[Ncep.nhc.nco_contacts] TSB Morning Rounds - Tue Sep 19, 2017

TSB Admin nhc.tsbadmin at noaa.gov
Tue Sep 19 13:29:46 UTC 2017

Here is today's summary of NHC computer operations:
--- Craig Mattocks


1. Intermittent dropouts of GOES-E imagery, horizontal black lines and  
blocks of missing imagery since Sunday. Branch Chief Hugh Cobb contacted  
the SDM, who eventually found a small subset of GOES-E/TPC/MERCATOR/[IR| 
SWIR|WV]_4km data where the reported condition was apparent.

IDSB Dataflow's Dustin Sheffler reported that NHC gets this data feed  
locally through the dish on their site, there was likely a processing issue  
with this data set. SAB and ESPC examined the data in N-AWIPS and did not  
see a problem with their imagery. The SDM took a screenshot of the imagery  
and opened a ticket with IDSB Dataflow, requesting assistance with the  
upstream source of the data.

TAFB attempted to retrieve this imagery using the "goes" command on the  
Sheephead workstation but the Python TCL/TK GUI program threw  
errors "TclError: out of stack space (infinite loop?)". It hangs if more  
than one instance of the program is run concurrently (code is not  
thread-safe). Matt Sardi investigated and determined that it does run  
successfully to completion as long as only one instance is run at a time.  
He will investigate whether this problem was introduced by recent  
system/library patches or changes due to the OS upgrade from RHEL 5 to 6,  
since the code hasn't changed since 2013.

Other items:
1. Web issues: Local 'Threats and Impacts' links to display for San Juan  
WFO and ER offices on the NWS Local Products Page not updating. TSB  
Developer Matt Sardi tried to update the Halo server to remedy these  
errors, but it then began restarting every 2 minutes, so he reverted to a  
previous version of the binary executable. This seems to have solved the  

2. Web issue: Text section of Key Messages graphic for Maria did not update  
properly; resolved when the above Halo web server problem was fixed by Matt  
Sardi. No other web graphics were impacted.

3. There has been an ongoing outage of ancillary forecast products  
(non-operational data of opportunity) from FNMOC/NRL in Monterey, CA since  
06 UTC Sept 16. This included DMSP/SSMI/S, WINDSAT and WW3 products. These  
data streams were restored at 2015 UTC yesterday.

4. TSB Developer Matt Onderlinde explained that, when the TC graphics are  
run multiple times ssimultaneously for the same storm, TSB gets strafed  
with unnecessary Graphics Production/web launch error warning messages via  
email; there are no actual problems.

5. Recon aircraft data ingest: 18 dropsonde/flight level reports in the  
vicinity of Hurricane Jose and 16 dropsonde/flight level reports in the  
vicinity of Hurricane Maria were ingested into the GFS model yesterday.

-------------- next part --------------
An HTML attachment was scrubbed...
URL: https://www.lstsrv.ncep.noaa.gov/pipermail/ncep.nhc.nco_contacts/attachments/20170919/bfa4b34b/attachment.html 

More information about the Ncep.nhc.nco_contacts mailing list