<html style="direction: ltr;">
<head>
<meta http-equiv="content-type" content="text/html; charset=utf-8">
<style type="text/css">body p { margin-bottom: 0cm; margin-top: 0pt; } </style>
</head>
<body style="direction: ltr;" bidimailui-charset-is-forced="true"
smarttemplateinserted="true" text="#000000" bgcolor="#FFFFFF">
Hello all:<br>
<br>
We are running WRF-Hydro in an operational flood forecasting system.
We download GFS data twice daily from
<a class="moz-txt-link-freetext" href="http://www.ftp.ncep.noaa.gov/data/nccf/com/gfs/prod/">http://www.ftp.ncep.noaa.gov/data/nccf/com/gfs/prod/</a> using wget. In
order to be sure that we are getting the full data files, we first
get the file sizes with the wget options: --spider
--server-response. We save the file sizes before download, then
compare the sizes after download. If there's any differences, we
restart the download.<br>
<br>
This mostly works well. When downloading the 0.5 degree data it
works perfectly. However when getting the 0.25 data we have noticed
that occasionally the sizes reported by the NCEP server appear to
not be the actual file sizes. This mismatch does not occur
consistently, but rather only about once in two weeks or so. When
this happens, then even waiting a few hours, and restarting the
simulation run does not help. Our wget --spider and comparison of
downloaded file sizes fails again. Then at the next GFS cycle
everything goes back to normal.<br>
<br>
Has anyone seen this behavior of wget receiving incorrect file
sizes? Is there some other recommended way to ensure that the
downloaded GFS data are the final, complete files.<br>
<br>
many thanks,<br>
Micha <br>
</body>
</html>