Patch.global.sql is not applied to LXD during restart with sudo systemctl reload snap.lxd.daemon

I have problem similar to [Solved] Wait for other cluster nodes to upgrade their versions

I have created patch.global.sql

root@server:/var/snap/lxd/common/lxd/database# ll
total 144
drwx------ 3 root root 4096 Jun 26 21:51 ./
drwx–x–x 17 root root 4096 Jun 26 21:25 …/
drwxr-xr-x 2 root root 4096 Jun 26 21:50 global/
-rw-r–r-- 1 root root 69632 Jun 26 18:06 local.db
-rw-r–r-- 1 root root 61440 Apr 6 2023 local.db.bak
-rw-r–r-- 1 root root 49 Jun 26 21:49 patch.global.sql

but it does not applied - no results in logs:
22:04:13 server systemd[1]: Reloading Service for snap application lxd.daemon.
22:04:13 server systemd[1]: Reloaded Service for snap application lxd.daemon.
22:04:14 server lxd.daemon[129476]: time=“2025-06-26T22:04:14+02:00” level=warning msg=“Dqlite: attempt 1: server ip:8443: dial: Failed connecting to HTTP endpoint “ip:8443”: dial tcp ip:8443: connect: connection refused”

What is wrong/missed?
root@server:/var/snap/lxd/common/lxd/database# cat patch.global.sql
UPDATE nodes SET api_extensions=402 WHERE id=9;

snap lxd version
5.21.3 LTS

Or is there other way to update database permanently ?
Or is it possible to remove entry with ID=9 completely ( since the server has HW outage during cluster LXD update)?

Incus is a continuation/fork of LXD that happened about two years ago. This discussion forum provides support to Incus.