In the new version thread several people mentioned they got a GFS too old message on completion of their forecast run. I got this on the previous version so there is a problem with GFS. I normally run with McMahon set for GFS, so I tried using Bohler instead but using that option I get no GFS data at all.
So for me Chris’ data is old data, shows as the midnight run from 16th August, but Bohler gives me nothing at all.
Stuart
Ditto symptoms for me too.
I have PM’s Chris McMahon and hope he will find a moment to take a look.
Stuart
For Tom’s information:
I’m have the same symptoms along with a runtime error 52: Bad name or number. Currently using WXSIM version 12.10.
The beta version 12.10.1 beta2 is also displaying the same runtime error. WXSIM then shuts down.
Hope this is of assistance.
I get this error meaage on EACH run now.
Same for me. Getting the error on every run.
Hi Tom,
I'm have the same symptoms along with a runtime error 52: Bad name or number. Currently using WXSIM version 12.10.
Same issue.
Thanks,
John
Don’t know what’s wrong, but will look ASAP. Visiting my dad in hospital this afternoon, so likely to be this evening before I get a chance to look.
Chris thanks for the update, give your Dad our best, if you get a chance later then great but family comes first.
Stuart
I just tried the Bohler download again and it is working now for me here in the UK, so if Chris’ data is old for you try Bohler and see if that works now.
Stuart
I took a quick look. The problem seems to be very slow downloads of the source data. I’ve tried extending the connection timeout to see if I can at least get the files down if not slower than usual. I don’t know if the connection problem is local to the server, at the far end or somewhere in between. That analysis will have to come later on.
Yes, Chris family definitely comes FIRST.
I’m seeing the “GFS 24 hours oldd” or " No Ready or GFS modle data found" on both Chris’s site and Bohler’s site so maybe it’s a NOAA problem.
mslog.txt - This was with Bohler’s data.
Thornhurst
8/17/2013
8:35:00 AM
No READY or GFS model data found
*
8/17/2013
8:41:38 AM
Cheers
MikeyM
This is strange now because I just looked at the data downloaded using Bohler and I can see data which to me looks like ready data and advection data timed at 0600 today 17th August and I got the error saying no data found. So there is something strange with the Bohler data. Having compared the old data form Chris and the new data from Bohler I cant see what is wrong with it yet I still get the error.
I think I need some advice from Tom and lets hope Chris manages to find time later to see if he can solve it.
Stuart
I’m having a hard time getting rid of the runtime error.
After downloading external data and local data using WXSIMATE I choose “import data” and I get a runtime error.
The progress box sticks at 93% (Scanning for READY or GFS). When clicking OK in the runtime error dialog box WXSIM shuts down.
If I choose import local data first then I don’t get the runtime error and when I subsequently choose import data the import is so fast I think it may not be importing anything at all.
Same issue here in Italy 
When I tried downloading the GFS from Bohler it was only 8129 bytes in size when WXSimate was downloading, I’m sure the file should be a lot larger than that.
Chris,
Its probably because i have seen that GFS datadownload has got really unrealible recently. Sometimes it downloads fast and sometimes it even stops halfway and even have missing files. The set of ca 85h what i download can came in 7 minutes but can also take 40 minutes.
From my own downloads i have seen it came unstable about 1-1.5 months ago and have not get any better since that.
// Henkka
OK, I’ve seen the messages and am checking into it!
I just found McMahon data to be old (00Z from 8/16), and Bohler data is up to date (06Z from 8/17), but there’s something about it which WXSIM seems to be rejecting. I suspect both problems have a common source. The first thing I’ll do is figure out why WXSIM doesn’t want to use the Bohler data (it’s well tested and normally works fine, which is why I wonder about a problem upstream).
I’ll post again soon!
Tom
I found the problem, and it certainly seems to be from the source. Out of 61 lines of data, 3 (just 3!) are bad. For example:
+24,1020.4,0.25,100.0,5.2,111.6,169.3,5869.4,17.5,16.4,15.1,13.7,11.7,9.0,5.7,95.4,93.0,93.2,94.7,92.9,94.5,95.5,39.9,86.9
+27,0.0,0.00,0.0,0.0,270.0,0.0,0.0,-273.1,-273.1,-273.1,-273.1,-273.1,-273.1,-273.1,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0
+30,1021.0,2.73,100.0,5.3,93.4,174.6,5871.9,18.0,17.5,16.8,14.1,11.4,8.7,6.3,92.9,93.4,92.1,88.2,85.0,92.6,97.6,81.6,73.7
The bad news is that we are very much at the mercy of NOAA’s servers. The good news is that, with so many people relying on this data (vastly more than just WXSIM users!), I can’t imagine they would go very long without fixing it.
It’s possible I could come up with a way to fill in gaps in data when this happens, but I don’t recall ever seeing this before, and I would hope it would never happen again! Perhaps I could have it use only data up to that point (of the first bad line), but in this case it’s only a day into the forecast.
The 12Z data should be coming on line in the next hour or two, so let’s see if it’s been fixed at that point.
Tom
I’ve not seen as many errors as Henkka, but today it’s really awful. It’s not managed to get 9 files downloaded in 4 hours (full set is 63 files) and not all of those are complete. Unfortunately in my case I only process the the files are completely downloaded so you either get data or nothing at all.