frp: add UCI service enable switches

[laipeng668]

📦 Package Details

Maintainer: N/A

Description:

Add an enabled option to the UCI init sections for frpc and frps,
allowing the two services to be enabled or disabled independently.

New default configurations set enabled to 0, so both services are
disabled by default on new installations. During configuration migration,
if an existing legacy configuration does not contain an enabled option,
initialize it to 1 to preserve the previous startup behavior. Do not
override explicitly configured user values.

When a service is disabled, skip both TOML generation and procd instance
registration.

Also stop configuration generation earlier when include validation
fails, before emitting any remaining TOML sections.

Align whitespace in frps.config with frpc.config by removing
blank separators between options.

Bump PKG_RELEASE to 2.

Depends on: openwrt/luci#9013

---

🧪 Run Testing Details

• OpenWrt Version: main/snapshot
• OpenWrt Target/Subtarget: qualcommax/aarch64
• OpenWrt Device: ipq6000-360v6

---

✅ Formalities

• I have reviewed the CONTRIBUTING.md file for detailed contributing guidelines.

If your PR contains a patch:

• It can be applied using git am
• It has been refreshed to avoid offsets, fuzzes, etc., using

  make package/<your-package>/refresh V=s

• It is structured in a way that it is potentially upstreamable
  _{(e.g., subject line, commit description, etc.)
  We must try to upstream patches to reduce maintenance burden.}

[openwrt-ai]

Commit checks

• 525eeff "frp: add UCI service enable switches" — "abort startup on include validation errors" reads as new behaviour, but _TOML_ERR raised by _append_conf_file was already caught by the pre-existing [ "$_TOML_ERR" = "0" ] || return 1 after the _emit_conf_section / _emit_http_plugin block, and nothing between the two points has side effects outside $CONF_FILE. The added check only moves the abort a few lines earlier — reword so it does not claim a fix that already existed.

---

Generated by Claude Code

[openwrt-ai]

Reviewed 1 new commit; no new issues found.

---

Generated by Claude Code

[kissingers]

Also I think if can enable and disable anytime, keep defualt no check more good than check, as can turn on anytime, and the defualt setting save resource for these want keep frp feature but not always trun on routers.

This is my first version, default is no check enable, but for the already bulid before version frp router, if change default to uncheck, how keep the before status(before same as now check)? because want keep same default result as before version so I have to change default enable now, but I dont think this is a good way.

If can set defualt disable, but if it is a upgrade router then auto check enable, the new router keep default disable will more good.
immortalwrt/packages@d0d829c

[laipeng668]

Okay, based on everyone’s suggestions, I’ve changed it so that the feature is disabled by default. This change will not affect the enabled status in existing users’ configuration profiles.

[kissingers]

When I use immortalwrt luci https://github.com/immortalwrt/luci , which pull from here, if reset router seting to default, and first time to frpc or frps set pages, if do nothing else but save config, find have 3 default seting deleted, that meanse the defaut reset para not same as first save para, can fix this problem?

[laipeng668]

@kissingers

This is perfectly normal—LuCI and the Packages need to match. Once all the code here has been merged, ImmortalWrt will work properly after syncing with these sources.

Until then, you can only handle it manually yourself, or install the files I compiled.

[openwrt-ai]

Reviewed 1 new commit.

---

Generated by Claude Code

[openwrt-ai]

Reviewed 1 new commit.

---

Generated by Claude Code

[kissingers]

I tested it, there are 2 problem:
1:
default setting problem, but this is a before problem not this PR problem, first time or reset the router and reboot, just at frpc and frps page save config, will 6 different for frpc and 4 different for frps. also this might luci problem but not packages problem.
frpc
uci del frpc.common.authentication_method
uci del frpc.common.protocol
uci del frpc.common.wire_protocol
uci del frpc.common.log_level
uci set frpc.cfg018539.user='root'
uci set frpc.cfg018539.group='root'

frps
uci del frps.common.authentication_method
uci del frps.common.log_level
uci set frps.cfg018539.user='root'
uci set frps.cfg018539.group='root'

2:
if use defualt windows format CR/LF for frpc.init and frps.init, server can not satart up, and logs:
daemon.notice: procd: /etc/rc.d/S99
frpc: /bin/sh: line 0: can't open /etc/rc.common
: no such file
aemon.notice: procd: /etc/rc.d/S99
frps: /bin/sh: line 0: can't open /etc/rc.common
: no such file

but if do nothing but only change to linux format LF. it work ok. only the frpc.ini and frps.ini need change, others need not change

debug:
root@myroute:~# /bin/sh /etc/rc.common
/etc/init.d/frpc status : not foundmon: /etc/init.d/frpc: line 2: : not foundmon: /etc/init.d/frpc: line 5: : not foundmon: /etc/init.d/frpc: line 9: : not foundmon: /etc/init.d/frpc: line 13: } : not foundmon: /etc/init.d/frpc: line 14: : not foundmon: /etc/init.d/frpc: line 17: } : not foundmon: /etc/init.d/frpc: line 18: : not foundmon: /etc/init.d/frpc: line 21: } : not foundmon: /etc/init.d/frpc: line 22: : not foundmon: /etc/init.d/frpc: line 19: { : not foundmmon: /etc/init.d/frpc: line 25: } : not foundmon: /etc/init.d/frpc: line 26: : not foundmon: /etc/init.d/frpc: line 23: { : not foundmon: /etc/init.d/frpc: line 29: } : not foundmon: /etc/init.d/frpc: line 30: /etc/rc.common: local: line 32: not in a function

[openwrt-ai]

Reviewed 1 new commit; no new issues found.

---

Generated by Claude Code

[kissingers]

After change frpc.init and frps.init to linux format LF, no other problem to switch enable and disable the 2 servers, but if keep default, can chanage the checkbox and can keep the checkbox after reboot, but never can startup

[laipeng668]

The second issue has been addressed, and the file format has been corrected to use LF line endings.

The first issue may be a fallback behavior intended for migrating configuration files. Since the input value is the same as the default value, it triggers a UCI warning. We can fix this later when there is time; I’m concerned that fixing it now might introduce other issues.