Repository navigation
EEBus: persist full-list writes to the remote cache - #34125
Merged
Merged
Conversation
Member
Author
|
@uhl the idea here is to fix that upstream since imho this will be an issue for every other consumer of non-partial writes. Wdyt? |
Rollup andig/eebus-go@f8d6395 = enbility/eebus-go dev (incl. #261) plus the still-open #255. Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
This comment has been minimized.
This comment has been minimized.
Contributor
|
❌ Backport failed, most likely a cherry-pick conflict. See the run logs. |
This comment has been minimized.
This comment has been minimized.
Contributor
|
✅ Opened backport pull request on |
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Fixes #34004. Upstream fix merged as enbility/eebus-go#261
Bumps the eebus-go fork to andig/eebus-go@f8d6395, which is upstream
dev(including the merged fix) plus the still-open enbility/eebus-go#255 that the previous rollup already carried.Devices that announce
loadControlLimitListDataas writeable withoutpartialsupport receive the complete limit list on every write, merged from the locally cached remote data. The remote replaces its list with what was written, but notifies the change only after the result.writeCurrentLimitDatawrites OPEV, awaits the result and then writes OSCEV, so the second write was built from the stale cache and reverted the obligation just written. On the Porsche PMCC/EVSE this leaves the obligation inactive after every write, inoffmode all limits inactive, and the car charges at maximum (charger logic error: disabled but charging). Same in #34066.This surfaced in 0.316.0 because the fork rebase in #33340 pulled in enbility/eebus-go#253, which stops skipping use cases announced with
useCaseAvailable=false. The PMCC announces OSCEV that way, so evcc never wrote OSCEV limits before and the two-write race never happened. Both traces in #34004 and #34066 show the PMCC notify after the OSCEV write with the OPEV ids reverted.The upstream change persists the written list into the remote feature cache right after sending and restores the previous state if the remote rejects the write. This applies to all full-list writes (load control limits and device configuration key values), so any sequence of writes to a device without partial write support is now consistent. #34071 addresses the same symptom by writing both limit sets in a single message; both changes are independent of each other.
🤖 Generated with Claude Code