SunSpec (model 124): refresh battery mode via InOutWRte_RvrtTms watchdog

[andig]

fixes #33037

The battery mode is written once per mode transition, but SunSpec model 124 reverts the storage control settings after InOutWRte_RvrtTms seconds (3600 s on a GEN24), so a hold silently expires mid-session. Instead of disabling the revert timer, the timer now acts as a watchdog: it is set to the watchdog interval when a non-normal mode is engaged and the mode is re-applied at half that interval.

• fronius-gen24, fronius-vertoplus, sunspec-inverter-control: batterymode wrapped in the watchdog plugin, InOutWRte_RvrtTms written on hold/charge/holdcharge
• new required watchdog parameter in seconds, default 60 — the battery returns to normal operation on its own if evcc stops refreshing
• new sunspec/read service so the parameter can be filled with the value the device currently uses

🤖 Generated with Claude Code

[sourcery-ai]

Hey - I've found 1 issue, and left some high level feedback:

• In the fronius-gen24 template you guard the negative maxchargerate with {{ if .maxchargerate }}-{{ .maxchargerate }}{{ end }}, but in fronius-vertoplus and sunspec-inverter-control you use -{{ .maxchargerate }} directly; consider making this consistent so configurations without maxchargerate set don’t generate invalid values.
• The new watchdog sequences for the different batterymode cases are largely duplicated across the three templates; you may want to extract common subsequences (e.g. setting InOutWRte_RvrtTms and StorCtl_Mod) via shared templates or YAML anchors to reduce the chance of future divergence.

Prompt for AI Agents

Please address the comments from this code review:

## Overall Comments
- In the fronius-gen24 template you guard the negative maxchargerate with `{{ if .maxchargerate }}-{{ .maxchargerate }}{{ end }}`, but in fronius-vertoplus and sunspec-inverter-control you use `-{{ .maxchargerate }}` directly; consider making this consistent so configurations without maxchargerate set don’t generate invalid values.
- The new watchdog sequences for the different batterymode cases are largely duplicated across the three templates; you may want to extract common subsequences (e.g. setting `InOutWRte_RvrtTms` and `StorCtl_Mod`) via shared templates or YAML anchors to reduce the chance of future divergence.

## Individual Comments

### Comment 1
<location path="templates/definition/meter/fronius-vertoplus.yaml" line_range="238-206" />
<code_context>
-            uri: {{ joinHostPort .host .port }}
-            id: 1
-            value: 124:0:OutWRte
-    - case: 3 # charge
-      set:
-        source: sequence
+          # device reverts to normal unless refreshed by the watchdog
+          - source: const
+            value: {{ durationSeconds .watchdog }} # s
+            set:
</code_context>
<issue_to_address>
**suggestion (bug_risk):** Charge mode depends on maxchargerate without guarding against it being unset or zero

Here `value: -{{ .maxchargerate }} # %` is used directly in case `3`. If `.maxchargerate` is unset or zero, this will produce an invalid/ineffective write to `OutWRte`. Other templates (e.g. GEN24) guard with `{{ if .maxchargerate }}`; please add a similar guard or a safe default so misconfigured `maxchargerate` doesn’t cause confusing device behavior.

Suggested implementation:

```
{{- if .maxchargerate }}
          - source: const
            value: -{{ .maxchargerate }} # %
{{- end }}

```

From the partial context, I can’t see the full `switch` and `case: 3 # charge` block. Please ensure that:
1. The `SEARCH` snippet (`- source: const` and `value: -{{ .maxchargerate }} # %`) exactly matches the current code inside `case: 3 # charge` in `templates/definition/meter/fronius-vertoplus.yaml`.
2. The `{{- if .maxchargerate }}` / `{{- end }}` wrapping is aligned with the rest of the templating style in this file (indentation, dash trimming, etc.) consistent with how GEN24 templates handle `maxchargerate`.
3. If the template uses a different structure (e.g., `source: sequence` with nested steps), the `if` guard should wrap only the single step that writes the negative `maxchargerate` (the charge limit) to the device, not the entire sequence.
</issue_to_address>

---

Sourcery is free for open source - if you like our reviews please consider sharing them ✨

• X
• Mastodon
• LinkedIn
• Facebook

_{Help me be more useful! Please click 👍 or 👎 on each comment and I'll use the feedback to improve your reviews.}

[iseeberg79]

Ich glaube laut Spezifikation ist InOutWRte_RvrtTms or RvrtTms in Model 124 optional. Ich bin nicht überzeugt, das es für den generischen sunspec-Meter passt.

Nachfolger wären Modelle 704/713, die man ggf. berücksichtigen müsste?

Haben Andere gen24-Nutzer den Wert einmalig auf 0 oder höher gesetzt bisher, oder ist das eine Anpassung des defaults der Firmware?

Im sunspec-Modell eines Kostal, der ja nicht Teil dieses PR ist, fehlt der Wert trotz notwendiger Watchdog-Steuerung, aber er implementiert auch das Modell 124 (oder die 700er) nicht.

[andig]

Does this in any form have negative impact? If not we could merge for those devices that need it.

[iseeberg79]

I think atm this might have more impact than use?

[andig]

Which impact do you see?

[iseeberg79]

double checked, even if I cannot test, the implementation should be fine. Even on watchdog is defined as zero it should be safe and any sunspec device should not fail on optional registers