bug: "Failed to fetch filter key extractor tables" causing compaction task to fail until restarting meta node

[hzxa21]

Describe the bug

We see these message flooding the compactor log and compaction tasks keeps failing until meta node restart
"Failed to fetch filter key extractor tables [{35157}. [[35157]] may be removed by meta-service. "

Error message/log

No response

To Reproduce

No response

Expected behavior

No response

How did you deploy RisingWave?

No response

The version of RisingWave

1.5.1

Additional context

No response

[zwang28]

It happened once before. I suspect there is a code path in meta node that a failed/cancelled stream job doesn't notify hummock manager to remove stale table ids.

[hzxa21]

May be related: #15355. Do we know whether this issue can occur without #15355? cc @zwang28

[zwang28]

duplicate of #15144

[zwang28]

May be related: #15355. Do we know whether this issue can occur without #15355? cc @zwang28

There have been quite some changes of meta's stream job management between v1.5.1 when this issue occurred, and latest v1.7.
I'll look into it in v1.7.