[
https://issues.apache.org/jira/browse/FLINK-35242?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
yux updated FLINK-35242:
------------------------
Description:
> Update: Changed `fine grained` terminology to avoid confusion between
> fine-grained job resource management feature.
Currently, Flink CDC allows three SE strategy: evolving it, ignoring it, or
throwing an exception. However such configuration strategy doesn't cover all
user cases and requires want more fine-grained strategy configuration.
This ticket suggests adding one more strategy "try_evolve" or
"evolve_when_available". It's basically like "evolving" option, but doesn't
throw an exception if such operation fails, which provides more flexibility.
Also, this ticket suggests allowing user to configure per-schema-event
strategy, so users could evolve some types of event (like rename column) and
reject some unwelcomed events (like truncate table, remove column).
was:
Currently, Flink CDC allows three SE strategy: evolving it, ignoring it, or
throwing an exception. However such configuration strategy doesn't cover all
user cases and requires want more fine-grained strategy configuration.
This ticket suggests adding one more strategy "try_evolve" or
"evolve_when_available". It's basically like "evolving" option, but doesn't
throw an exception if such operation fails, which provides more flexibility.
Also, this ticket suggests allowing user to configure per-schema-event
strategy, so users could evolve some types of event (like rename column) and
reject some dangerous events (like truncate table, remove column).
Summary: Add per-type schema evolution behavior configuration (was:
Add fine-grained schema evolution strategy)
> Add per-type schema evolution behavior configuration
> ----------------------------------------------------
>
> Key: FLINK-35242
> URL: https://issues.apache.org/jira/browse/FLINK-35242
> Project: Flink
> Issue Type: Improvement
> Components: Flink CDC
> Reporter: yux
> Assignee: yux
> Priority: Major
> Labels: pull-request-available
> Fix For: cdc-3.2.0
>
> Original Estimate: 336h
> Remaining Estimate: 336h
>
> > Update: Changed `fine grained` terminology to avoid confusion between
> > fine-grained job resource management feature.
> Currently, Flink CDC allows three SE strategy: evolving it, ignoring it, or
> throwing an exception. However such configuration strategy doesn't cover all
> user cases and requires want more fine-grained strategy configuration.
> This ticket suggests adding one more strategy "try_evolve" or
> "evolve_when_available". It's basically like "evolving" option, but doesn't
> throw an exception if such operation fails, which provides more flexibility.
> Also, this ticket suggests allowing user to configure per-schema-event
> strategy, so users could evolve some types of event (like rename column) and
> reject some unwelcomed events (like truncate table, remove column).
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
