Open gspetro-NOAA opened 1 year ago
I might suggest fixing TASKTHROTTLE. That is a useful Rocoto option when running extremely large/complicated experiments with many jobs.
@JeffBeck-NOAA the taskthrottle tag exists through the rocoto:
section of the config, so this variable wouldn't apply to Rocoto workflows. Right now 1000 is the default, which can be seen in parm/wflow/default_workflow.yaml
.
To implement a different taskthrottle tag just overwrite that default in your config.yaml:
rocoto:
attrs:
taskthrottle: "10"
@mkavulich Just to be 100% clear, are you saying that TASKTHROTTLE
isn't needed as a separate config_defaults variable because it can already be used/implemented in the rocoto:
section of that file without a separate entry elsewhere?
Description
This issue will be used to track variables that appear in
config_defaults.yaml
but should potentially be removed.CRONTAB_LINE
: This variable does not appear to be configurable. It is set automatically insetup.py
.EXTRN_MDL_SYSBASEDIR_ICS
: This functionality did exist at one point, but it seems to have been removed from all Tier 1 platforms, so now this variable just duplicates the function ofEXTRN_MDL_SOURCE_BASEDIR_ICS
.EXTRN_MDL_SYSBASEDIR_LBCS
: Same as above but for LBCS.TASKTHROTTLE
: Introduced in #508 but doesn't seem to have any functionality. Bug to fix? Or should it be removed?FIXcrtmupp
: Doesn't seem to have any functionality. Came in as part of a GSI analysis task.Solution
Consult with CMs to determine whether these variables should be removed or whether there is reason to keep them in (e.g., they are the subject of future development).
Requirements**
The new code should remove variables that are either (a) unused in the workflow or (b) are not configurable.
Acceptance Criteria (Definition of Done)
Each variable is addressed, and either removed or kept depending on the input of CMs/reviewers.