correct benchmark diag_tables for coupled model configurations; move …

…bm_ic directory out of inputdata directory; use aws ec2 for CI test; auto-rt fixes (#426)

* Modifications to run tests on aws ec2 instances
* Change layout numbers. Include all utests (#7)
* Remove unnecessary sleep in build_test.yml and buidl_status_check.py
* Increase sleep period in build_status_check.py to meet 60 requests/hr
* Add sleep back to build_status.check.py
* Minor changes. Change CI to develop branch
* update diag_tables
*add config variable for diag_table
*set up diag_table templates for benchmarks (pre/v16) versions
* fix diag table variable for v16 tests
* add 35 v16 bmark frac test
* add missing diag_table in bmark restart tests
* Move bm_ic directory out of input-data directory
* Replace sys exit with valueError in build_status_check.py
* updates for v16 stability
* remove un-used dz_min=6 parameter setting in bmark v16 tests
* add psm_bc variable and set to value of 1 (default 0) for v16 tests
* set dt_atmos=225 in v16 tests
* updates for dated BM_IC directory
* add INPUTDATA_ROOT_BMIC to rt.sh
* remove BM_IC variable in cpld_bmark_run.IN and use explict path to specify IC sources for ATM,MOM6,CICE5 and WW3 (only for 35d test)
* add variables in parm/input.benchmark_v16.nml.IN
* set dddmp = 0.2 and FSICL to 0 in v16 bmark tests
* modify v16 tests
* keep FSICL at default value for now
* reduce time for v16 wave test to 6hr
* update RTPWD to 20130308; skip-ci
* fix cpld_bmarkfrac_wave_v16 test
* fix timestamps on files for comparison to baseline after switching to 6hour test length
* remove restart file from WW3 comparison. Currently the restart file writes out only at hour 12. This is set in ww3_multi.inp. A change to allow writing a restart at hour 6 would require an update to the ww3 input data directory

Co-authored-by: MinsukJi-NOAA <[email protected]>
Co-authored-by: Brian Curtis <[email protected]>
Co-authored-by: climbfuji <[email protected]>

.github/workflows/main.yml → .github/workflows/build_test.yml

@@ -1,17 +1,16 @@
-name: Pull Request Tests
+name: Pull request tests
 
 on:
  push:
- branches:
- - develop
- pull_request:
- branches:
- - develop
+ branches: ['develop']
+ pull_request_review:
+ types: [submitted]
+ branches: ['develop']
 
 jobs:
  setup:
  name: Set up
- runs-on: ubuntu-latest
+ runs-on: ubuntu-20.04
 
  outputs:
  tn: ${{ steps.parse.outputs.tn }}
@@ -46,7 +45,8 @@ jobs:
  build:
  name: Build (${{ matrix.bld_set }})
  needs: setup
- runs-on: ubuntu-latest
+ if: github.event_name == 'push' || (github.event.review.state == 'approved' && toJson(github.event.pull_request.requested_reviewers) == '[]')
+ runs-on: ubuntu-20.04
 
  strategy:
  fail-fast: false
@@ -85,8 +85,7 @@ jobs:
  utest:
  name: Unit test (${{ needs.setup.outputs.tn }}, ${{ matrix.test_set }})
  needs: [setup,build]
- runs-on: ubuntu-latest
- #runs-on: self-hosted
+ runs-on: self-hosted
 
  strategy:
  fail-fast: false
@@ -101,7 +100,7 @@ jobs:
  - name: Prepare artifacts
  run: |
  tar xvjf artifact.tar.bz2 && rm -f artifact.tar.bz2
- sudo docker load --input ${{ needs.setup.outputs.img }}.tar.gz && rm -f ${{ needs.setup.outputs.img }}.tar.gz
+ docker load --input ${{ needs.setup.outputs.img }}.tar.gz && rm -f ${{ needs.setup.outputs.img }}.tar.gz
 
  - name: Run utest
  run: ./ci.sh -n ${{ needs.setup.outputs.tn }} -r ${{ matrix.test_set }}
@@ -117,5 +116,7 @@ jobs:
  if: ${{ always() }}
  run: |
  rm -f ci.sh ci.test
- sudo docker rm my-container && sudo docker rmi ${{ needs.setup.outputs.img }}:latest
- sudo docker volume rm DataVolume
+ docker stop my-container && docker rm my-container && docker rmi ${{ needs.setup.outputs.img }}:latest
+ docker volume rm DataVolume
+ #docker rmi minsukjinoaa/fv3-input-data:input-data-20210115
+ rm -f memory_stat

.github/workflows/manage.yml → .github/workflows/manage_workflows.yml

@@ -2,21 +2,22 @@ name: Manage workflows
 
 on:
  workflow_run:
- workflows: ["Pull Request Tests"]
+ workflows: ["Pull request tests"]
  types:
  - requested
 
 jobs:
  job1:
- name: Job 1
- runs-on: ubuntu-latest
+ name: Cancel workflows
+ runs-on: ubuntu-20.04
 
  steps:
  - name: Checkout codes
  uses: actions/checkout@v2
 
  - name: Check if skip-ci is requested
  run: |
+ sleep 40
  cd ${GITHUB_WORKSPACE}/tests/ci
  repo="${GITHUB_API_URL}/repos/${GITHUB_REPOSITORY}/actions/runs"
  tr_id=$(cat ${GITHUB_EVENT_PATH} | ./json_helper.py get_trigger_id)

.github/workflows/start_runners.yml

@@ -0,0 +1,33 @@
+name: Start runners
+
+on:
+ workflow_run:
+ workflows: ["Pull request tests"]
+ types:
+ - requested
+
+jobs:
+ job1:
+ name: Start AWS runners
+ runs-on: ubuntu-20.04
+
+ steps:
+ - name: Check out codes
+ uses: actions/checkout@v2
+
+ - name: Configure AWS
+ uses: aws-actions/configure-aws-credentials@v1
+ with:
+ aws-access-key-id: ${{ secrets.AWS_ACCESS_KEY_ID }}
+ aws-secret-access-key: ${{ secrets.AWS_SECRET_ACCESS_KEY }}
+ aws-region: us-east-1
+
+ - name: Check the status of build and start self-hosted runners
+ env:
+ aws_instance_id: ${{ secrets.AWS_INSTANCE_ID }}
+ run: |
+ cd ${GITHUB_WORKSPACE}/tests/ci
+ conclusion=$(cat ${GITHUB_EVENT_PATH} | ./build_status_check.py)
+ if [[ $conclusion == "success" ]]; then
+ aws ec2 start-instances --instance-ids $aws_instance_id
+ fi

.github/workflows/stop_runners.yml

@@ -0,0 +1,25 @@
+name: Stop runners
+
+on:
+ workflow_run:
+ workflows: ["Pull request tests"]
+ types:
+ - completed
+
+jobs:
+ job1:
+ name: Stop AWS runners
+ runs-on: ubuntu-20.04
+
+ steps:
+ - name: Configure AWS
+ uses: aws-actions/configure-aws-credentials@v1
+ with:
+ aws-access-key-id: ${{ secrets.AWS_ACCESS_KEY_ID }}
+ aws-secret-access-key: ${{ secrets.AWS_SECRET_ACCESS_KEY }}
+ aws-region: us-east-1
+
+ - name: Stop self-hosted runners
+ env:
+ aws_instance_id: ${{ secrets.AWS_INSTANCE_ID }}
+ run: aws ec2 stop-instances --instance-ids $aws_instance_id