BRDC Y2K Report

RAVE worked flawlessly at the start of the new millenium.

Unfortunately, a couple of external pieces of software did not. While they are not a part of the RAVE system, the BRDC is dependent on them anyway. What follows is a brief summary of what went wrong and what actions have been taken.
 

Accessing SYNOP data

SYNOP precipitation observations are not a BRDC product but they are needed for the RR product.

Despite a new real-time database, the only dependable way of accessing SYNOP data is through very old routines developed by SMHI and which run under VMS. These routines were secured for Y2K during the autumn of 1999, but they couldn't be built on our new Alphas since the compilers were placed on one machine and the libs on another. I failed when trying to update the routines used by the BRDC and the old routines stopped working on January 1. Since then, I have received assistance in getting these routines working and they are operating as of January 4. No loss of SYNOP data has resulted.
 

BRDC archive

The archive is based on DEC/Compaq Networker software and a 5-tape DLT juke-box. Version 4.4 was being used at the turn of the year and the software commenced writing a completely new archive on January 1. It no longer recognized the previous archive. This means that some data from the BRIDGE Pilot Phase (August 16 - September 20, 1999) have been overwritten since this was the tape in the first slot at the time. In order to fix the problem, the computer's clock had to be set back to 1999 before Networker could be upgraded to version 5.2. The former archive can now be managed along with new data from 2000.
 

Daniel Michelson, 2000-01-11