Merge pull request 'Reduce SUS FDIR Events' (#806) from sus-fdir-reduce-events into main

Reviewed-on: #806
Reviewed-by: Robin Müller <muellerr@irs.uni-stuttgart.de>

CHANGELOG.md

@@ -52,6 +52,9 @@ will consitute of a breaking change warranting a new major release:
   code.
 - Added a 3 second delay in the EIVE system between commanding all PL components except the SUPV,
   and the SUPV itself OFF when the power level becomes low or critical.
+- SUS FDIR should now trigger less events. The finish event is now only triggered once the
+  SUS has been working properly for a minute again. It will then display the number of periods
+  during which the SUS was not working as well as the maximum amount of invalid messages.
 
 # [v7.1.0] 2023-10-11
 

mission/acs/SusHandler.cpp

@@ -97,7 +97,7 @@ ReturnValue_t SusHandler::interpretDeviceReply(DeviceCommandId_t id, const uint8
   // appear sometimes for the SUS device: Allow invalid message up to a certain threshold
   // before triggering FDIR reactions.
   if (reply->tempRaw == 0xfff and not waitingForRecovery) {
-    if (invalidMsgCounter == 0) {
+    if (invalidMsgCounter == 0 and invalidMsgPeriodCounter == 0) {
       triggerEvent(TEMPERATURE_ALL_ONES_START);
     } else if (invalidMsgCounter == susMax1227::MAX_INVALID_MSG_COUNT) {
       triggerEvent(DeviceHandlerIF::DEVICE_WANTS_HARD_REBOOT);
@@ -110,8 +110,17 @@ ReturnValue_t SusHandler::interpretDeviceReply(DeviceCommandId_t id, const uint8
     return returnvalue::OK;
   }
   if (invalidMsgCounter > 0) {
-    triggerEvent(TEMPERATURE_ALL_ONES_RECOVERY, invalidMsgCounter);
+    invalidMsgPeriodCounter++;
+    if (invalidMsgCounter > invalidMsgCounterMax) {
+      invalidMsgCounterMax = invalidMsgCounter;
+    }
     invalidMsgCounter = 0;
+    invalidMsgCountdown.resetTimer();
+  }
+  if (invalidMsgCountdown.hasTimedOut() and invalidMsgPeriodCounter > 0) {
+    triggerEvent(TEMPERATURE_ALL_ONES_RECOVERY, invalidMsgPeriodCounter, invalidMsgCounterMax);
+    invalidMsgPeriodCounter = 0;
+    invalidMsgCounterMax = 0;
   }
   dataset.setValidity(true, true);
   dataset.tempC = max1227::getTemperature(reply->tempRaw);

mission/acs/SusHandler.h

@@ -20,8 +20,9 @@ class SusHandler : public DeviceHandlerBase {
   //! [EXPORT] : [COMMENT] Detected invalid values, starting invalid message counting
   static constexpr Event TEMPERATURE_ALL_ONES_START =
       event::makeEvent(SUBSYSTEM_ID, 0, severity::MEDIUM);
-  //! [EXPORT] : [COMMENT] Detected valid values again, resetting invalid message counter.
+  //! [EXPORT] : [COMMENT] Detected valid values for a prolonged time again, resetting all counters.
-  //! P1: Invalid message counter.
+  //! P1: Number of periods with invalid messages.
+  //! P2: Maximum invalid message counter.
   static constexpr Event TEMPERATURE_ALL_ONES_RECOVERY =
       event::makeEvent(SUBSYSTEM_ID, 1, severity::INFO);
 
@@ -54,8 +55,13 @@ class SusHandler : public DeviceHandlerBase {
   susMax1227::SusDataset dataset;
   acs::SusRequest request{};
   uint8_t susIdx;
+
+  //  After 1 minute, trigger the event for the invalid messages.
+  Countdown invalidMsgCountdown = Countdown(60000);
   bool waitingForRecovery = true;
   uint32_t invalidMsgCounter = 0;
+  uint32_t invalidMsgCounterMax = 0;
+  uint32_t invalidMsgPeriodCounter = 0;
 
   uint32_t transitionDelay = 1000;
   bool goToNormalMode = false;