When the project command is run, a pseudo-Miriad dataset (called project) is created parallel to the data, i.e. in $HATDATA/date/project_name/project This directory contains the contents of the observing command file, the created setups, a history log of the scan and loop commands, as well as a file called variables. It is instructive to look at these files and insure that the setup files contain what you expected from your setup commands and follow the history of the observations to insure that sources were observed (or understand why some were not).
The variables file contains information used to keep track of how many times a setup has been observed. If a setup uses the maxobs parameter, there will be an obscnt entry in the variables file for that setup indicating the number of times the setup has been observed so far.
The variables file is used to insure that some setups are only observed maxobs times no matter how many times a script has been run on the same day. Within checker mode, however, this mode would act to confuse the user rather than help them. This is because everytime the user runs the project command to test their script, they are generally testing the script from a natural beginning; not recovering from a terminated script. To this end, the variables file is automatically reset (using resetproject) everytime the project command is run within checker. To override this default, unset the environment variable RESET_VAR prior to running the project command.