You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
It seems that release 0.9.7 breaks MQTT in a new way. With the new version, enabling MQTT does not cause data to be published on the revpi/io/# channel.
There is an error message in the logs that seems to relate to this, although I have no proof that is actually the case.
2022-05-04 11:48:26 [ERROR ] 'Connect' object has no attribute ‘__evt_wdtoggle’
Downgrading to 0.9.6 resolves the problem that data is not written to MQTT, but obviously reintroduces the problem of not committing data to the I/O ports when using the revpi/set/# channels. I am back to running a patched version.
The object scan to find IOs on core devices uses the `__slot__` attribute, which include very private objects. That objects are not accessible by `getattr` - Now we are using `dir()`.
Missing objects from parent classes.
Ignored the export flag from piCtory.
closing #9
It seems that release 0.9.7 breaks MQTT in a new way. With the new version, enabling MQTT does not cause data to be published on the
revpi/io/#
channel.There is an error message in the logs that seems to relate to this, although I have no proof that is actually the case.
Downgrading to 0.9.6 resolves the problem that data is not written to MQTT, but obviously reintroduces the problem of not committing data to the I/O ports when using the
revpi/set/#
channels. I am back to running a patched version.Please see https://revolutionpi.com/forum/viewtopic.php?t=3409 for the report from user GardG, who spotted the problem first.
The text was updated successfully, but these errors were encountered: