Документ взят из кэша поисковой машины. Адрес оригинального документа : http://www.naic.edu/~astro/spectral_line/handbook/app/node6.html
Дата изменения: Fri Feb 14 18:29:01 2003
Дата индексирования: Sun Dec 23 03:44:06 2007
Кодировка:
Поисковые слова: martian surface

TROUBLE SHOOTING AT THE TELESCOPE

• At some point you realize that data is not (or is no longer) being written to the disk This can be caused by two things:
  • You have inadvertently turned off the disk output. This can happen a number of ways, including by jamming up the network lines by trying to move your (very large) data files across the network while observing.// To fix, just type "disc on". Then type "disk status" to be sure that you were successful.//
  • The disk is full. This shouldn't happen, but it does. To see if this is your problem, go to some other window on observer and type "df -k /export/obs8". If the capacity is listed as 100To fix, first double (and triple) check that this is your problem. Once you are convinced it is, you need to get someone from the computer department to help you out. Ask the operator to contact someone for you.//
• Once you star taking data, message "Cannot write to logfile" appears and datataking stops
  This is caused by the fact that you (hopefully) installed everything into your project directory as dtusr while the datataking system, which needs to write to the logfile, is actually "vxw".
  To fix, just go into the SunOS window and change the mode of your log file so that anyone can write to it. That is, type:
  
  chmod a+w a9999.log
• Although you can type on the datataking window it doesn't appear to be responding to your commands.
  This may be caused by you having typed either a square or curly bracket ([ or {) on the datataking window. You can do this without realizing it through a combination of the control key and other characters.
  To fix, just type a whole bunch of closing square brackets (]]]]), hit return, and then hit a bunch of closing curly brackets (}}}}}).
• Datataking is not communicating with the correlator.
  This is probably caused by random network problems.
  To fix, try just exiting the datataking program and restarting it. (Don't forget to move your data first!). If that doesn't work, try logging out of observer (again after moving your data and then exiting vxWorks) and restarting. If you try both of these things a few times and nothing is working, have the operator take a look at the correlator for you and make sure nothing is wrong. As a last resort, exit out of observer and have the operator reboot the correlator.
• Tracking is off - Program starts taking data before it gets on sou rce
  This may be caused by the pointing tolerance being set too high. Look in the upper right corner of the pointing monitor window. The variable "tolA" should be set to 10.0. If it isn't, this is your problem. To fix, go into the "datataking" window and type "pnt tol 10.0" to reset the pointing tolerance to 10 seconds.

If all else fails, reboot.