Run a self test on slurm (#2)

* Run a self test on slurm

* Use apt to install ansible

* Fix lint error

* Fix slurm install issues

* Enable debugging options

* Better debugging

* Try set StorageType

* Type in config variable

* Fix pid file

* Add basic miniwdl config for slurm

* Install singularity too

* Use different singularity role

* Correct position for config flag

* Fix typo in path name

* Simpler playbook

* Include a partition

* Correct accounting storage

* Add testing configuration in vagrant

* Update CI playbook

* remove directory

* Run on latest ubuntu

* Increase verbosity

* Check slurm logs

* Remove verbosity flags

* Set downloads to 1 core to enable scheduling

* Update memory and cpus

* Don't fail on slurm logs

* Configure realnode

* Use a number for realmemory

* Check slurm logs after miniwdl run

* Check on failure

* Fix broken or

* Adapt real memory to be correct

* Add QC test and test coverage

* Add missing command

* Use correct module name

* Pin version for SLURM role

* Install aptainer via deb package to avoid costly compile of singularity

* Update name to reflect task

.github/workflows/ci.yml

@@ -24,3 +24,38 @@ jobs:
         run: pip install tox
       - name: Lint
         run: tox -e lint
+
+  test:
+    runs-on: ubuntu-22.04
+    steps:
+      - uses: actions/checkout@v2.3.4
+        with:
+          submodules: recursive
+      - name: Set up Python 3.7
+        uses: actions/setup-python@v2
+        with:
+          python-version: 3.7
+      - name: Install miniwdl-slurm
+        run: pip install . coverage
+      - name: install ansible
+        run: sudo apt-get install ansible
+      - name: download slurm role
+        run: sudo ansible-galaxy install galaxyproject.slurm,1.0.1
+      - name: install slurm and singularity on this node
+        run: sudo ansible-playbook tests/slurm_playbook.yml -vv
+      - name: Check slurm logs
+        run: "sudo bash -c 'cat /var/log/slurm/* || true'"
+      - name: run miniwdl self test
+        run: miniwdl run-self-test --cfg tests/miniwdl.cfg
+        timeout-minutes: 3
+      - name: Check slurm logs
+        run: "sudo bash -c 'cat /var/log/slurm/* || true'"
+        if: ${{ failure() }}
+      - name: Run QC workflow
+        run: >- 
+          coverage run --source=miniwdl_slurm -m WDL run 
+          -i tests/integration/single_end.json 
+          --cfg ../miniwdl.cfg 
+          QC.wdl Cutadapt.cores=1 Cutadapt.timeMinutes=10
+        working-directory: "tests/QC"
+      - uses: codecov/codecov-action@v3

.gitmodules

@@ -0,0 +1,3 @@
+[submodule "tests/QC"]
+	path = tests/QC
+	url = https://github.com/biowdl/QC.git

setup.py

@@ -44,7 +44,7 @@
             "slurm_singularity=miniwdl_slurm:SlurmSingularity"
         ],
     },
-    classifiers= [
+    classifiers=[
         "Development Status :: 3 - Alpha",
         "License :: OSI Approved :: MIT License",
         "Programming Language :: Python :: 3 :: Only",

tests/Vagrantfile

@@ -0,0 +1,16 @@
+# -*- mode: ruby -*-
+# vi: set ft=ruby :
+
+Vagrant.configure("2") do |config|
+  config.vm.box = "debian/bullseye64"
+
+  config.vm.hostname = "vagrantslurm"
+  config.vm.provider "libvirt" do |lv|
+      lv.memory = 8192
+      lv.cpus = 4
+  end
+
+  config.vm.provision "ansible" do |ansible|
+    ansible.playbook = "vagrant_playbook.yml"
+  end
+end

tests/miniwdl.cfg

@@ -0,0 +1,19 @@
+[scheduler]
+container_backend=slurm_singularity
+task_concurrency=200
+
+[singularity]
+exe = ["apptainer"]
+
+run_options = [
+        "--containall"
+    ]
+
+image_cache = "$PWD/miniwdl_singularity_cache"
+
+[task_runtime]
+# Set downloads to 1 core to make sure github actions can schedule the job.
+download_defaults = {
+        "cpu": 1,
+        "memory": "1G"
+    }

tests/slurm_playbook.yml

@@ -0,0 +1,40 @@
+- name: Slurm all in One
+  hosts: localhost
+  become: true
+  vars:
+    slurm_config_dir: "/etc/slurm"
+    slurm_roles: ['controller', 'exec', 'dbd']
+    slurmdbd_config:
+      StorageType: "accounting_storage/none"
+      DbdHost: "localhost"
+      PidFile: "run/slurmdbd.pid"
+    slurm_config:
+      AccountingStorageType: "accounting_storage/none"
+      SlurmctldLogFile: "/var/log/slurm/slurmctld.log"
+      SlurmdLogFile: "/var/log/slurm/slurmd.log"
+      SlurmctldPidFile: "/run/slurmctld.pid"
+      SlurmdPidFile: "/run/slurmd.pid"
+      StateSaveLocation: "/var/lib/slurm/slurmctld"
+      SlurmdSpoolDir: "/var/spool/slurm/slurmd"
+    slurm_nodes:
+      - name: "localhost"
+        CPUs: 1
+        RealMemory: 6144  # in MB. CI nodes have about 7 GB on Github CI.
+    slurm_partitions:
+      - name: local
+        Default: YES
+        MaxTime: UNLIMITED
+        Nodes: "localhost"
+  pre_tasks:
+    - name: Install apptainer
+      apt:
+        update_cache: true
+        cache_valid_time: 36000
+        install_recommends: false
+        deb: https://github.com/apptainer/apptainer/releases/download/v1.0.3/apptainer_1.0.3_amd64.deb
+    - name: create spool directory
+      file:
+        state: directory
+        path: "/var/spool/slurm/slurmd"
+  roles:
+    - role: galaxyproject.slurm

tests/vagrant_playbook.yml

@@ -0,0 +1,40 @@
+- name: Slurm all in One
+  # Vagrant playbook for testing before updating CI playbook.
+  hosts: all
+  become: true
+  vars:
+    slurm_roles: ['controller', 'exec', 'dbd']
+    slurmdbd_config:
+      StorageType: "accounting_storage/none"
+      DbdHost: "localhost"
+    slurm_config:
+      AccountingStorageType: "accounting_storage/none"
+      SlurmctldLogFile: "/var/log/slurm/slurmctld.log"
+      SlurmdLogFile: "/var/log/slurm/slurmd.log"
+      SlurmctldPidFile: "/run/slurmctld.pid"
+      SlurmdPidFile: "/run/slurmd.pid"
+      StateSaveLocation: "/var/lib/slurm/slurmctld"
+      SlurmdSpoolDir: "/var/spool/slurm/slurmd"
+    slurm_nodes:
+      - name: "localhost"
+        CPUs: 1
+        RealMemory: 6144
+    slurm_partitions:
+      - name: local
+        Default: YES
+        MaxTime: UNLIMITED
+        Nodes: "localhost"
+    slurm_config_dir: /etc/slurm
+  pre_tasks:
+    - name: Install apptainer
+      apt:
+        update_cache: true
+        cache_valid_time: 36000
+        install_recommends: false
+        deb: https://github.com/apptainer/apptainer/releases/download/v1.0.3/apptainer_1.0.3_amd64.deb
+    - name: create spool directory
+      file:
+        state: directory
+        path: "/var/spool/slurm/slurmd"
+  roles:
+    - role: galaxyproject.slurm