[
https://issues.apache.org/jira/browse/NIFI-5583?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16644496#comment-16644496
]
ASF GitHub Bot commented on NIFI-5583:
--------------------------------------
Github user yoshiata commented on the issue:
https://github.com/apache/nifi/pull/2997
We need no separate processor.
I added a saparate processor because I was concerned that adding code to
existing processors would make complicated.
I will change to add code to existing processors.
> Make Change Data Capture (CDC) processor for MySQL refer to GTID
> ----------------------------------------------------------------
>
> Key: NIFI-5583
> URL: https://issues.apache.org/jira/browse/NIFI-5583
> Project: Apache NiFi
> Issue Type: New Feature
> Components: Extensions
> Affects Versions: 1.7.1
> Reporter: Yoshiaki Takahashi
> Priority: Major
>
> When connecting to a MySQL cluster consisting of multiple hosts, existing CDC
> processor can not automatically switch to other host when the connected host
> goes down.
> The reason is that file names and positions of binary log may be different
> for each host in the same cluster.
> In such a case it is impossible to continue reading files from the same
> position.
> In order to continue reading in such cases, a processor referring to the GTID
> (Global Transaction ID) is necessary.
> GTID is ID that uniquely identifies a transaction, and it is recorded in
> binary log of MySQL.
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
