fix(meta): fix column binding for fragment ids #16133
Conversation
| let actors: Vec<(ActorId, Vec<ActorId>)> = Actor::find() | ||
| let actors: Vec<(ActorId, ActorUpstreamActors)> = Actor::find() |
There was a problem hiding this comment.
How did you find this is wrong? 🤡
There was a problem hiding this comment.
I was running some tests locally and the meta failed to restart when load_backfill_fragment_ids. 🥵
There was a problem hiding this comment.
Wait, does it means that this is a serious bug and we need to patch it ASAP?
There was a problem hiding this comment.
😄 no worries, these part of codes are not in the release 1.8.
There was a problem hiding this comment.
Yes, this can be triggered whenever any fragment is created (why?), like CREATE TABLE t(), and then restart risingwave.
src/meta/src/controller/fragment.rs
Outdated
| upstream_ids | ||
| .into_inner() | ||
| .into_iter() | ||
| .flat_map(|(_, ids)| ids.into_iter()) | ||
| .collect(), |
There was a problem hiding this comment.
I think we can return the UpstreamActorIds as-is here, and only map once in the caller site
yezizp2012
force-pushed
the
fix/sql-column-binding
branch
from
5ac5d67 to
b0391c8
Compare
Didn't find a way while using the APIs of sea-orm. 🥵 We may only rely on e2e to prevent it. |
I think we can have a test case like simply stop and restart meta... 🤔 |
|
Wait, don't we already have this? |
Waiting for support in Madsim: #15612 |
I hereby agree to the terms of the RisingWave Labs, Inc. Contributor License Agreement.
What's changed and what's your intention?
Fixes from #16090 .
Checklist
./risedev check(or alias,./risedev c)Documentation
Release note
If this PR includes changes that directly affect users or other significant modifications relevant to the community, kindly draft a release note to provide a concise summary of these changes. Please prioritize highlighting the impact these changes will have on users.