[jira] [Comment Edited] (UNOMI-627) Persist migration status, and recover in case of failure

["Kevan Jahanshahi (Jira)" <ji...@apache.org>]

    [ https://issues.apache.org/jira/browse/UNOMI-627?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17578888#comment-17578888 ] 

Kevan Jahanshahi edited comment on UNOMI-627 at 8/12/22 9:35 AM:
-----------------------------------------------------------------

PR: [https://github.com/apache/unomi/pull/473]

migration steps status are persisted in a JSON file: /data/migration/history.json

it contains each steps already completed, that wont be executed again in case of next attempts using recovery from the history.

in case history.json is found by the migration command, a question will be ask to the user at the beginning of the migration, to choose if you want to recover from last failure or redo the complete migration from the beginning.

This is also configurable to avoid question to be prompted to the console using conf file:

*org.apache.unomi.migration.cfg and property:* 

*recoverFromHistory = true*


was (Author: jkevan):
PR: [https://github.com/apache/unomi/pull/473]

migration status is persisted in a JSON file: /data/migration/history.json

it contains each steps already completed, that wont be executed again in case of next attempts using recovery from the history.

in case history.json is fount, a question will be ask to the user at the beginning of the migration, to choose if you want to recover from last failure or redo the complete migration from the beginning.

This is also configurable to avoid question to be prompted to the console using conf file:

*org.apache.unomi.migration.cfg and property:* 

*recoverFromHistory = true*

> Persist migration status, and recover in case of failure
> --------------------------------------------------------
>
>                 Key: UNOMI-627
>                 URL: https://issues.apache.org/jira/browse/UNOMI-627
>             Project: Apache Unomi
>          Issue Type: Improvement
>    Affects Versions: 2.0.0
>            Reporter: Kevan Jahanshahi
>            Assignee: Kevan Jahanshahi
>            Priority: Major
>
> In case a migration script fail because ES connection close or ES out of memory or any reason.
> IT could be great to have this options:
>  * rollback ?
>  * being able to start back from the failure.
>  * having clear info on current state when a script failed.
> the main idea is to bring more robustness to current migration system.



--
This message was sent by Atlassian Jira
(v8.20.10#820010)