Troubleshooting
Possible problems with the pipeline
Firstly, if you do have problems when running the pipeline, its probably not your fault! Since the pipeline is currently fully automatic, the chances are that if you encounter an error while running it, it means that either there is a temporary problem with the system (like the server is down) or the data you have selected can not be processed properly with the current version of the pipeline.
The only influence you can have over the pipeline is in what data files you choose for processing. Some of the most common problems (which may be solved by changing your data selection) are:
- You are unauthorised to access the data you have selected. See the Authorisation help page.
- The pipeline will stop processing if your job is taking more than 1 hr. Try choosing a smaller set of files.
- Selecting a set of files with incomplete metadata (for example, no primary calibrator) will mean the data can not be processed.
- The pipeline can only process single-pointing continuum imaging data. All other data will be ignored.
Reporting an error
Please report any errors you encounter by contacting Tara Murphy. This will help us rectify immediate problems (like the service being down) and improve the pipeline and archive for future use. When you report an error, please give the error message you received, and also the names of the FITS files you were trying to reduce when you got the error.
Error messages
If the pipeline processing is successful you will get a message "Data Reduction Completed" in which case you can download your plots and images. If a problem occurs you will get a red error message with a brief description. This section lists descriptions of the error messages you might get when using the pipeline.
- No primary calibrator means that the pipeline was not able to identify a primary calibrator within the data you selected. You might want to go back and check the details of the files you selected before pressing the reduce button. If there is a primary calibrator available, it means the pipeline is not identifying it (which most probably means it is an unusual case). In this case you will have to download and reduce the data yourself.
- No calibrated targets means that there are no target sources in your selection that have a suitable calibrator. You might want to go back and check the details of the files you selected before pressing the reduce button. If you think there is an appropriate calibrator present in the data you have selected, please let us know so we can improve the matching algorithm.
- The data you selected could not be calibrated means that either there was not enought data to obtain a solution, or that the target-calibrator matches made by the pipeline were incorrect. You can check both of these things by looking at the table of matches printed out during processing, and looking at the details of the files you have choosen in the archive query results page.
- No data in your selection can be imaged with the current pipeline. means either that there was not enough metadata to allow imaging, or that the data is not continuum data. In the first case there should be error messages in the metadata table printed in the previous section. In the second case it is probably because the data you have selected is spectral line data, millimetre data or some other type of data that the pipeline can't process. We plan to extend the pipeline to cver these cases in the future.
- You have not selected any files for processing means that you haven't actually highlighted any of the files on the ATOA results page. Go back to the results page and select some files (or use 'select all') before pressing reduce.
- No connection established to service. means either the pipeline service is currently unavailable or there is some problem trying to connect from your location. Contact Tara Murphy to report this and/or try again later.
There are also some internal errors that you can't do anything about directly, however if you report them we can improve the service. These are probably caused by either problems with the metadata, or a bug in the pipeline that your particular case has triggered. Some of the internal errors you may encounter are:
- Filename contains illegal characters
- Could not fetch the required files for processing
- The data could not be checked for its imageability
- The data could not be filtered
- There was a problem preparing the data for processing
- The server could not allocate any file space
- The required data files could not be loaded
- There was an error during the editing process
- The data you selected could not be calibrated
- Error preparing to image data
- The data you selected could not be imaged
- Could not retrieve URL
|
