The JSON encoding should only be used for debugging purposes! However
this change avoids showing errors in case there are unknown fields in the
JSON payload. This would happen when for example the MQTT Forwarder
and ChirpStack uses a different API version (which in case of Protobuf
would be fine, as long as the major version remains the same).
This removes the device lock that is stored in Redis to avoid Class-A
and Class-B / Class-C overlapping. Instead, it consistently sets the
scheduler_run_after timestamp, which indicates if the Class-B /
Class-C scheduler should consider scheduling downlinks for a particular
device.
This also updates the get_with_class_b_c_queue_items function to make
sure that multiple ChirpStack instances can execute the same query at
the same time, without getting the same set of data (as the device
records are locked for update and then updated with a
scheduler_run_after timestamp in the future).
This crate can be used to build external ChirpStack integrations.
The plan is to move all non-HTTP based integrations to external
repositories. The reason is that these integrations usually require
quite some external dependencies as these rely on their own SDKs.
Historically, the tx power was defined as ERP. The Semtech UDP Packet
Forwarder protocol was never explicit if the "powe" field was in ERP or
EIRP. However, as the antenna gain in the Semtech UDP Packet Forwarder
is documented as dBi and is deducted from the "powe" field, this means
that "powe" is in EIRP.
This commit corrects all values to EIRP, to make sure that if the
antenna gain is properly configured (which should be), that the correct
rf_power is used by the gateway.
This can be useful when using multiple NetIDs (e.g. after a merge of two
networks, or when using multiple NetIDs). Previously, any other NetID
than the configured net_id would be considered roaming.
This provides more flexibility to custom ADR algorithm implementations.
E.g. the device variables can be used to store per-device variables that
might overwrite normal ADR logic.
This implements end-to-end encryption between the end-device and
end-application. The encrypted AppSKey or SessionKeyID is forwarded to
the end-application which should be able to decrypt or request the
AppSKey to decrypt the uplink payload. As well the end-application will
be able to enqueue encrypted application payloads.
Using this mechanism, ChirpStack will never have access to the uplink
and downlink application-payloads.
This solves #282 because the issue was that auto created measurements
did not have a name and without the measurements tab open, the error
was not visible.
Before ChirpStack v4.5.0 forceRender was not used on the Tabs.TabPane,
in which case the validation would not run on unrendered tabs, which
explains why the form could be submitted even with the empty
measurement names.
Closes#282.
