[SCHEDULER-194]  AutoFocus failures are ignored
Type Limitation
Priority Medium
Severity Major
Component AcquireScheduler Script, Dispatch Scheduler Engine
Fixed In Version [3.3] 3.3
Versions Affected [3.0.1, 3.0] 3.0 Hot Fix 1, 3.0
Severity Closed
Resolution Complete
Reported By Bob Denny
Resources Bob Denny
Start Date 11/28/2010

Description 
For both the new idle-time AF and the ones done in the course of running an Observation, failures of AF are not detected. The periodic AF time is advanced in either case.

This is going to require some way to tell, after a run or a dispatcher-initiated AF, wheter that AF succeeded or failed, and only if it succeeded should the periodic AF time be advanced.

Comments
1/10/2011 2:19:53 PM   Bob Denny 
Oops forgot to post engine.cs as a change for this logging. Oh well, closing this issue back down.
1/10/2011 11:17:22 AM   Bob Denny 
Reopened - Sjoberg posted a log that shows the sequencer LastAFTime not being updated. Probably failed AFs but who knows??? Need to check the LastAFTime before and after an attempt to AF, and report in the Engine log if it appears that the AF failed (lastAFTime didn't change).
11/29/2010 1:57:46 PM   Bob Denny 

 

  SVN Comment
 

 

  Author
  rbdenny
 

 

  Repository
  svn+ssh://rbdenny@a2_svn_dc3/home/rbdenny/svn/astro/scheduler
 

 

  SVN Revision
  126
 

 

  Affected files
  /trunk/Help/Advanced.htm (Modified)
/trunk/Help/RelNotes.htm (Modified)
/trunk/Scheduler/ACPSequencer.cs (Modified)
/trunk/Scheduler/Engine.cs (Modified)
/trunk/Scheduler/ISequencer.cs (Modified)
/trunk/Scheduler/SimulatorSequencer.cs (Modified)
/trunk/SchedulerSetup/SchedulerSetup.vdproj (Modified)

 

 

  Check-in comment
  Periodic AF now watches for AF failures and extends from the last successful AF. ALso, a non Gemini issue, fix the Day/Night mode do it opens and does periodic AF.
GEM:194
11/28/2010 8:36:32 PM   Bob Denny 
OK, linked issue ACP-519 has LastAfTime.txt with the Julian date of the last successful AF.
11/28/2010 8:05:10 PM   Bob Denny 
This is going to be a bit difficult, especially for Observations with explicit AF specified. It's going to require some sort of "last successful AF" file that can be read when checking if it's time for AF.






The right place for this is in AcquireSupport... and it should go in for ACP 6! See grouped GEM issue. The ACP sequencer will have a new property LastAfTime which will directly read the file. Then use this at the time you make the decision on when to do the next periodic AF. It should actually simplify things in the dispatcher's logic. 







Must implement the LastAFTime property in the Simulator Sequencer too!