CARMENES cookbook


Errors are shown as follows:

- Message or description
Cause or periodicity, if known

- Ice.ConnectionLostException error=0
Extremely random and frequent
Soved by resetting ICS, sometimes CUs
- Ice.ConnectionLost (Time Out NCU3222)
Extremely random
Soved by resetting ICS and NIR CU
- Err-4: expected count rate too low (or too high ?)
Solved by opening the ExpMeters
- Err-5: unknown error
Seen when selecting SCIENCE mode from SAFE mode
Solved by trying again
- Err-7: no object found and/or AG window doesn't respond (image frozen)
Happens rarely (last time 04-01-2020)
Change to work at pop-up AG windows instead, or reset ICS (see here)
- Err-11: telescope int error
Not sure
Solved by resetting the AG computer
- Nack-4: exposure already running
If exposure is stopped for any reason
Set to Sleep and Awake again, or reset ICS
- Nack-xx: ???????????????? Observation missing by NIR
Device response error popup on the ICS GUI.
Generally, due to GEIRS hangup. NIR cannot be used anymore.
At night, first try nireset-hardreset.
If that is not enough, use nireset-reboot + nireset-hardreset.
- Nack-31: Wrong parameter. Delay time (~start time)
It is due to the NIR Read & Reset count parameter set above 800s
The solution is to lower that number below 800 (it usually is around
half of the ExpTime)
- Lamp state mismatch
Defective lamp
Solved by changing lamp
- Network.cpp:2169
Scheduler opened too early before observations.
Solved by restarting Scheduler + ICS
- Time out AGC 1897 (or 1907)
Happens rarely (last time 18-08-2019)
Solved by stopping guider by hand and re-acquiring obj.
- Time out FFE
Probs in the FE
Solved by resetting the FE controller

- NIR subsystem device error
Solved by performing a sleep-wake up + science

- Broken promise
Occurrence: This situation essentially means that a promise was made to produce a value within the time specified in the “future” parameter, but that promise was not fulfilled before the associated “future” was finished
Promise is a standard C++ object used for signaling if an operation gets completed.
Carmenes example: Sending a signal if the observation gets completed of if the fits file was received correctly.

- Future already retrieved
Ocurrence: This error message occurs when the value of the future is attempted to be retrieved more than once.
This probably can occur in two channel mode, when due to system overload, wrongly handled exceptions or other factors one of the channels finishes the task unexpectedly/significantly faster.
Future is a standard C++ package it. It provides a mechanism for retrieving the result or waiting for it to become available.
Carmenes example: when we start the exposure we have the current time, the length of the exposure and some overheads based on the observing modes. Based on this time values added up we know how long we need to wait for the exposure/readout to finish.
Solved by acknowledging and repeating Observe command

- InvalidFocustepException
Happens when doing the focus sequence, it dosen't see stars.
Because clouds, dome closed, ...

- Pipeline error for LACK of IDL licence (pipeline does not start, not even in earlier nights already reduced).
After restarting Barnard or Lalande it might be that the process which abilitates the license stopsIn the proper computer (Barnard or Lalande).
Use the command: /usr/local/exelis/idl83/bin/lmgrd &