Hour ending   Problem                 Cause
  (UTC)
-------------------------------------------------------------
20130508-   ####: No 6-hourly analysis from OHRFC. 
19Z 20130425- #####
              No hourly analysis from OHRFC.
20130411-20130412: 
              No 6-hourly analysis from MARFC covering 12Z 11th-12Z 12th Apr.
              No hourly analysis from MARFC from 20Z 11th to 00Z 12th.
20130403-20130412: No 6-hourly analysis from OHRFC.  
18Z 20121011- 12Z 20121014: No 6-hourly analysis from NERFC
20120823-20120829:
              Missing 6-hourly analysis from NCRFC
20120515-20120913: 
              Missing 6-hourly analysis from OHRFC
20120229-20120301: 
              Original (data produced in real time) 6h accumulation ending 
              at 18Z 29 Feb, 00/06/12Z 1 Mar, 24h accumulation ending at 
              12Z 1 Mar carry no bitmap, because only 4 RFCs sent in their
              6-hourly analysis files for this time period, among them NWRFC,
              whose GRIB header specified no bitmap.  NWRFC has an RFC id of
              159, and when RFCs that have "higher" id numbers (#160-162) 
              are missing, NWRFC's GRIB header got carried over to the 
              Stage IV.  This problem was later fixed, and Stage IV data 
              obtained from NCAR CODIAC archive for this time period would
              have the problem corrected (i.e. with bitmap indicating large
              areas of missing data, since we only received data from four
              RFCs).

20110502, 12,14,15,16Z: bad hourly analyses from OHRFC.  

2011042412:   Bad analysis from OHRFC.  Three files were received from OHRFC
              for this hour: the first two, received at 12:20Z and 13:21Z, 
              appeared to be OK (max value: 34.875mm), but the last file, 
              received at 19:21Z, has a max value of 86.125mm.

20091002-1003 No hourly analysis during 16Z 2 - 00Z 3 Oct.  No 6-hourly 
              analysis for 6-hour periods ending 18Z 2 Oct, 00/06/12Z 3 Oct.
              No 24h accumulation ending 12Z 20091003.  Reason: data outage,
              did not receive hourly/6-hourly analyses from any RFC for these
              times.  
             
20090318      for the 24h period ending 12Z 20090318, many 6-hourly regional
              analyses from the RFCs were not received, resulting in sparse
              coverage in both 6-hourly and 24-hourly analyses for this period.

20090309-0310 From around 18Z 9 Mar to about 14Z 10 Mar, NCEP did not receive
              incoming regional analyses from the RFCs through TOC, resulting
              in sparse coverage for hourly Stage IV from 11Z 9th to 06Z 10th.
       
20090127      6h and 24h accumulations ending 2009012712 showed over 300mm
              of precipitation in western KY/TN area.  This is likely the
              result of contamination by the heavy ice in this region (thanks
              to Richard Grumm of NWS State College, PA for input).  
20081102-
  20081103:   The following files are missing: 
                 6-hourly accumulations ending 2008110118
                 All 20081102 files
                 6-hourly accumulations ending 2008110300
                 24-hour accumulations ending 2008110312

              Cause of this outage: around 13Z 31 Oct, we began receiving from 
              CBRFC what appeared to be forecasts 2 or 3 days into the future: 
              for example, in 31 Oct, there are files with GRIB header that 
              indicate that they are valid at 60-66 hours (with a reference 
              time of 12Z 31 Oct).   This has caused some problems in our 
              mosaic processing since it was set up to handle 'past' (analysis)
              files, not files with future time stamps.  Turned out CBRFC began
              migrating to a different machine and somehow the forecast files 
              got into the analysis stream sent to us.  On 4 Nov CBRFC 
              discontinued these rogue sends and things went back to normal.  

2004101902-03: Hourly analysis files corrupted, cannot be de-compressed.
20040303,06-12Z: no hourly analysis.  Received no regional analyses from
              the RFCs during this time.
2003082619-2003082718: 
              No hourly analysis 2003082619-2003082700.  Sparse hourly
              coverage 2003082701-2003082718.  06Z analysis ending 
              2003082700 is missing.  06Z analysis ending 2003082618 had 
              input from CBRFC only.  
              Reason: power outage during the evening of Tue 26 Aug.  Even
              after the computers recovered, the outage-related backlogs
              caused disruption in incoming dataflow.  
2003071123-2003071206:
              No hourly analysis.  Major network disruption of over 13 
              hours during the first half (UTC, approx.) of Saturday 12 Jul.  
              No regional hourly analysis received for these hours.  Coverage
              of hourly analysis from 07-12Z are spotty.