix Version/s:     (was: 2.8)
                   2.9

> Do first checkpoint on node start before join to topology
> ---------------------------------------------------------
>
>                 Key: IGNITE-10625
>                 URL: https://issues.apache.org/jira/browse/IGNITE-10625
>             Project: Ignite
>          Issue Type: Improvement
>          Components: cache
>    Affects Versions: 2.4
>            Reporter: Pavel Kovalenko
>            Priority: Major
>             Fix For: 2.9
>
>
> If a node joins to active cluster we do the first checkpoint during PME when partition states have restored here 
> {code:java}
> org.apache.ignite.internal.processors.cache.distributed.dht.topology.GridDhtPartitionTopology#afterStateRestored 
> {code}
> In IGNITE-9420 we moved logical recovery phase before joining to topology and currently when a node joins to active cluster it already has all recovered partitions. It means that we can safely do the first checkpoint after all logical updates are applied. This change will accelerate PME process if there were a lot of applied updates during recovery.



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