Merge pull request #125 from CIROH-UA/test_ds_opts

Test datastream cli options.
* Removed mount-s3
* Fixed configure-datastream.py and added test
* Update docs in terraform

.github/workflows/build_test_docker_x86.yaml

@@ -39,7 +39,7 @@ jobs:
         
     - name: Install packages for datastream
       run: |
-        curl -L -O https://s3.amazonaws.com/mountpoint-s3-release/latest/x86_64/mount-s3.deb && curl -L -O https://github.com/lynker-spatial/hfsubsetCLI/releases/download/v1.1.0/hfsubset-v1.1.0-linux_amd64.tar.gz && tar -xzvf hfsubset-v1.1.0-linux_amd64.tar.gz && sudo mv ./hfsubset /usr/bin/hfsubset && sudo apt install ./mount-s3.deb && sudo apt-get install git pip pigz awscli python3.9 -y        
+        curl -L -O https://github.com/lynker-spatial/hfsubsetCLI/releases/download/v1.1.0/hfsubset-v1.1.0-linux_amd64.tar.gz && tar -xzvf hfsubset-v1.1.0-linux_amd64.tar.gz && sudo mv ./hfsubset /usr/bin/hfsubset && sudo apt-get install git pip pigz awscli python3.9 -y        
 
     - name: Build docker containers
       run : |        

.github/workflows/build_test_push_docker_x86.yaml

@@ -33,7 +33,7 @@ jobs:
         
     - name: Install packages for datastream
       run: |
-        curl -L -O https://s3.amazonaws.com/mountpoint-s3-release/latest/x86_64/mount-s3.deb && curl -L -O https://github.com/lynker-spatial/hfsubsetCLI/releases/download/v1.1.0/hfsubset-v1.1.0-linux_amd64.tar.gz && tar -xzvf hfsubset-v1.1.0-linux_amd64.tar.gz && sudo mv ./hfsubset /usr/bin/hfsubset && sudo apt install ./mount-s3.deb && sudo apt-get install git pip pigz awscli python3.9 -y        
+        curl -L -O https://github.com/lynker-spatial/hfsubsetCLI/releases/download/v1.1.0/hfsubset-v1.1.0-linux_amd64.tar.gz && tar -xzvf hfsubset-v1.1.0-linux_amd64.tar.gz && sudo mv ./hfsubset /usr/bin/hfsubset && sudo apt-get install git pip pigz awscli python3.9 -y        
 
     - name: Build docker containers
       run : |        

.github/workflows/test_datastream_options.yml

@@ -0,0 +1,118 @@
+name: Test Datastream Options
+
+on:
+  push:
+    branches:
+      - main
+
+  pull_request:
+    branches:
+        - main  
+
+jobs:
+  test-datastream-options:
+    runs-on: ubuntu-latest
+    steps:
+    - name: Checkout code
+      uses: actions/checkout@v2  
+
+    - name: Set up QEMU
+      uses: docker/setup-qemu-action@v3
+
+    - name: Set up Docker Buildx
+      uses: docker/setup-buildx-action@v3      
+
+    - name: Configure AWS
+      run: |
+        aws configure set aws_access_key_id ${{ secrets.aws_access_key_id }}
+        aws configure set aws_secret_access_key ${{ secrets.aws_secret_access_key }}
+        aws configure set region us-east-1    
+
+    - name: Build docker containers
+      run : |        
+        ./scripts/docker_builds.sh -b        
+        
+    - name: Install packages for datastream
+      run: |
+        curl -L -O https://github.com/lynker-spatial/hfsubsetCLI/releases/download/v1.1.0/hfsubset-v1.1.0-linux_amd64.tar.gz && tar -xzvf hfsubset-v1.1.0-linux_amd64.tar.gz && sudo mv ./hfsubset /usr/bin/hfsubset && sudo apt-get install git pip pigz awscli python3.9 -y        
+
+    - name: Get geopackage from hfsubset
+      run: |
+        hfsubset -w medium_range -s nextgen -v 2.1.1 -l divides,flowlines,network,nexus,forcing-weights,flowpath-attributes,model-attributes -o palisade.gpkg -t hl "Gages-09106150"
+
+    - name: Base test and NWM_RETRO_V3
+      run: |        
+        sudo rm -rf $(pwd)/data/datastream_test
+        ./scripts/stream.sh -s 202006200100 -e 202006200200 -C NWM_RETRO_V3 -d $(pwd)/data/datastream_test -g $(pwd)/palisade.gpkg -R $(pwd)/configs/ngen/realization_sloth_nom_cfe_pet.json
+
+    - name: Cache resource directory
+      run: |
+        mkdir ./data/cache
+        cp -r ./data/datastream_test/datastream-resources ./data/cache
+        cp -r ./data/cache/datastream-resources ./data/cache/datastream-resources-no-forcings
+        cp -r ./data/cache/datastream-resources ./data/cache/datastream-resources-missing
+        sudo rm -rf ./data/cache/datastream-resources-no-forcings/ngen-forcings 
+
+    - name: Resource directory test missing all
+      if: always()
+      run: |
+        sudo rm -rf $(pwd)/data/datastream_test
+        sudo rm -rf ./data/cache/datastream-resources-missing/ngen-forcings 
+        sudo rm -rf ./data/cache/datastream-resources-missing/config/*
+        ./scripts/stream.sh -r ./data/cache/datastream-resources-missing -R $(pwd)/configs/ngen/realization_sloth_nom_cfe_pet.json -s 202006200100 -e 202006200200 -C NWM_RETRO_V3 -d $(pwd)/data/datastream_test -g $(pwd)/palisade.gpkg
+
+    - name: Resource directory test
+      if: always()
+      run: |
+        sudo rm -rf $(pwd)/data/datastream_test
+        ./scripts/stream.sh -r ./data/cache/datastream-resources -s 202006200100 -e 202006200200 -C NWM_RETRO_V3 -d $(pwd)/data/datastream_test
+
+    - name: Forcings sources option test NWM_RETRO_V2
+      if: always()
+      run: |
+        sudo rm -rf $(pwd)/data/datastream_test
+        ./scripts/stream.sh -r ./data/cache/datastream-resources-no-forcings -s 201906200100 -e 201906200200 -C NWM_RETRO_V2 -d $(pwd)/data/datastream_test        
+        
+    
+    - name: Forcings sources option test NWM_OPERATIONAL_V3
+      if: always()
+      run: |
+        sudo rm -rf $(pwd)/data/datastream_test
+        TODAY=$(env TZ=US/Eastern date +'%Y%m%d')
+        ./scripts/stream.sh -r ./data/cache/datastream-resources-no-forcings -s $TODAY"0100" -e $TODAY"0200" -C NWM_OPERATIONAL_V3 -d $(pwd)/data/datastream_test   
+        
+    - name: Forcings sources option test NOMADS_OPERATIONAL
+      if: always()
+      run: |        
+        sudo rm -rf $(pwd)/data/datastream_test
+        TODAY=$(env TZ=US/Eastern date +'%Y%m%d')
+        ./scripts/stream.sh -r ./data/cache/datastream-resources-no-forcings -s $TODAY"0100" -e $TODAY"0200" -C NOMADS_OPERATIONAL -d $(pwd)/data/datastream_test  
+        
+    - name: Test hfsubset options
+      if: always()
+      run: |
+        sudo rm -rf $(pwd)/data/datastream_test
+        ./scripts/stream.sh -s 202006200100 -e 202006200200 -C NWM_RETRO_V3 -d $(pwd)/data/datastream_test -I "Gages-09106150" -i hl -v 2.1.1 -R $(pwd)/configs/ngen/realization_sloth_nom_cfe_pet.json
+
+    - name: S3 write out test
+      if: always()
+      run: |        
+        sudo rm -rf $(pwd)/data/datastream_test
+        ./scripts/stream.sh -s 202006200100 -e 202006200200 -C NWM_RETRO_V3 -d $(pwd)/data/datastream_test -g $(pwd)/palisade.gpkg -R $(pwd)/configs/ngen/realization_sloth_nom_cfe_pet.json -S ngen-datastream -o git_actions_test/
+        aws s3api wait object-exists --bucket ngen-datastream --key git_actions_test/ngen-run.tar.gz   
+        aws s3api delete-object --bucket ngen-datastream --key git_actions_test/ngen-run.tar.gz
+
+    - name: DAILY today test
+      if: always()
+      run: |        
+        sudo rm -rf $(pwd)/data/datastream_test
+        ./scripts/stream.sh -s DAILY -C NWM_OPERATIONAL_V3 -d $(pwd)/data/datastream_test -g $(pwd)/palisade.gpkg -R $(pwd)/configs/ngen/realization_sloth_nom_cfe_pet.json
+
+    - name: DAILY pick day test
+      if: always()
+      run: |        
+        sudo rm -rf $(pwd)/data/datastream_test
+        ./scripts/stream.sh -s DAILY -e 202006200000 -C NWM_RETRO_V3 -d $(pwd)/data/datastream_test -g $(pwd)/palisade.gpkg -R $(pwd)/configs/ngen/realization_sloth_nom_cfe_pet.json       
+        
+        
+  

INSTALL.md

@@ -1,55 +1,11 @@
-# Install Instructions for ngen-datastream
-These steps are provided [scripted](#scripts) or [step-by-step](#step-by-step). Note some steps are specific to either `x86` or `aarch64`
-
-## Prerequisites
-* Linux OS
-* Docker
-* git
-
-These instructions assume launching an instance from a blank Amazon 2023 Linux image. Steps may vary depending on your specific linux distribution.
-
-## Scripted
-1) Clone this repository
-```
-git clone https://github.com/CIROH-UA/ngen-datastream.git
-```
-2) Execute the startup script
-```
-cd ngen-datastream && ./scripts/install.sh
-```
-`aws_configure` if you intend to mount an s3 bucket or reference a bucket in the configuration.
-
-You're ready to run ngen-datastream!
-
-## Step-by-step 
-`$USER=($whomami)`
-
-For `x86`, `PKG_MNGR="dnf"`
-
-For `aarch64`, `PKG_MNGR="yum"`
-
-1) Update package manager and install packages
-```
-sudo $PKG_MNGR update -y
-sudo $PKG_MNGR install git pip python pigz awscli -y
-```
-`x86` : `sudo $PKG_MNGR update -y
-sudo $PKG_MNGR install dnf-plugins-core -y`
-
-2) install packages from internet
-
-`x86` : `curl -L -O https://s3.amazonaws.com/mountpoint-s3-release/latest/x86_64/mount-s3.rpm` 
-
-`aarch64` : `curl -L -O https://s3.amazonaws.com/mountpoint-s3-release/latest/arm64/mount-s3.rpm`
-```
-sudo dnf update -y
-sudo dnf ./mount-s3.rpm
-```
-3) clone this repository
-```
-git clone https://github.com/CIROH-UA/ngen-datastream.git
-```
-`aws_configure` if you intend to mount an s3 bucket or reference a bucket in the configuration.
-
-You're ready to run ngen-datastream!
+To run ngen-datastream, clone this repository onto a linux machine and make sure the packages below are installed.
+
+## Required Packages
+* docker 
+* git 
+* pip 
+* python>=3.9
+* pigz 
+* awscli 
+* [hfsubset](https://github.com/lynker-spatial/hfsubsetCLI)
 

README.md

@@ -30,7 +30,7 @@ or run with cli args
   -f, --NWM_FORCINGS_DIR    <Path to nwm forcings directory> 
   -F, --NGEN_FORCINGS       <Path to ngen forcings directory, tarball, or netcdf> 
   -N, --NGEN_BMI_CONFS      <Path to ngen BMI config directory> 
-  -S, --S3_MOUNT            <Path to mount s3 bucket to>  
+  -S, --S3_BUCKET           <s3 bucket to write output to>
   -o, --S3_PREFIX           <File prefix within s3 mount> 
   -n, --NPROCS              <Process limit> 
   -y, --DRYRUN              <True to skip calculations> 
@@ -63,8 +63,8 @@ To see what's happening in `ngen-datastream` step-by-step, see the [breakdown do
 | NWM_FORCINGS_DIR | Path to local directory containing nwm files. Alternatively, these file could be stored in RESOURCE_DIR as nwm-forcings. |  |
 | NGEN_BMI_CONFS | Path to local directory containing NextGen BMI configuration files. Alternatively, these files could be stored in RESOURCE_DIR under `config/`.  See here for [directory structure](#configuration-directory-ngen-runconfig). |  |
 | NGEN_FORCINGS            | Path to local ngen forcings directory holding ngen forcing csv's or parquet's. Also accepts tarball or netcdf. Alternatively, this file(s) could  be stored in RESOURCE_DIR at `ngen-forcings/`. |  |
-| S3_MOUNT            | Path to mount S3 bucket to. `ngen-datastream` will copy outputs here. |  |
-| S3_PREFIX           | Prefix to prepend to all files when copying to s3 |
+| S3_BUCKET           | AWS S3 Bucket to write output to |  |
+| S3_PREFIX           | Path within S3 bucket to write to |
 | DRYRUN             | Set to "True" to skip all compute steps. |
 | NPROCS              | Maximum number of processes to use in any step of  `ngen-datastream`. Defaults to `nprocs - 2` |  |
 

python_tools/src/python_tools/configure_datastream.py

@@ -151,13 +151,28 @@ def create_confs(args):
     conf = config_class2dict(args)
     realization = args.realization_file
     geo_base = args.gpkg.split('/')[-1]
+
+    if "OPERATIONAL" in args.forcing_source:    
+        retro_or_op = "operational"         
+        if "V3" in args.forcing_source:
+            urlbaseinput = 7
+        if "NOMADS" in args.forcing_source:
+            urlbaseinput = 1                   
+    elif "RETRO" in args.forcing_source:   
+        retro_or_op = "retrospective"          
+        if "V2" in args.forcing_source:
+            urlbaseinput = 1
+        if "V3" in args.forcing_source:
+            urlbaseinput = 4   
 
     if conf['globals']['start_date'] == "DAILY":
         if conf['globals']['end_date'] != "":
             # allows for a "daily" run that is not the current date
             start_date = datetime.strptime(conf['globals']['end_date'],'%Y%m%d%H%M')
         else:
             start_date = datetime.now(tz.timezone('US/Eastern'))
+            retro_or_op="operational"
+            urlbaseinput=7
         today = start_date.replace(hour=1, minute=0, second=0, microsecond=0)
         tomorrow = today + timedelta(hours=23)
 
@@ -168,8 +183,8 @@ def create_confs(args):
         end = tomorrow.strftime('%Y%m%d%H%M')
         start_realization =  today.strftime('%Y-%m-%d %H:%M:%S')
         end_realization =  tomorrow.strftime('%Y-%m-%d %H:%M:%S')
-        nwm_conf = create_conf_nwm(start, end, "operational",7)
-        fp_conf = create_conf_fp(start, end, conf['globals']['nprocs'],args.docker_mount,args.forcing_split_vpu,"operational",geo_base) 
+        nwm_conf = create_conf_nwm(start, end, retro_or_op, urlbaseinput)
+        fp_conf = create_conf_fp(start, end, conf['globals']['nprocs'],args.docker_mount,args.forcing_split_vpu,retro_or_op,geo_base) 
     else: 
         start = conf['globals']['start_date']
         end   = conf['globals']['end_date']
@@ -185,19 +200,6 @@ def create_confs(args):
             nwm_conf = {}
             fp_conf  = create_conf_fp(start, end, conf['globals']['nprocs'], args.docker_mount, args.forcing_split_vpu,retro_or_op,geo_base) 
         else:
-            if "OPERATIONAL" in args.forcing_source:    
-                retro_or_op = "operational"         
-                if "V3" in args.forcing_source:
-                    urlbaseinput = 7
-                if "NOMADS" in args.forcing_source:
-                    urlbaseinput = 1                   
-            elif "RETRO" in args.forcing_source:   
-                retro_or_op = "retrospective"          
-                if "V2" in args.forcing_source:
-                    urlbaseinput = 1
-                if "V3" in args.forcing_source:
-                    urlbaseinput = 4   
-
             nwm_conf = create_conf_nwm(start,end, retro_or_op,urlbaseinput)
             fp_conf  = create_conf_fp(start, end, conf['globals']['nprocs'], args.docker_mount, args.forcing_split_vpu,retro_or_op,geo_base) 
 

python_tools/tests/test_configurer.py

@@ -102,5 +102,33 @@ def test_conf_daily():
     start = datetime.strptime(data['time']['start_time'],"%Y-%m-%d %H:%M:%S")
     assert start.day == datetime.today().day
 
+    with open(CONF_NWM,'r') as fp:
+        data = json.load(fp)   
+
+    assert data['urlbaseinput'] == 7
+
+
+def test_conf_daily_pick():
+    inputs.start_date = "DAILY"
+    inputs.end_date   = "202006200000"
+    create_confs(inputs)
+    assert os.path.exists(CONF_NWM)
+    assert os.path.exists(CONF_FP)
+    assert os.path.exists(CONF_DATASTREAM)
+    assert os.path.exists(REALIZATION_META_USER)   
+    assert os.path.exists(REALIZATION_META_DS)   
+    assert os.path.exists(REALIZATION_RUN)  
+
+    with open(REALIZATION_RUN,'r') as fp:
+        data = json.load(fp) 
+
+    start = datetime.strptime(data['time']['start_time'],"%Y-%m-%d %H:%M:%S")
+    assert start.day == datetime.strptime(inputs.end_date,"%Y%m%d%H%M%S").day  
+
+    with open(CONF_NWM,'r') as fp:
+        data = json.load(fp)
+
+    assert data['urlbaseinput'] == 4  
+
 
 

research_datastream/terraform/ARCHITECTURE.md

@@ -3,7 +3,7 @@ This document provides a technical look at the AWS infrastructure that is create
 
 The AWS State Machine diagram
 
-![statemachine](../docs/images/statemachine.jpg)
+![statemachine](../../docs/images/statemachine.jpg)
 
 Each green box represents a distinct AWS Lambda Function that is responsible for a task such as spawning an instance from an Amazon Machine Image or issuing commands to an instance. Together these lambdas coordinate the task of creating a cloud based host to which a user can issue commands. 
 

research_datastream/terraform/GETTING_STARTED.md

@@ -33,7 +33,7 @@ This will set your user to `root`. Type `exit` to return back to `ec2-user`. The
 
 Second, install required packages
 ```
-curl -L -O https://s3.amazonaws.com/mountpoint-s3-release/latest/arm64/mount-s3.rpm && curl -L -O https://github.com/lynker-spatial/hfsubsetCLI/releases/download/v1.1.0/hfsubset-v1.1.0-linux_arm64.tar.gz && tar -xzvf hfsubset-v1.1.0-linux_arm64.tar.gz && sudo mv ./hfsubset /usr/bin/hfsubset && sudo dnf install ./mount-s3.rpm git pip pigz awscli python -y
+curl -L -O https://github.com/lynker-spatial/hfsubsetCLI/releases/download/v1.1.0/hfsubset-v1.1.0-linux_arm64.tar.gz && tar -xzvf hfsubset-v1.1.0-linux_arm64.tar.gz && sudo mv ./hfsubset /usr/bin/hfsubset && sudo dnf install git pip pigz awscli python -y
 ```
 
 Next, clone this repository

research_datastream/terraform/README.md

@@ -9,21 +9,34 @@ See [here](https://github.com/CIROH-UA/ngen-datastream/tree/main/terraform/ARCHI
 * Terraform
 * Linux
 
-## Building AWS State Machine
+## Build AWS Infrastructure
+Construct AWS State Machine, Lambdas, Policies, and Roles. See [here](https://github.com/CIROH-UA/ngen-datastream/tree/main/terraform/ARCHITECTURE.md) for a more indepth explanation of the infrastrucutre.
 1) Open a terminal, log into AWS account
-2) Customize resource names by editing `variables.tfvars`.
+2) Customize resource names by editing `variables.tfvars`. Names must be unqiue and not correspond to already existing resources.
 3) Build the state machine with Terraform
 ```
 cd terraform
 terraform init
 terraform apply -var-file=./variables.tfvars
 ```
 
-## Execute
+## Execute AWS State Machine
+This command will start the aws state machine, which will start and manage an ec2 instance to run the datastream command. This step will
 ```
 aws stepfunctions start-execution \
 --state-machine-arn arn:aws:states:us-east-1:###:stateMachine:<sm_name> \
 --name <your-execution-name> \
 --input "file://<path-to-execution-json>" \
 --region us-east-1
 ```
+
+## Tear Down AWS Infrastructure
+```
+terraform destroy -var-file=./variables.tfvars
+```
+
+## Partial Success (`terraform apply failure`)
+`terraform apply` will fail if some of the resources already exist with the names defined in `variables.tfvars`. These resources must be either manually destroyed or imported. A script exists [here](https://github.com/CIROH-UA/ngen-datastream/blob/main/research_datastream/terraform/test/import_resources.sh) to automate importing any existing resources. Remove all spaces from variable file if using this script.
+```
+./test/import_resources.sh <path-to-variables.tfvars>
+```

research_datastream/terraform/lambda_functions/streamcommander/lambda_function.py

@@ -39,8 +39,6 @@ def lambda_handler(event, context):
         if "s3_bucket" in ds_options:
             bucket = ds_options["s3_bucket"]
             prefix = ds_options["object_prefix"]
-            event['commands'].append("runuser -l ec2-user -c 'mkdir -p /home/ec2-user/ngen-datastream/data/mount'")
-            event['commands'].append(f"runuser -l ec2-user -c 'mount-s3 {bucket} /home/ec2-user/ngen-datastream/data/mount'")
         nprocs = ds_options["nprocs"]
         start = ds_options["start_time"]
         end = ds_options["end_time"]
@@ -51,7 +49,7 @@ def lambda_handler(event, context):
         subset_type = ds_options["subset_id_type"]
         command_str = f"runuser -l ec2-user -c 'cd /home/ec2-user/ngen-datastream && ./scripts/stream.sh -s {start} -e {end} -C {forcing_source} -I {subset_id} -i {subset_type} -v {hf_version} -d $(pwd)/data/datastream -R {realization} -n {nprocs}"
         if "s3_bucket" in ds_options: 
-            command_str += f" -S $(pwd)/data/mount -o {prefix}"
+            command_str += f" -S {bucket} -o {prefix}"
         command_str += '\''
         event['commands'].append(command_str)