bug: recover_backfill_state panic

[zwang28]

Describe the bug

The stream job being backfilled is SINK INTO TABLE. It includes NOW.

"panicInfo": "panicked at src/stream/src/executor/backfill/no_shuffle_backfill.rs:607:13:\nassertion left == right failed\n left: true\n right: false"

The compute node has been scaled down from 6C to 2C. I suspect it's related because the assertion mentions: All vnode partitions should have same state (no scale-in supported).

Error message/log

No response

To Reproduce

No response

Expected behavior

No response

How did you deploy RisingWave?

No response

The version of RisingWave

v2.0.2

Additional context

No response

[zwang28]

Currently scaling is not supported for background backfilling job

So this is an expected error.
It will be better if we don't have to drop the affected sinks to recover the cluster. @kwannoel

[kwannoel]

Opened an issue here #19142 for the corresponding feature. Closing as this is not a bug.