There are several reasons why a job will not run. The first reason is due to the job resource requirements. It is possible that the cluster is full and you have to wait for available resources (processors, etc.)
It is also possible that the job may have experienced an error in the run script. In which case the status would be "Eqw". You can query a job's status by entering the following:
qstat -explain c -j _Job-ID_
where _Job-ID_ is the Grid Engine job number.