[
https://issues.apache.org/jira/browse/IGNITE-20143?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=18079679#comment-18079679
]
Dmitry Pavlov commented on IGNITE-20143:
----------------------------------------
[~av] [~timonin.maksim], I understand this was an internal API cleanup. But it
still hurts
Please take a look at recovery case from [IGNITE-21899] — what kind of monster
this migration already turned into. We now have recursive deep scans, broken
binary metadata recovery, dumps, auto-repair logic, etc. This already required
significant effort from 4 community contributors just to safely recover old
TeamCity Bot persistent storage from 2022.
Could we please avoid removing old fields/types “blindly” in future
compatibility changes? Even seemingly harmless cleanup in persistent objects
can later require extremely invasive recovery logic for legacy DBs.
For long-lived Ignite persistence, schema compatibility tends to be much more
expensive than the original cleanup itself.
> GridIntList & GridLongList initial cleanup
> ------------------------------------------
>
> Key: IGNITE-20143
> URL: https://issues.apache.org/jira/browse/IGNITE-20143
> Project: Ignite
> Issue Type: Sub-task
> Reporter: Anton Vinogradov
> Assignee: Anton Vinogradov
> Priority: Major
> Fix For: 2.16
>
> Time Spent: 20m
> Remaining Estimate: 0h
>
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
