[Ncep.list.rfcmemo] RFC_MEMO_080221
M. Farley
micki.farley at noaa.gov
Thu Feb 21 16:10:53 EST 2008
-------------- next part --------------
February 21, 2008
MEMORANDUM FOR: Distribution E
FROM: Maxine M. Brown
Acting Deputy Director, NCEP Central Operations (NCO)
SUBJECT: Scheduled Operational Changes and Upgrades
Item 153, NAWIPS version 5.11.1 on the CCS, was reattempted on Tuesday, 19 February and withdrawn after problems occurred in the 18Z GFS. A problem was uncovered running a script that used GDPROF to generate metafiles. GDPROF has been corrected and tested in the workstation environment and on the CCS.
Item 194, which corrected a table entry that sets the maximum number of grids allowed in NAWIPS GEMPAK files, was implemented on Tuesday, 19 February and withdrawn as it was unclear if it was causing failures noted above in the 18Z GFS. Items 153 and 194 will be reattempted on Monday, 25 February.
Items 159, 160, 161, 162, 163 and 165, removing all X-server dependencies from the Model Analysis Forecast (MAF) scripts, were postponed until Tuesday, 26 February following implementation of items 153 and 194.
Items 185 and 186 were implemented at 20Z on Thursday, 14 February as emergency changes to correct errors in the HYSPLIT model, and were immediately removed after failing in production. Item 186 was not required as a part of this package and has been withdrawn. An additional change was required to complement item 185 for the HYSPLIT modifications. It is being thoroughly tested with item 185.
Items 195-197, setting up NCWCP testbed DNS, Directory Services and access to the RHN Satellite server were all implemented, but not on the dates scheduled. Each item was implemented one day later than advertised.
__________________________________________________
The following emergency item was implemented this past week:
206. HOURLY RUC - IBM Job JRUC2A_FCST. (Ma, NCO). Script JRUC2A_FCST.sms.prod was modified to change system parameters allowing a full core dump file when a failure occurs. Done at 15Z on 19 February.
__________________________________________________
---- SOFTWARE IMPLEMENTATIONS ----
The following items will be implemented at 12Z on Tuesday, 26 February:
207. SCRIPT - IBM Job JSREF_WRF_POST. (Du, EMC). Script sref_wrfprdgen.sh generates various grids of GRIB data for the SREF WRF members. The script is being modified to add grid 221 at the request of AWC and GSI and to remove existing grid 104 to save disk space since grid 104 is no longer being used.
208. SCRIPT - IBM Job JSREF_RSM_POST. (Du, EMC). Script exsref_rsm_post.sh.sms creates various grids of GRIB data for the SREF RSM members. The script is being modified to add grid 221 at the request of AWC and GSI and to remove existing grid 104 to save disk space since grid 104 is no longer being used.
209. SCRIPT - IBM Job JSREF_ETA_POST. (Du, EMC). Script exsref_eta_post.sh.sms creates various grids of GRIB data for the SREF Eta members. The script is being modified to add grid 221 at the request of AWC and GSI and to remove existing grid 104 to save disk space since grid 104 is no longer being used.
210. DBNET - System NCOSRV. (Freeman, NCO). The DBNET parsing table on NCOSRV is being modified to fetch 20km RUC GRIB2 files from the CCS and send alerts to AWC. These files are required for AWC operations.
211. SCRIPTS - IBM Job JUVI_GEMPAK. (Magee, NCO). Job script JUVI_GEMPAK is being modified to remove X-server dependencies. Script exuvi_gempak.sh.sms is being modified to use software fonts necessitated by the removal of the X-servers, and fix file uv_index.tbl is being changed to correct the black and white color definitions. Daily X-server-related job failures will discontinue so users will have access to the output gif images.
212. GENESIS TRACKER - IBM Job JTRACKER.SMS.PROD. (Lou, EMC). The program is being modified to ensure the storm counter will not exceed the maximum allowed. Occasionally the existing code fails when executing the Canadian Ensemble forecasts model when the storm counter limit is exceeded.
213. SMS - IBM Jobs GFS_PGRB2B_0P5DEG_XX and GFS_POST_ANL_XX, where XX=00,06,12,18. (Cokley, NCO). The SMS definition file is being modified to change the trigger for the start of jobs GFS_PGRB2B_0P5DEG_XX from the beginning to the completion of jobs GFS_POST_ANL_XX. The change will ensure that the master file is available to create the pgrb2b.anl index file.
214. DBNET - Systems CSS, NCOSRV, CF, TGP06/07. (Freeman, NCO. The dbn_admin.sh is being modified to make it more resilient. The script is currently starting daemons that are already running. To correct this, the script is being modified to ensure the ps command, which reports on active processes, creates a non-empty output file before searching the output to determine if daemons are already running.
215. VERF_PRECIP_PCPCONFORM - IBM Job VERF_PRECIP_GETPPT. (Zhu, EMC). This program is being modified to change the GRIB product definition section flag for the originating center from 98 (ECMWF) to 07 (NCEP). The character length for filename variables file1 and file2 are being increased from 80 to 100 to assure filenames are not truncated.
216. QUATERLY UPGRADE PACKAGE - Systems LUNA, HAZE, ZEPHYR, DEW, MIST. (Ward, NCO). This change announces the Q2 Quarterly CCS Operating System upgrade. Many software upgrades, firmware upgrades and vulnerabilities are being addressed. The changes will be deployed on each system as follows: LUNA, HAZE, ZEPHYR, DEW, MIST. Each CCS system will be out of service for 24 hours during the upgrade. The specific dates and times will be advertised in the CCS Message of the Day (MOTD) and in the Schedule of Events. Luna will be attempted on 21 February. If all goes well, HAZE and HPSS upgrades will be implemented at 12Z on 26 February.
217. BIG_BROTHER - Job System NCOSRV. (O'Reilly, NCO). Big Brother monitoring on NCOSRV is being modified to add monitoring of new message queue "MSG_dude-pc.hfo.noaa.gov" for the new HFO backup system. Monitoring of message queue "MSG_public.hfo.noaa.gov" is being deleted as the public system is no longer in service.
218. DBNET - System NCOSRV. (O'Reilly, NCO). DBNet configuration on the NCOSRV cluster is being modified to set up new message queue and daemon for "MSG_dude-pc.hfo.noaa.gov", a new HFO backup DBNet system. The configuration changes will include adding the pushing of DBNet alerts to new machine. Changes also include removing queue and related daemon "MSG_public.hfo.noaa.gov", since this machine is no longer in use by HFO.
219. DBNET - Systems TPG06/07. (O'Reilly, NCO). The parsing.tbl on TGP06/07 is being modified to begin parsing RUC BGRB13 alerts and pulling the data appropriately. This change will give external users access to the native grid 13km RUC data.
220. DBNET - System CCS. (Cooke, NCO). The parsing and scp tables are being modified on the CCS to begin sending the NAM 20KM output in both parallel and production to the NCEP FTP server.
221. WAVE MULTI 1 - IBM Job JWAVE_MULTI_1_POST. (Chawla, EMC). This program is being modified to extend the Global domain out to 90N by interpolating the grid information from the forecast run in the post processing step. This change will allow more wave information in the Northern waters of Alaska for the local WFOs to get better forecast guidance.
222. TSM. (Vargo, NCO). This change Installs the TSM client on OPCPR and configures it for backups.
223. TSM. (Vargo, NCO). This change Installs the TSM client on OPCAR and configures it for backups.
224. TSM. (Vargo, NCO). This change Installs the TSM client on OPCOL and configures it for backups.
225. TSM. (Vargo, NCO). This change Installs the TSM client on OPCPH and configures it for backups.
226. TSM. (Vargo, NCO). This change Installs the TSM client on OPCAH and configures it for backups.
227. TSM. (Vargo, NCO). This change Installs the TSM client on OPCBKUP and configures it for backups.
228. TSM. (Glotfelty, NCO). This change allows TSM clients in the NCWCP testbed to store backup data on dmztsm01. This allows testing of backup and recovery procedures.
229. SnapMirror. (Brunner, NCO). This is the initial installation of the NetApp lifecycle replacement server along with the setup of the Netapp storage system. The system will be connected to the operational and research networks. SnapMirror will be configured to mirror data between the old and new NetApp systems for data replication.
230. Directory Services. (Glotfelty, NCO). This change adds Directory Services to OPC forecaster workstation OPCBKUP to maintain transparency of current functionality.
The following items will be implemented at 12Z on Wednesday, 27 February:
231. BIG_BROTHER - System CCS. (O'Reilly, NCO). This change will remove monitoring of the ingest processes on the DBNet dev nodes of the CCS, "mist_ioddev" and "dew_ioddev". There are only ingest processes on "mist_iodprod" and "dew_iodprod". We are falsely monitoring these on the "dev" machines.
232. COMPUTE FARM. (Ojo, NCO). This change performs a failover test of the HPC and OPC services on the Compute Farm. To prove that clients ssh keys will work after services are failed over between blades.
233. McIDAS. (Ojo, NCO). The latest version of McIDAS is being installed on CPC's compute farm blade.
The following items will be implemented at 12Z on Thursday, 28 February:
234. DISASTER RECOVERY TAPES. (Smith, NCO). Disaster Recovery tapes will be created and transported to NCEP's off-site storage area in Silver Spring.
235. DEVONPROD - System Mist. (Ward, NCO). DevOnProd access on Mist is being suspended for 24 hours. This change will help determine if development users on the Production system are causing the slowdown of operational jobs.
236. DBNet – System VNCOSRV. (Glotfelty, NCO). DBNet is being installed on the NCWCP testbed VNCOSRV system to transfer a subset of model data to the testbed storage system. Test data will be current and functionally equivalent to a subset of the live model data improving the quality of testing in preparation for the move to NCWCP.
__________________________________________________
---- HARDWARE IMPLEMENTATIONS ----
There are no hardware implementations scheduled for next week.
__________________________________________________
---- OPERATIONAL ANNOUNCEMENTS ----
1. There will be an immediate moratorium for all data flow changes from the Central Computer System until the Force10 router upgrade is complete in mid February. This moratorium will impact the staging of data from parallel runs for proposed model upgrades and any outstanding requests for additional data from the current operational suite.
2. Upcoming major changes to NCEP's production suite and details on any product or timing changes associated with upcoming implementations can be viewed at the following website:
http://www.nco.ncep.noaa.gov/pmb/changes/
Test data for each of these proposed changes should be available six weeks prior to implementation. Anyone wishing to receive formal notice of the final pre-implementation testing and the opportunity to provide feedback on the proposed changes can subscribe to the Model Evaluation mailing list via:
https://lstsrv.ncep.noaa.gov/mailman/listinfo/ncep.list.modelevalinfo
3. Maintenance of subscriptions to the Scheduled Operational Changes and Upgrades mailing list can be performed via the following website:
https://lstsrv.ncep.noaa.gov/mailman/listinfo/ncep.list.rfcmemo
-------------- next part --------------
A non-text attachment was scrubbed...
Name: micki_farley.vcf
Type: text/x-vcard
Size: 323 bytes
Desc: not available
Url : https://lstsrv.ncep.noaa.gov/pipermail/ncep.list.rfcmemo/attachments/20080221/bf3c0e35/micki_farley.vcf
More information about the Ncep.list.rfcmemo
mailing list