Starting at 16:33 UTC July 6 a filled log directory led to sh commands failing on redirecting logs to be created in that directory. A typo in the cleanup routine failed to clean our all the files after the normal 4 days. As a result they have been accumulating since the beginning of the mission.  It is unexpected that the sh failed when it could not redirect anymore.  

All data products that failed have been created and put in the appropriate directories. However, this will mean a failure in the data latency requirements for some of the GMI L1B and GPROF and a few combined products.