[Ncep.nhc.nco_contacts] TSB Morning Rounds - Mon May 01, 2017

Christopher Mello - NOAA Federal christopher.mello at noaa.gov
Mon May 1 22:32:27 UTC 2017


Jeff,
The data us there now. It appears we identified the problem this evening.
Edex appears to be getting overwhelmed with data failing to decode the
entire contents of the nwps data file. Unlike other models, the nwps model
data resides in one file. For example...The GFS ingest consists seperate
file for each hour.. nwps data is getting dropped.

  So What changed recently  to cause this problem? Goesr ingest...


Ok so how can we relieve the problem....


short term...

1.  I ingest the model data a second time after a 5 minute sleep period.
So the data will have to fail twice to not get ingested...
2. I will give instructions on how to manually ingest the data a third time
from one of the menus Tuesday..

Longer term....We need to address the data flow issues of goes-r and manage
the legacy sbn data flow. We can discuss the options this week.

Regards
Chris







On Mon, May 1, 2017 at 4:48 PM, Jeffrey Lewitsky - NOAA Federal <
jeffrey.lewitsky at noaa.gov> wrote:

> Chris,
>
> How do we run this script?
>
> The 18z run for the Atlantic/NH2 just finished but only the first 24
> hours came into GFE.
>
> Jeff
>
> On 5/1/17, Christopher Mello - NOAA Federal <christopher.mello at noaa.gov>
> wrote:
> > All,
> > *nwps update*. I wrote a script that will re-ingest  the nwps data a
> second
> > time into GFE after the nwps finishes running. This will help ensure all
> > the available  hours gets ingested into GFE...
> >
> > If the hourly data is still missing...then it is likely an upstream issue
> > with the model out side of AWIPS2. We will observe the nwps data in GFE
> > this week to find out.
> >
> > Chris
> >
> > On Mon, May 1, 2017 at 9:52 AM, TSB Admin <nhc.tsbadmin at noaa.gov> wrote:
> >
> >> Here is today's summary of NHC computer operations:
> >> --- Mello
> >>
> >> ----------
> >>
> >> *TAFB:*
> >>
> >>
> >> TAFB:
> >> 1. TAFB reported no 00z Atlantic/NH2 NWPS data ingested into D2D/GFE
> >> beyond 60 hours. This problem continues to randomly occur. Also, no 21z
> >> SREF SBN data ingested into D2D.
> >>
> >> Follow up. The 00Z and 06Z nwps model runs ingested fully overnight
> >> Sunday
> >> for both domains. The problem might be EDEX getting overwhelmed at
> >> certain
> >> times (EDEX congestion might also cause the reported SREF ingest
> >> failure).
> >>
> >> Possible Solution. Mello will create an nwps Desktop icon today.
> Clicking
> >> on this icon will reingest all the current AWIPS2 nwps model data into
> >> GFE.
> >> This script will determine if the problem is an EDEX ingest issue. If
> the
> >> data does not update beyond 60 hours, then the nwps model run is failing
> >> before it gets into AWIPS2.
> >>
> >> 2. Matt Sardi continues to work with NCO to resolve the NAWIPS
> >> performance
> >> issues. Matt identified some X configuration errors on both the
> >> TILEFISH'S
> >> and BALLYHOO's computers. Forecasters reported that TILEFISH was stable
> >> but
> >> sluggish at times over the weekend.
> >> *HSU:*
> >>
> >> HSU:
> >> 1. TSB/NCO will swap out the last old ATCF monitor in HSU this week(The
> >> Pacific Desk).
> >>
> >> 2. NCO, Joe Zajic (HQ) and Monica Bozeman are still troubleshooting
> >> multiple issues with GOES-16 data flow. Expect sporadic data outages
> >> while
> >> developers clean up the software.
> >> *Other items:*
> >>
> >> *NHC RFCs:*
> >>
> >>
> >>
> >
> >
> > --
> > Chris Mello
> > National Hurricane Center
> > TSB
> >
>



-- 
Chris Mello
National Hurricane Center
TSB
-------------- next part --------------
An HTML attachment was scrubbed...
URL: https://www.lstsrv.ncep.noaa.gov/pipermail/ncep.nhc.nco_contacts/attachments/20170501/3868853e/attachment-0001.html 


More information about the Ncep.nhc.nco_contacts mailing list