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
Is your feature request related to a problem? Please describe.
The CFE_SB_TransmitMsg() function has a boolean parameter to indicate if the sequence number should be incremented automatically based on the route. This route-based sequence number overrides whatever was in the buffer.
However, this only works on telemetry packets. There is a test that checks specifically if the packet is TLM, and only updates the sequence number if so.
This means that all locally-generated messages of the "CMD" variety - which includes the various TIME messages e.g. 1Hz, and all "SEND_HK" packets generated by SCH - always have a sequence number of 0 in their header.
Describe the solution you'd like
The sequence number should probably be valid on all these - otherwise there is no way to tell if a message was missed.
Additional context
It is not clear why only locally-generated TLM packets would have a valid sequence, but not locally-generated CMD packets.
Note SCH_LAB currently calls CFE_SB_TransmitMsg() with the value set false -- so this would have to change too in order to get valid sequence numbers in SEND_HK packets.
Requester Info
Joseph Hickey, Vantage Systems, Inc.
The text was updated successfully, but these errors were encountered:
Assuming agreement here - its only a one-line change to remove the check which limits this to TLM packets. But my previous PR #1154 affects it so should wait for that one to get merged before fixing this.
Is your feature request related to a problem? Please describe.
The
CFE_SB_TransmitMsg()
function has a boolean parameter to indicate if the sequence number should be incremented automatically based on the route. This route-based sequence number overrides whatever was in the buffer.However, this only works on telemetry packets. There is a test that checks specifically if the packet is TLM, and only updates the sequence number if so.
This means that all locally-generated messages of the "CMD" variety - which includes the various TIME messages e.g. 1Hz, and all "SEND_HK" packets generated by SCH - always have a sequence number of 0 in their header.
Describe the solution you'd like
The sequence number should probably be valid on all these - otherwise there is no way to tell if a message was missed.
Additional context
It is not clear why only locally-generated TLM packets would have a valid sequence, but not locally-generated CMD packets.
Note SCH_LAB currently calls
CFE_SB_TransmitMsg()
with the value setfalse
-- so this would have to change too in order to get valid sequence numbers in SEND_HK packets.Requester Info
Joseph Hickey, Vantage Systems, Inc.
The text was updated successfully, but these errors were encountered: