[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: problem accessing NCOM aggregation

Charles & Scott,

I had problems just using the old FMRC catalog syntax with the new 4.2
TDS, even though they were supposed to work.

I had to start using the new FMRC specification before things worked
properly again.

-Rich


On Wed, Jan 12, 2011 at 2:37 PM, Charles Carleton
<address@hidden> wrote:
> We reverted versions of THREDDS back to a previous version.  It turns out
> that the 'tau' variable never works properly even when the others do.  We
> still haven't figured out why the newer version had problems.
>
> -----Original Message-----
> From: Rich Signell [mailto:address@hidden]
> Sent: Tuesday, January 11, 2011 1:31 PM
> To: Charles Carleton
> Subject: Re: problem accessing NCOM aggregation
>
> Try bringing up the ncom URL below in toolsUI and see on you can make a
> plot.
>
> -Rich
>
> On Jan 11, 2011, at 14:01, Charles Carleton <address@hidden>
> wrote:
>
>> Tomcat has been restarted.  Can you send me the request that provoked that
>> error?
>>
>> -----Original Message-----
>> From: address@hidden [mailto:address@hidden] On Behalf Of Rich
>> Signell
>> Sent: Tuesday, January 11, 2011 9:34 AM
>> To: charles.carleton
>> Cc: Scott Cross; John Wilkin (IMCS)
>> Subject: Fwd: problem accessing NCOM aggregation
>>
>> Charles,
>>
>> I'm having the same problem that John Wilkin is having with the NCOM
>> aggregated data
>> I'm getting:
>>
>> Inconsistent array length read: 1165128303 != 1914731274
>>
>> which indicates that the cache is screwed up.  Can you restart tomcat
>> or clear the cache and let us know?
>>
>> Thanks,
>> Rich
>>
>>
>> ---------- Forwarded message ----------
>> From: John Wilkin (IMCS) <address@hidden>
>> Date: Tue, Jan 11, 2011 at 10:16 AM
>> Subject: Re: cf-compliance and Hycom
>> To: Rich Signell <address@hidden>
>>
>>
>> Rich,
>>
>> Now that I have that sorted out, code that was working yesterday on
>> ncom is breaking. Can you test a nj_subsetgrid request to
>>
> http://edac-dap.northerngulfinstitute.org/thredds/dodsC/ncom_reg1_agg/NCOM_R
>> egion_1_Aggregation_best.ncd
>>
>> I'm wondering if in updating nj_tbx I've broken something else. But
>> the NGI server was down for a while yesterday, so maybe the problem is
>> there.
>>
>> Thanks, John.
>>
>> --
>> John L Wilkin | address@hidden | ph: +1-609-630-0559
>> web: marine.rutgers.edu/~wilkin | cal: tinyurl.com/jwilkincalendar
>>
>>
>>
>>
>> On Tue, Jan 11, 2011 at 10:03 AM, Rich Signell <address@hidden> wrote:
>>> John,
>>> What?  Something worked?!
>>> Sweet!   Thanks for the positive feedback!
>>> -Rich
>>>
>>> On Tue, Jan 11, 2011 at 9:58 AM, John Wilkin (IMCS) <address@hidden>
>> wrote:
>>>> Rich,
>>>>
>>>> You're awesome!
>>>>
>>>> I had started down the road of writing custom code to wrangle those
>>>> stupid dates in the Hycom file (integers like 20101218) to figure out
>>>> what time index to request, and similarly parse the wacky longitudes
>>>> ...  all very messy.
>>>>
>>>> Your ncml renders the dataset compliant with the nj_tbx routines so I
>>>> can go back to the code that I had already written for ncom. My work
>>>> is done.
>>>>
>>>> Interoperability rocks...
>>>>
>>>> Cheers,  John.
>>>>
>>>> --
>>>> John L Wilkin | address@hidden | ph: +1-609-630-0559
>>>> web: marine.rutgers.edu/~wilkin | cal: tinyurl.com/jwilkincalendar
>>>>
>>>>
>>>>
>>>>
>>>> On Mon, Jan 10, 2011 at 5:34 PM, Rich Signell <address@hidden> wrote:
>>>>> John,
>>>>> Actually, I think technically that code is CF compliant, it's just
>>>>> that there are some longitude values that need to have 360 or 720
>>>>> removed to have them fit between -180 and 180.
>>>>>
>>>>> When I convert the longitude values to be within -180 to 180, it works
>> fine:
>>>>>
>>>>> Try loading this:
>>>>>
>>>>>
>>
> http://geoport.whoi.edu/thredds/fileServer/usgs/data1/rsignell/models/hycom/
>> glb_analysis.ncml
>>>>>
>>>>> In Matlab:
>>>>>
>>>>>
>>
> url='http://geoport.whoi.edu/thredds/fileServer/usgs/data1/rsignell/models/h
>> ycom/glb_analysis.ncml';
>>>>> nc=mDataset(url);
>>>>> [lon,lat]=nj_lonlat(nc,'temperature');
>>>>> [jj,ii]=lonlat2ij(lon,lat,[-71.5 -65.0 39.5 46]);
>>>>> t=nc{'temperature'}(1,1,jj,ii);
>>>>> g=nc{'temperature'}(1,1,jj,ii).grid;
>>>>> pcolorjw(g.lon,g.lat,double(t));colorbar
>>>>> title(datestr(g.time))
>>>>>
>>>>>
>>>>> On Mon, Jan 10, 2011 at 3:52 PM, John Wilkin (IMCS)
>> <address@hidden> wrote:
>>>>>> Eoin, Rich,
>>>>>>
>>>>>> Do I recall there was there an action item at the SURA kickoff meeting
>>>>>> that someone in the CI group would help get
>>>>>> http://tds.hycom.org/thredds/dodsC/glb_analysis.html
>>>>>> into CF compliance?
>>>>>>
>>>>>> It sure would accelerate things I'm doing to have interoperability of
>>>>>> my code designed for CF servers.
>>>>>>
>>>>>> Thanks, John.
>>>>>>
>>>>>> --
>>>>>> John L Wilkin | address@hidden | ph: +1-609-630-0559
>>>>>> web: marine.rutgers.edu/~wilkin | cal: tinyurl.com/jwilkincalendar
>>>>>>
>>>>>
>>>>>
>>>>>
>>>>> --
>>>>> Dr. Richard P. Signell   (508) 457-2229
>>>>> USGS, 384 Woods Hole Rd.
>>>>> Woods Hole, MA 02543-1598
>>>>>
>>>>
>>>
>>>
>>>
>>> --
>>> Dr. Richard P. Signell   (508) 457-2229
>>> USGS, 384 Woods Hole Rd.
>>> Woods Hole, MA 02543-1598
>>>
>>
>>
>>
>> --
>> Dr. Richard P. Signell   (508) 457-2229
>> USGS, 384 Woods Hole Rd.
>> Woods Hole, MA 02543-1598
>>
>
>



-- 
Rich Signell
81 Queen St
Falmouth, MA