storage: synchronize replica destruction on readOnlyCmdMu

[tamird]

This change is

[tamird]

This causes TestReplicaCommandQueueInconsistent to deadlock. I'm investigating.

[tamird]

This ready for a look.

[petermattis]

---

Reviewed 1 of 1 files at r1, 1 of 1 files at r2, 1 of 1 files at r3, 1 of 1 files at r4, 1 of 1 files at r5, 2 of 2 files at r6.
Review status: 2 of 3 files reviewed at latest revision, 1 unresolved discussion, all commit checks successful.

---

pkg/storage/store.go, line 2034 at r7 (raw file):

      log.Fatalf(ctx, "unexpectedly overlapped by %v", placeholder)
  }
  delete(s.mu.replicaPlaceholders, rep.RangeID)

Was there a purpose to the movement of this deletion? Seemed more consistent where it was previously (next to the other map deletions).

---

Comments from Reviewable

[tamird]

Review status: 2 of 3 files reviewed at latest revision, 1 unresolved discussion, all commit checks successful.

---

pkg/storage/store.go, line 2034 at r7 (raw file):

Previously, petermattis (Peter Mattis) wrote…

Was there a purpose to the movement of this deletion? Seemed more consistent where it was previously (next to the other map deletions).

It was to more closely match the bodies of `addPlaceholderLocked` and `removePlaceholderLocked`.

---

Comments from Reviewable