Difference between revisions of "Staff Priorities - Masha"

Revision as of 17:52, 18 March 2015

CSEP:
- Canterbury experiment: ran inventory of daily forecasts using realtime catalogs, missing:
  - Sebastian's 4 forecasts for 2011/03/27, tried to reprocess, got the same failure. Collected inputs, contacted Sebastian regarding the issue
  - STEPCoulomb for 2011/07/02 had a SEGFAULT, reprocessed OK
- David Harte from NZ reported nightly processing failure, connected to NZTC to investigate processing on their certification and operational servers. Got back to David with solutions.

CSEP:
- Failures on csep-op due to CMT catalog not being available over the weekend, reprocess Oceanic Transform Faults jobs
- Got response from Sebastian, investigate raw catalog, reprocess all forecasts for 2011/03/27 if it's bad catalog
- Update code base for Canterbury image, ran evaluation with RELM tests resetting forecast start time

Blocked: None

Followups: None

@@ Line 3: / Line 3: @@
 === Yesterday ===
 * CSEP:
-** Registered for SSA
+** Canterbury experiment: ran inventory of daily forecasts using realtime catalogs, missing:
-** Configuration files for evaluation of all Canterbury forecast groups
+*** Sebastian's 4 forecasts for 2011/03/27, tried to reprocess, got the same failure. Collected inputs, contacted Sebastian regarding the issue
+*** STEPCoulomb for 2011/07/02 had a SEGFAULT, reprocessed OK
+** David Harte from NZ reported nightly processing failure, connected to NZTC to investigate processing on their certification and operational servers. Got back to David with solutions.
 === Today ===
 * CSEP:
-** Failures on csep-op and csep-cert due to CMT catalog not being available over the weekend, reprocess csep-op jobs
+** Failures on csep-op due to CMT catalog not being available over the weekend, reprocess Oceanic Transform Faults jobs
-** Canterbury 1-day forecasts using realtime catalog finished, investigate couple of failures, reprocess missing forecasts
+** Got response from Sebastian, investigate raw catalog, reprocess all forecasts for 2011/03/27 if it's bad catalog
 ** Update code base for Canterbury image, ran evaluation with RELM tests resetting forecast start time