luci-app-rustdesk-server: fixes firmware update causes configuration loss.

[smallprogram]

Fixes #8935
This fixes the issue where configuration settings were lost during a firmware upgrade.
Signed-off-by: David Mandy smallprogramzhusir@gmail.com

Pull request details

Description

fixes firmware update causes configuration loss.

---

Tested on

OpenWrt version: OpenWrt 25.12.2
LuCI version: LuCI openwrt-25.12 branch
Web browser(s): Chrome

---

Checklist

• (Nice to have) Includes what Issue it closes (e.g. openwrt/luci#issue-number).
• (Nice to have) Includes what it depends on (e.g. openwrt/packages#pr-number in sister repo).

[openwrt-ai]

Reviewed 1 new commit; the diagnosis is right — uci-defaults scripts ship in the new rootfs and re-run on the first boot after sysupgrade, so the unguarded batch was resetting enabled/enabled_relay back to 0. Comments inline on the implementation.

---

Generated by Claude Code

[openwrt]

Formality Check: Failed

We checked this pull request against the contribution guidelines. Here is what needs your attention:

🛑 CRITICAL ERRORS

Commit 5c4400c - luci-app-rustdesk-server: fixes firmware update causes config loss:

• No Signed-off-by matches commit author (David Mandy <smallprogramzhusir@gmail.com>). Please add a 'Signed-off-by: David Mandy smallprogramzhusir@gmail.com' line that matches this name and email exactly.

Tip

Do not close this pull request to make corrections. Instead, modify your existing commits (e.g. git commit --amend) and update the branch using git push --force-with-lease --force-if-includes. The checks will re-run automatically.

---

_{Something broken? Consider reporting an issue.
Running version 0e15d1c deployed on 2026-08-12 22:36:32 CEST}

[openwrt-ai]

Reviewed 2 new commits. The touch removal and the indentation change both address the earlier inline comments, and the guard itself is correct now.

The design question from the previous review is still open, though: the shipped conffile root/etc/config/rustdesk-server already contains the rustdesk-server section with enabled '0' and enabled_relay '0', so @rustdesk-server[0] resolves on a fresh install too and the guarded batch is unreachable on every path. That thread was resolved without a reply — if the script is being kept deliberately rather than deleted, the commit message is the place to say why.

Commit checks

• c33e042 "luci-app-rustdesk-server: fixes firmware update causes config loss" — the subject and body describe the sysupgrade fix, but this commit's diff contains only the re-indentation of the here-doc body; the actual fix landed in bc36bd5. Squashing the series (see inline) makes the message match the change.

---

Generated by Claude Code

[openwrt-ai]

nit: the re-indent in c33e042 moved the here-doc body but left the terminator at column 0, so the block now reads as unbalanced. It still works — <<- strips leading tabs from the delimiter line as well as the body — so this is cosmetic only, but indenting it to match the uci -q batch line it closes is what the rest of the block now does.

Suggested change

	EOF
	EOF

---

Generated by Claude Code

[openwrt-ai]

nit: this guard is now spread over three commits for what is a single 7-line change — bc36bd5 introduces it together with a touch line, ebd811f deletes that touch line again, and c33e042 only re-indents the here-doc body. Neither of the last two stands on its own; they are review fixups to the first.

Squashing all three into one commit would make the series bisectable and would also clear the FormalityCheck / Git & Commits failure, since the offending Update applications/... subject on ebd811f disappears with the commit.

---

Generated by Claude Code

[efahl]

I think if you squash the commits, and add
Fixes: https://github.com/openwrt/luci/issues/8935
to the commit message, you'll have it.

[smallprogram]

I think if you squash the commits, and add Fixes: https://github.com/openwrt/luci/issues/8935 to the commit message, you'll have it.

Resubmitted the PR. #8946