windy/vps
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

Record W1N-104 CSG auto dual-stack deploy on hass.windy.lan.

Browse Source 
end1 IPv6 is on, wlan0 stays off, and the live custom component is de01914
with ip_family=auto after the IPv4 blackhole.
This commit is contained in:
windyboy
2026-08-14 16:11:03 +08:00
parent fcb76d3d5a
commit 88eaefda33
1 changed files with 22 additions and 17 deletions
Download Patch File Download Diff File Expand all files Collapse all files
hosts/hass.windy.lan.md
+22 -17
View File
@@ -39,7 +39,8 @@ recovery codes in this repository.
| Interface | Address / role |
|---|---|
| `end1` | `192.168.55.11/24`; primary LAN55 address |
| `end1` | IPv4 static `192.168.55.11/24` (gw `.254`, DNS `192.168.66.36`); IPv6 SLAAC `auto` with GUA on `240e:3bd:231:2082::/64`; primary LAN55 NIC |
| `wlan0` | Supervisor **disabled** (verified 2026-08-14, W1N-104); IPv6 remains off on this RTL8821CS radio |
| `wg0` | `10.13.13.2/32`; WireGuard (add-on / integration tunnel) |
| `hassio` / `docker0` | internal HAOS Docker bridges (`172.30.32.0/23`, `172.30.232.0/23`) |
@@ -109,24 +110,28 @@ getent hosts hass.windy.lan
## Local patches (custom components)
**`china_southern_power_grid_stat` IPv4 aiohttp (verified 2026-08-14, W1N-102):**
**`china_southern_power_grid_stat` configurable IP family (verified 2026-08-14, W1N-104):**
`/config/custom_components/china_southern_power_grid_stat` was replaced from
GitHub `windyboy/china_southern_power_grid_stat` `master` (`eb8b174`, W1N-101).
The live client uses HA `async_get_clientsession(hass, family=socket.AF_INET)`
and bounded aiohttp timeouts; it no longer uses blocking `requests`. Backup of
the 2026-02-02 HACS/`requests` tree:
`china_southern_power_grid_stat.bak-20260814-w1n102`. HACS still tracks
`CubicPill/china_southern_power_grid_stat` — a HACS update would overwrite this
fork. This directory is a file copy, not a git clone; do not `git pull` in
place.
GitHub `windyboy/china_southern_power_grid_stat` `master` (`de01914`, W1N-103;
previous live tree was `eb8b174` / W1N-102). The live client uses HA
`async_get_clientsession` with a stored `ip_family` option (`ipv4` / `auto` /
`ipv6`; default `ipv4`). Entry `01KGCQDSZCF523A9X6SV3BZ1B9` is set to
`ip_family: auto`. Backups:
`china_southern_power_grid_stat.bak-20260814-w1n102` (HACS/`requests`) and
`china_southern_power_grid_stat.bak-20260814-w1n104` (`eb8b174`). HACS still
tracks `CubicPill/china_southern_power_grid_stat` — a HACS update would
overwrite this fork. This directory is a file copy, not a git clone; do not
`git pull` in place.
After deploy, Core loaded the new code (15:21:45). `verify_login` failed in ~5s
with `CSGTransportError` (no executor hang). Home PPPoE IPv4 to CSG is currently
blackholed: `curl -4 https://95598.csg.cn` from `gw` times out to
`218.19.148.218`; `curl -6` from `gw` returns 200 via `240e:f9:8060::1:16`.
HA `end1`/`wlan0` have `ipv6.method: disabled` (W1N-85) and only `fe80`
addresses, so HA cannot use that IPv6 path. Remote IPv4 to the same A record
still works. Re-apply after any HACS/`CubicPill` update.
Home PPPoE IPv4 to CSG is still blackholed (`curl -4` to `218.19.148.218:443`
times out). `end1` IPv6 is enabled (`ipv6.method: auto`); from HA,
`curl -6 https://95598.csg.cn` returns HTTP 200 via `240e:f9:8060::1:16`.
After this deploy, Core started 16:07:15 CST; first `verify_login` timed out,
reload of the entry then succeeded (`state: loaded` at 16:08:50) and CSG
sensors updated. Some cost/ladder sensors stay `unknown` because CSG's own
`get_month_daily_cost_detail` calendar API returns a marketing-system SQL
error — transport to `95598.csg.cn` is working. Re-apply after any
HACS/`CubicPill` update.
**`tianqi` weather recorder patch (verified 2026-08-13, W1N-75):**
`/config/custom_components/tianqi/weather.py` has a local patch adding