Data archiving

In the current set-up the prefix for the data files which identifies the date on which the data are taken changes automatically at noon UT each day. When the data archiving is done the system will write all the data with a given prefix (in principle that for the previous night) to DVDs (one copy for the observers and one for our own archive). After the data are burned, the files that were saved are moved to subdirectories with the corresponding instrument specific prefix. If the archiving is done before noon UT (which is normally the case) any data data taken later will not be included in the data that is being burned. A side effect is also that if the data has been moved after the burning, any (re)start of the data acquisition program before noon can generate file names that already exist. In general the data taken during the day are test data, but also those might be worthwhile to save while it would also mean that one has to keep track of which data is worth saving and which aren't (which is at least subjective, if not dangerous).

The way the system should work is that the data prefix is changed before the files are selected for burning. One way would be to only do the data archiving after noon UT, but as this is often done in the morning (sometimes because the observers want their copy early) it would be best if the prefix is at the moment the data archiving process is started. The prefix is set in the detector control program, but as we are going to get a new data data storage system for use with the new data acquisition program and detector controllers it seems better to make any specific changes only in the new software but we are looking for a way how to solve this in the mean time.