From e3473857c1dd9c555b3921d00e520b7903254b73 Mon Sep 17 00:00:00 2001
From: Deezzir <deezzir@gmail.com>
Date: Fri, 6 Sep 2024 11:29:31 -0400
Subject: [PATCH 01/36] Update structure to comply with bootstack template

---
 Makefile       | 55 +++++++++++++++++++++++++++++++++++++++++
 pyproject.toml | 66 ++++++++++++++++++++++++++++++++++++++++++++++++++
 rename.sh      | 17 +++++++++++++
 tox.ini        | 41 +++++++++++++++++++++++++++++++
 4 files changed, 179 insertions(+)
 create mode 100644 Makefile
 create mode 100644 pyproject.toml
 create mode 100755 rename.sh
 create mode 100644 tox.ini

diff --git a/Makefile b/Makefile
new file mode 100644
index 0000000..3394bab
--- /dev/null
+++ b/Makefile
@@ -0,0 +1,55 @@
+# This is a template `Makefile` file for snaps
+# This file is managed by bootstack-charms-spec and should not be modified
+# within individual snap repos. https://launchpad.net/bootstack-charms-spec
+
+PYTHON := /usr/bin/python3
+
+PROJECTPATH=$(dir $(realpath ${MAKEFILE_LIST}))
+SNAP_NAME=$(shell cat ${PROJECTPATH}/snap/snapcraft.yaml | grep -E '^name:' | awk '{print $$2}')
+SNAP_FILE=${PROJECTPATH}/${SNAP_NAME}.snap
+
+help:
+	@echo "This project supports the following targets"
+	@echo ""
+	@echo " make help - show this text"
+	@echo " make clean - remove unneeded files"
+	@echo " make build - build the snap"
+	@echo " make lint - run lint checkers"
+	@echo " make reformat - run lint tools to auto format code"
+	@echo " make unittests - run the tests defined in the unittest subdirectory"
+	@echo " make functional - run the tests defined in the functional subdirectory"
+	@echo " make test - run lint, proof, unittests and functional targets"
+	@echo ""
+
+lint:
+	@echo "Running lint checks"
+	@tox -e lint
+
+unittests:
+	@echo "Running unit tests"
+	@tox -e unit -- ${UNIT_ARGS}
+
+test: lint unittests functional
+	@echo "Tests completed for the snap."
+
+reformat:
+	@echo "Reformat files with black and isort"
+	@tox -e reformat
+
+build:
+	@echo "Building the snap"
+	@snapcraft --use-lxd
+	@bash -c ./rename.sh
+
+clean:
+	@echo "Cleaning snap"
+	@snapcraft clean --use-lxd
+	@echo "Cleaning existing snap builds"
+	@rm -rf ${SNAP_FILE}
+
+functional: build
+	@echo "Executing functional tests using built snap"
+	@TEST_SNAP=${SNAP_FILE} tox -e func -- ${FUNC_ARGS}
+
+# The targets below don't depend on a file
+.PHONY: help clean build lint reformat unittests functional test
\ No newline at end of file
diff --git a/pyproject.toml b/pyproject.toml
new file mode 100644
index 0000000..8af02ae
--- /dev/null
+++ b/pyproject.toml
@@ -0,0 +1,66 @@
+# This is a template `pyproject.toml` file for snaps
+# This file is managed by bootstack-charms-spec and should not be modified
+# within individual snap repos. https://launchpad.net/bootstack-charms-spec
+
+[tool.flake8]
+ignore = ["C901", "D100", "D101", "D102", "D103", "W503", "W504"]
+exclude = ['.eggs', '.git', '.tox', '.venv', '.build', 'build', 'report']
+max-line-length = 99
+max-complexity = 10
+
+[tool.black]
+line-length = 99
+exclude = '''
+/(
+    | .eggs
+    | .git
+    | .tox
+    | .venv
+    | .build
+    | build
+    | report
+)/
+'''
+
+[tool.isort]
+profile = "black"
+skip_glob = [
+  ".eggs",
+  ".git",
+  ".tox",
+  ".venv",
+  ".build",
+  "build",
+  "report"
+]
+
+[tool.pylint]
+max-line-length = 99
+ignore = ['.eggs', '.git', '.tox', '.venv', '.build', 'report', 'tests']
+
+[tool.mypy]
+warn_unused_ignores = true
+warn_unused_configs = true
+warn_unreachable = true
+disallow_untyped_defs = true
+exclude = ['.eggs', '.git', '.tox', '.venv', '.build', 'report', 'tests']
+
+## Ignore unsupported imports
+[[tool.mypy.overrides]]
+ignore_missing_imports = true
+module = ["setuptools"]
+
+[tool.coverage.run]
+relative_files = true
+source = ["."]
+omit = ["tests/**", "docs/**", "lib/**", "snap/**", "build/**", "setup.py"]
+
+[tool.coverage.report]
+fail_under = 100
+show_missing = true
+
+[tool.coverage.html]
+directory = "tests/unit/report/html"
+
+[tool.coverage.xml]
+output = "tests/unit/report/coverage.xml"
\ No newline at end of file
diff --git a/rename.sh b/rename.sh
new file mode 100755
index 0000000..3b14d01
--- /dev/null
+++ b/rename.sh
@@ -0,0 +1,17 @@
+#!/bin/bash
+# This is a template `rename.sh` file for snaps
+# This file is managed by bootstack-charms-spec and should not be modified
+# within individual snap repos. https://launchpad.net/bootstack-charms-spec
+
+snap=$(grep -E "^name:" snap/snapcraft.yaml | awk '{print $2}')
+echo "renaming ${snap}_*.snap to ${snap}.snap"
+echo -n "pwd: "
+pwd
+ls -al
+echo "Removing previous snap if it exists"
+if [[ -e "${snap}.snap" ]];
+then
+    rm "${snap}.snap"
+fi
+echo "Renaming snap here."
+mv ${snap}_*.snap ${snap}.snap
\ No newline at end of file
diff --git a/tox.ini b/tox.ini
new file mode 100644
index 0000000..6827ecb
--- /dev/null
+++ b/tox.ini
@@ -0,0 +1,41 @@
+# This is a template `tox.ini` file for snaps
+# This file is managed by bootstack-charms-spec and should not be modified
+# within individual snap repos. https://launchpad.net/bootstack-charms-spec
+
+[tox]
+skipsdist=True
+envlist = lint, unit, func
+skip_missing_interpreters = True
+
+[testenv]
+basepython = python3
+setenv = PYTHONPATH={toxinidir}
+
+[testenv:lint]
+commands =
+    pflake8
+    pylint --recursive=y .
+    black --check --diff --color .
+    isort --check --diff --color .
+deps =
+    black
+    flake8
+    pyproject-flake8
+    flake8-docstrings
+    pep8-naming
+    flake8-colors
+    colorama
+    isort
+    pylint
+    {[testenv:func]deps}
+
+[testenv:reformat]
+envdir = {toxworkdir}/lint
+deps = {[testenv:lint]deps}
+commands =
+    black .
+    isort .
+
+[testenv:unit]
+
+[testenv:func]
\ No newline at end of file

From 89b10bd31d28e27f0eb1e259f718106812c0dd87 Mon Sep 17 00:00:00 2001
From: Deezzir <deezzir@gmail.com>
Date: Fri, 6 Sep 2024 11:29:45 -0400
Subject: [PATCH 02/36] Add install instructions to the README

---
 README.md | 12 +++++++++---
 1 file changed, 9 insertions(+), 3 deletions(-)

diff --git a/README.md b/README.md
index 08f5469..704667e 100644
--- a/README.md
+++ b/README.md
@@ -1,4 +1,4 @@
-# dcgm-snap
+# DCGM Snap
 
 This is a snap delivering NVIDIA dcgm components.
 The snap consists of [dcgm](https://developer.nvidia.com/dcgm) and [dcgm-exporter](https://github.com/NVIDIA/dcgm-exporter).
@@ -8,5 +8,11 @@ The snap consists of [dcgm](https://developer.nvidia.com/dcgm) and [dcgm-exporte
 You can build the snap locally by using the command:
 
 ```shell
-snapcraft --use-lxd
-```
\ No newline at end of file
+make build
+```
+
+## Install the snap
+
+```shell
+snap install --dangerous ./dcgm.snap
+```

From b531437b39d1407d9123116f958bdec382606b16 Mon Sep 17 00:00:00 2001
From: Deezzir <deezzir@gmail.com>
Date: Fri, 6 Sep 2024 11:30:00 -0400
Subject: [PATCH 03/36] Add test structure and snap setup

---
 tests/functional/test_init.py      | 13 +++++++++++++
 tests/functional/test_snap_dcgm.py |  5 +++++
 2 files changed, 18 insertions(+)
 create mode 100644 tests/functional/test_init.py
 create mode 100644 tests/functional/test_snap_dcgm.py

diff --git a/tests/functional/test_init.py b/tests/functional/test_init.py
new file mode 100644
index 0000000..8dce7e9
--- /dev/null
+++ b/tests/functional/test_init.py
@@ -0,0 +1,13 @@
+import os
+import subprocess
+
+import pytest
+
+
+@pytest.fixture(scope="session", autouse=True)
+def install_tempest_snap():
+    snap = os.environ["TEST_SNAP"]
+    process = subprocess.run(["sudo", "snap", "install", "--dangerous", snap])
+    assert 0 == process.returncode
+    yield
+    subprocess.run(["sudo", "snap", "remove", "--purge", "tempest"])
diff --git a/tests/functional/test_snap_dcgm.py b/tests/functional/test_snap_dcgm.py
new file mode 100644
index 0000000..c3215e7
--- /dev/null
+++ b/tests/functional/test_snap_dcgm.py
@@ -0,0 +1,5 @@
+import subprocess
+
+
+def test_network_interface():
+    pass

From 74f7c35e7c5ffa2e23de023e1904913ec039ea05 Mon Sep 17 00:00:00 2001
From: Deezzir <deezzir@gmail.com>
Date: Fri, 6 Sep 2024 15:15:17 -0400
Subject: [PATCH 04/36] Update .gitignore

---
 .gitignore | 20 +++++++++++++++++++-
 1 file changed, 19 insertions(+), 1 deletion(-)

diff --git a/.gitignore b/.gitignore
index f24fcd3..74a9ea4 100644
--- a/.gitignore
+++ b/.gitignore
@@ -1,8 +1,21 @@
 # This is a template `.gitignore` file for snaps
+# This file is managed by bootstack-charms-spec and should not be modified
+# within individual snap repos. https://launchpad.net/bootstack-charms-spec
+
+# Byte-compiled / optimized / DLL files
+__pycache__/
+*.py[cod]
+*$py.class
 
 # Tests files and dir
+.pytest_cache/
+.coverage
+.tox
+.venv
 reports/
 **/report/
+htmlcov/
+.mypy_cache
 
 # Log files
 *.log
@@ -17,9 +30,14 @@ reports/
 # version data
 repo-info
 
+# Python builds
+deb_dist/
+dist/
+*.egg-info/
+
 # Snaps
 *.snap
 
 # Builds
 .build/
-build/
+build/
\ No newline at end of file

From ab9f50283ec041c6488f008b86b38ee833b0a3a2 Mon Sep 17 00:00:00 2001
From: Deezzir <deezzir@gmail.com>
Date: Fri, 6 Sep 2024 15:32:27 -0400
Subject: [PATCH 05/36] Make dcgm-exporter into a daemon

---
 snap/snapcraft.yaml | 6 ++++++
 1 file changed, 6 insertions(+)

diff --git a/snap/snapcraft.yaml b/snap/snapcraft.yaml
index 973814a..e3dab38 100644
--- a/snap/snapcraft.yaml
+++ b/snap/snapcraft.yaml
@@ -29,6 +29,12 @@ apps:
     plugs:
       - network-bind
       - opengl
+    daemon: simple
+    install-mode: disable
+    restart-condition: on-failure
+    restart-delay: 2s
+    environment:
+      DCGM_HOME_DIR: "${SNAP_COMMON}"
   dcgmi:
     command: usr/bin/dcgmi
     plugs:

From 931c25dd4b7a98e6fa6382cc73fdfb22df619344 Mon Sep 17 00:00:00 2001
From: Deezzir <deezzir@gmail.com>
Date: Fri, 6 Sep 2024 15:32:58 -0400
Subject: [PATCH 06/36] Add simple smoke test for dcgm-exporter endpoint

---
 tests/functional/conftest.py       | 27 +++++++++++++++++++++++++++
 tests/functional/test_init.py      | 13 -------------
 tests/functional/test_snap_dcgm.py | 16 ++++++++++++++--
 tox.ini                            |  8 +++++++-
 4 files changed, 48 insertions(+), 16 deletions(-)
 create mode 100644 tests/functional/conftest.py
 delete mode 100644 tests/functional/test_init.py

diff --git a/tests/functional/conftest.py b/tests/functional/conftest.py
new file mode 100644
index 0000000..00bd5ea
--- /dev/null
+++ b/tests/functional/conftest.py
@@ -0,0 +1,27 @@
+import os
+import subprocess
+from time import sleep
+
+import pytest
+
+
+@pytest.fixture(scope="session", autouse=True)
+def install_dcgm_snap():
+    """Install the snap and enable dcgm-exporter service for testing."""
+    snap = os.environ["TEST_SNAP"]
+    dcgm_exporter_service = "snap.dcgm.dcgm-exporter.service"
+    assert 0 == subprocess.run(["sudo", "snap", "install", "--dangerous", snap]).returncode
+
+    subprocess.run(["sudo", "systemctl", "enable", "--now", dcgm_exporter_service])
+    sleep(5)  # Give some time for the service to start
+
+    assert (
+        0
+        == subprocess.run(
+            ["sudo", "systemctl", "is-active", "--quiet", dcgm_exporter_service]
+        ).returncode
+    )
+
+    yield
+
+    subprocess.run(["sudo", "snap", "remove", "--purge", "dcgm"])
diff --git a/tests/functional/test_init.py b/tests/functional/test_init.py
deleted file mode 100644
index 8dce7e9..0000000
--- a/tests/functional/test_init.py
+++ /dev/null
@@ -1,13 +0,0 @@
-import os
-import subprocess
-
-import pytest
-
-
-@pytest.fixture(scope="session", autouse=True)
-def install_tempest_snap():
-    snap = os.environ["TEST_SNAP"]
-    process = subprocess.run(["sudo", "snap", "install", "--dangerous", snap])
-    assert 0 == process.returncode
-    yield
-    subprocess.run(["sudo", "snap", "remove", "--purge", "tempest"])
diff --git a/tests/functional/test_snap_dcgm.py b/tests/functional/test_snap_dcgm.py
index c3215e7..3bd9e5b 100644
--- a/tests/functional/test_snap_dcgm.py
+++ b/tests/functional/test_snap_dcgm.py
@@ -1,5 +1,17 @@
 import subprocess
+from time import sleep
 
 
-def test_network_interface():
-    pass
+def test_dcgm_exporter_endpoint():
+    """Smoke test of the dcgm-exporter service and its endpoint."""
+    dcgm_exporter_endpoint = "localhost:9400/metrics"
+    subprocess.run(["sudo", "snap", "disconnect", "dcgm:network-bind"])
+
+    # The Endpoint should be unavailable with the networking plug
+    assert 0 != subprocess.run(["curl", dcgm_exporter_endpoint]).returncode
+
+    subprocess.run(["sudo", "snap", "connect", "dcgm:network-bind"])
+    print("reconnect")
+    sleep(5) # should be sufficient for the service to restart
+    
+    assert 0 == subprocess.run(["curl", dcgm_exporter_endpoint]).returncode
diff --git a/tox.ini b/tox.ini
index 6827ecb..72eca81 100644
--- a/tox.ini
+++ b/tox.ini
@@ -38,4 +38,10 @@ commands =
 
 [testenv:unit]
 
-[testenv:func]
\ No newline at end of file
+[testenv:func]
+deps =
+    pytest
+passenv =
+    TEST_*
+commands =
+    pytest {toxinidir}/tests/functional {posargs:-v}
\ No newline at end of file

From b4f58774dca78449e1e0b11714baa71cba5e56d9 Mon Sep 17 00:00:00 2001
From: Deezzir <deezzir@gmail.com>
Date: Fri, 6 Sep 2024 15:33:08 -0400
Subject: [PATCH 07/36] Update README.md

---
 README.md | 1 +
 1 file changed, 1 insertion(+)

diff --git a/README.md b/README.md
index 704667e..15a1f12 100644
--- a/README.md
+++ b/README.md
@@ -15,4 +15,5 @@ make build
 
 ```shell
 snap install --dangerous ./dcgm.snap
+sudo systemctl enable snap.dcgm.dcgm-exporter.service --now # enable and start dcgm-exporter service
 ```

From 36fd7579d178627b3b0d731ae7b9754a23c717a9 Mon Sep 17 00:00:00 2001
From: Deezzir <deezzir@gmail.com>
Date: Fri, 6 Sep 2024 15:34:06 -0400
Subject: [PATCH 08/36] Fix format

---
 tests/functional/test_snap_dcgm.py | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/tests/functional/test_snap_dcgm.py b/tests/functional/test_snap_dcgm.py
index 3bd9e5b..1750cec 100644
--- a/tests/functional/test_snap_dcgm.py
+++ b/tests/functional/test_snap_dcgm.py
@@ -12,6 +12,6 @@ def test_dcgm_exporter_endpoint():
 
     subprocess.run(["sudo", "snap", "connect", "dcgm:network-bind"])
     print("reconnect")
-    sleep(5) # should be sufficient for the service to restart
-    
+    sleep(5)  # should be sufficient for the service to restart
+
     assert 0 == subprocess.run(["curl", dcgm_exporter_endpoint]).returncode

From e257ad711162faeeb78666bde52fe81d5836bd7b Mon Sep 17 00:00:00 2001
From: Deezzir <deezzir@gmail.com>
Date: Fri, 6 Sep 2024 16:19:22 -0400
Subject: [PATCH 09/36] Remove unneccesary envar for dcgm-exporter daemon

---
 snap/snapcraft.yaml | 2 --
 1 file changed, 2 deletions(-)

diff --git a/snap/snapcraft.yaml b/snap/snapcraft.yaml
index e3dab38..da5936e 100644
--- a/snap/snapcraft.yaml
+++ b/snap/snapcraft.yaml
@@ -33,8 +33,6 @@ apps:
     install-mode: disable
     restart-condition: on-failure
     restart-delay: 2s
-    environment:
-      DCGM_HOME_DIR: "${SNAP_COMMON}"
   dcgmi:
     command: usr/bin/dcgmi
     plugs:

From 5d5047bf8b87ed26a4f52f8e7906937e8ed578f4 Mon Sep 17 00:00:00 2001
From: Deezzir <deezzir@gmail.com>
Date: Wed, 11 Sep 2024 10:11:21 -0400
Subject: [PATCH 10/36] Un-omit the test directory for the additional coverage

---
 pyproject.toml | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/pyproject.toml b/pyproject.toml
index 8af02ae..049c82c 100644
--- a/pyproject.toml
+++ b/pyproject.toml
@@ -53,7 +53,7 @@ module = ["setuptools"]
 [tool.coverage.run]
 relative_files = true
 source = ["."]
-omit = ["tests/**", "docs/**", "lib/**", "snap/**", "build/**", "setup.py"]
+omit = ["docs/**", "lib/**", "snap/**", "build/**", "setup.py"]
 
 [tool.coverage.report]
 fail_under = 100
@@ -63,4 +63,4 @@ show_missing = true
 directory = "tests/unit/report/html"
 
 [tool.coverage.xml]
-output = "tests/unit/report/coverage.xml"
\ No newline at end of file
+output = "tests/unit/report/coverage.xml"

From 5be41618f6bb74ec4db50b1bb33f9829e08c0446 Mon Sep 17 00:00:00 2001
From: Deezzir <deezzir@gmail.com>
Date: Wed, 11 Sep 2024 10:22:41 -0400
Subject: [PATCH 11/36] Remove redundant unit test commands

---
 Makefile | 7 +------
 tox.ini  | 4 +---
 2 files changed, 2 insertions(+), 9 deletions(-)

diff --git a/Makefile b/Makefile
index 3394bab..f5f7cff 100644
--- a/Makefile
+++ b/Makefile
@@ -16,7 +16,6 @@ help:
 	@echo " make build - build the snap"
 	@echo " make lint - run lint checkers"
 	@echo " make reformat - run lint tools to auto format code"
-	@echo " make unittests - run the tests defined in the unittest subdirectory"
 	@echo " make functional - run the tests defined in the functional subdirectory"
 	@echo " make test - run lint, proof, unittests and functional targets"
 	@echo ""
@@ -25,10 +24,6 @@ lint:
 	@echo "Running lint checks"
 	@tox -e lint
 
-unittests:
-	@echo "Running unit tests"
-	@tox -e unit -- ${UNIT_ARGS}
-
 test: lint unittests functional
 	@echo "Tests completed for the snap."
 
@@ -52,4 +47,4 @@ functional: build
 	@TEST_SNAP=${SNAP_FILE} tox -e func -- ${FUNC_ARGS}
 
 # The targets below don't depend on a file
-.PHONY: help clean build lint reformat unittests functional test
\ No newline at end of file
+.PHONY: help clean build lint reformat unittests functional test
diff --git a/tox.ini b/tox.ini
index 72eca81..a0fd1a7 100644
--- a/tox.ini
+++ b/tox.ini
@@ -36,12 +36,10 @@ commands =
     black .
     isort .
 
-[testenv:unit]
-
 [testenv:func]
 deps =
     pytest
 passenv =
     TEST_*
 commands =
-    pytest {toxinidir}/tests/functional {posargs:-v}
\ No newline at end of file
+    pytest {toxinidir}/tests/functional {posargs:-v}

From 349ee7a7c5df640431be848a7d9d48390c7b1239 Mon Sep 17 00:00:00 2001
From: Deezzir <deezzir@gmail.com>
Date: Wed, 11 Sep 2024 11:50:52 -0400
Subject: [PATCH 12/36] Make dcgm-exporter listen address configurable

---
 snap/hooks/configure            |  8 +++++++-
 snap/local/run_dcgm_exporter.sh | 14 ++++++++++++++
 snap/snapcraft.yaml             |  4 ++--
 3 files changed, 23 insertions(+), 3 deletions(-)
 create mode 100644 snap/local/run_dcgm_exporter.sh

diff --git a/snap/hooks/configure b/snap/hooks/configure
index 5d9de78..6364787 100644
--- a/snap/hooks/configure
+++ b/snap/hooks/configure
@@ -3,4 +3,10 @@
 # Register config options if unset,
 # so users can see available options by running
 # `sudo snap get dcgm`.
-[ -z "$(snapctl get nv-hostengine-port)" ] && snapctl set nv-hostengine-port=
+if [ -z "$(snapctl get nv-hostengine-port)" ]; then
+    snapctl set nv-hostengine-port=""
+fi
+
+if [ -z "$(snapctl get dcgm_exporter_listen)" ]; then
+    snapctl set dcgm_exporter_listen=""
+fi
diff --git a/snap/local/run_dcgm_exporter.sh b/snap/local/run_dcgm_exporter.sh
new file mode 100644
index 0000000..ad14d56
--- /dev/null
+++ b/snap/local/run_dcgm_exporter.sh
@@ -0,0 +1,14 @@
+#!/bin/bash
+set -euo pipefail
+
+# Build the argument list for the dcgm-exporter command
+args=()
+
+# Add the dcgm-exporter-port option if it is set. Default: “:9400”
+dcgm_exporter_listen="$(snapctl get dcgm_exporter_listen)"
+
+if [ -n "$dcgm_exporter_listen" ]; then
+    args+=("-a" "$dcgm_exporter_listen")
+fi
+
+exec "$SNAP/bin/dcgm-exporter" "${args[@]}"
diff --git a/snap/snapcraft.yaml b/snap/snapcraft.yaml
index da5936e..e5f356e 100644
--- a/snap/snapcraft.yaml
+++ b/snap/snapcraft.yaml
@@ -25,14 +25,13 @@ package-repositories:
 
 apps:
   dcgm-exporter:
-    command: bin/dcgm-exporter
+    command: run_dcgm_exporter.sh
     plugs:
       - network-bind
       - opengl
     daemon: simple
     install-mode: disable
     restart-condition: on-failure
-    restart-delay: 2s
   dcgmi:
     command: usr/bin/dcgmi
     plugs:
@@ -61,6 +60,7 @@ parts:
     override-build: |
       craftctl default
       chmod +x run_nv_hostengine.sh
+      chmod +x run_dcgm_exporter.sh
   dcgm-exporter:
     plugin: go
     stage-packages: [datacenter-gpu-manager]

From 48a50591d9113f18338c7218d5707402eb3f2cf3 Mon Sep 17 00:00:00 2001
From: Deezzir <deezzir@gmail.com>
Date: Wed, 11 Sep 2024 12:00:54 -0400
Subject: [PATCH 13/36] Remove unittest from Makefile

---
 Makefile | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/Makefile b/Makefile
index f5f7cff..6c285f2 100644
--- a/Makefile
+++ b/Makefile
@@ -17,14 +17,14 @@ help:
 	@echo " make lint - run lint checkers"
 	@echo " make reformat - run lint tools to auto format code"
 	@echo " make functional - run the tests defined in the functional subdirectory"
-	@echo " make test - run lint, proof, unittests and functional targets"
+	@echo " make test - run lint, proof, and functional targets"
 	@echo ""
 
 lint:
 	@echo "Running lint checks"
 	@tox -e lint
 
-test: lint unittests functional
+test: lint functional
 	@echo "Tests completed for the snap."
 
 reformat:

From 818bd7af3ca886edcb7236abf0e1adb9d5e02857 Mon Sep 17 00:00:00 2001
From: Deezzir <deezzir@gmail.com>
Date: Wed, 11 Sep 2024 16:10:44 -0400
Subject: [PATCH 14/36] Improve dcgm-exporter endpoint test

---
 snap/hooks/configure               |  4 +--
 snap/local/run_dcgm_exporter.sh    |  2 +-
 tests/functional/conftest.py       | 24 ++++++++++++-----
 tests/functional/test_snap_dcgm.py | 42 +++++++++++++++++++++++-------
 4 files changed, 52 insertions(+), 20 deletions(-)

diff --git a/snap/hooks/configure b/snap/hooks/configure
index 6364787..5a0d5dd 100644
--- a/snap/hooks/configure
+++ b/snap/hooks/configure
@@ -7,6 +7,6 @@ if [ -z "$(snapctl get nv-hostengine-port)" ]; then
     snapctl set nv-hostengine-port=""
 fi
 
-if [ -z "$(snapctl get dcgm_exporter_listen)" ]; then
-    snapctl set dcgm_exporter_listen=""
+if [ -z "$(snapctl get dcgm-exporter-listen)" ]; then
+    snapctl set dcgm-exporter-listen=""
 fi
diff --git a/snap/local/run_dcgm_exporter.sh b/snap/local/run_dcgm_exporter.sh
index ad14d56..10ec838 100644
--- a/snap/local/run_dcgm_exporter.sh
+++ b/snap/local/run_dcgm_exporter.sh
@@ -5,7 +5,7 @@ set -euo pipefail
 args=()
 
 # Add the dcgm-exporter-port option if it is set. Default: “:9400”
-dcgm_exporter_listen="$(snapctl get dcgm_exporter_listen)"
+dcgm_exporter_listen="$(snapctl get dcgm-exporter-listen)"
 
 if [ -n "$dcgm_exporter_listen" ]; then
     args+=("-a" "$dcgm_exporter_listen")
diff --git a/tests/functional/conftest.py b/tests/functional/conftest.py
index 00bd5ea..24e21c0 100644
--- a/tests/functional/conftest.py
+++ b/tests/functional/conftest.py
@@ -1,6 +1,6 @@
 import os
 import subprocess
-from time import sleep
+import time
 
 import pytest
 
@@ -10,18 +10,28 @@ def install_dcgm_snap():
     """Install the snap and enable dcgm-exporter service for testing."""
     snap = os.environ["TEST_SNAP"]
     dcgm_exporter_service = "snap.dcgm.dcgm-exporter.service"
-    assert 0 == subprocess.run(["sudo", "snap", "install", "--dangerous", snap]).returncode
+
+    assert (
+        0 == subprocess.run(["sudo", "snap", "install", "--dangerous", snap]).returncode
+    ), f"Failed to install {snap}"
 
     subprocess.run(["sudo", "systemctl", "enable", "--now", dcgm_exporter_service])
-    sleep(5)  # Give some time for the service to start
 
-    assert (
-        0
-        == subprocess.run(
+    dcgm_exporter_is_active = (
+        lambda: subprocess.call(
             ["sudo", "systemctl", "is-active", "--quiet", dcgm_exporter_service]
-        ).returncode
+        )
+        == 0
     )
 
+    timeout = 30  # seconds
+    start_time = time.time()
+
+    while not dcgm_exporter_is_active():
+        if time.time() - start_time > timeout:
+            assert False, f"Failed to start {dcgm_exporter_service} service"
+        time.sleep(5)
+
     yield
 
     subprocess.run(["sudo", "snap", "remove", "--purge", "dcgm"])
diff --git a/tests/functional/test_snap_dcgm.py b/tests/functional/test_snap_dcgm.py
index 1750cec..d02d4a8 100644
--- a/tests/functional/test_snap_dcgm.py
+++ b/tests/functional/test_snap_dcgm.py
@@ -1,17 +1,39 @@
 import subprocess
-from time import sleep
+import time
 
 
 def test_dcgm_exporter_endpoint():
-    """Smoke test of the dcgm-exporter service and its endpoint."""
-    dcgm_exporter_endpoint = "localhost:9400/metrics"
-    subprocess.run(["sudo", "snap", "disconnect", "dcgm:network-bind"])
+    """Test of the dcgm-exporter service and its endpoint."""
+    endpoint = "localhost:9400/metrics"
+    timeout = 60  # seconds
+    start_time = time.time()
 
-    # The Endpoint should be unavailable with the networking plug
-    assert 0 != subprocess.run(["curl", dcgm_exporter_endpoint]).returncode
+    def query_endpoint():
+        return subprocess.run(["curl", endpoint], text=True, capture_output=True)
 
-    subprocess.run(["sudo", "snap", "connect", "dcgm:network-bind"])
-    print("reconnect")
-    sleep(5)  # should be sufficient for the service to restart
+    while (result := query_endpoint()).returncode != 0 or not result.stdout.strip():
+        if time.time() - start_time > timeout:
+            assert False, f"Failed to reach '{endpoint}' of the dcgm-exporter service"
+        time.sleep(5)
 
-    assert 0 == subprocess.run(["curl", dcgm_exporter_endpoint]).returncode
+    assert "DCGM_FI_DRIVER_VERSION" in result.stdout, "No dcgm exported metrics found"
+
+
+def test_dcgm_configs():
+    """Test snap configuratin."""
+    pass
+
+
+def test_dcgm_nv_hostengine():
+    """Test of the dcgm-nv-hostengine service and its endpoint."""
+    pass
+
+
+def test_dcgmi():
+    """Test of the dcgmi command."""
+    pass
+
+
+def test_dcgmproftesters():
+    """Test of the dcgmproftesters."""
+    pass

From c201332045e92a079ab14b4473ddf7ae5fdef8ce Mon Sep 17 00:00:00 2001
From: Deezzir <deezzir@gmail.com>
Date: Wed, 11 Sep 2024 16:12:53 -0400
Subject: [PATCH 15/36] Add missing trailing whitespace

---
 .gitignore | 2 +-
 rename.sh  | 2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)

diff --git a/.gitignore b/.gitignore
index 74a9ea4..b6b2573 100644
--- a/.gitignore
+++ b/.gitignore
@@ -40,4 +40,4 @@ dist/
 
 # Builds
 .build/
-build/
\ No newline at end of file
+build/
diff --git a/rename.sh b/rename.sh
index 3b14d01..6d645e1 100755
--- a/rename.sh
+++ b/rename.sh
@@ -14,4 +14,4 @@ then
     rm "${snap}.snap"
 fi
 echo "Renaming snap here."
-mv ${snap}_*.snap ${snap}.snap
\ No newline at end of file
+mv ${snap}_*.snap ${snap}.snap

From 36a913f5893d00c5743989014f9f6c1330d2c6e8 Mon Sep 17 00:00:00 2001
From: Deezzir <deezzir@gmail.com>
Date: Wed, 11 Sep 2024 19:52:27 -0400
Subject: [PATCH 16/36] Add simple tests for other components

---
 tests/functional/test_snap_dcgm.py | 43 ++++++++++++++++++++++++------
 1 file changed, 35 insertions(+), 8 deletions(-)

diff --git a/tests/functional/test_snap_dcgm.py b/tests/functional/test_snap_dcgm.py
index d02d4a8..eaf1f57 100644
--- a/tests/functional/test_snap_dcgm.py
+++ b/tests/functional/test_snap_dcgm.py
@@ -1,3 +1,4 @@
+import json
 import subprocess
 import time
 
@@ -19,21 +20,47 @@ def query_endpoint():
     assert "DCGM_FI_DRIVER_VERSION" in result.stdout, "No dcgm exported metrics found"
 
 
-def test_dcgm_configs():
-    """Test snap configuratin."""
-    pass
-
-
 def test_dcgm_nv_hostengine():
-    """Test of the dcgm-nv-hostengine service and its endpoint."""
-    pass
+    """Check the dcgm-nv-hostengine service."""
+    assert 0 == subprocess.call(
+        ["sudo", "systemctl", "is-active", "--quiet", "snap.dcgm.nv-hostengine.service"]
+    ), "DCGM NV Hostengine service is not running"
 
 
 def test_dcgmi():
     """Test of the dcgmi command."""
-    pass
+    result = subprocess.run(["dcgm.dcgmi", "discovery", "-l"], capture_output=True, text=True)
+    print(result)
+    assert "GPU ID" in result.stdout.strip(), "DCGMI is not working"
 
 
 def test_dcgmproftesters():
     """Test of the dcgmproftesters."""
     pass
+
+
+def test_dcgm_port_configs():
+    """Test snap port configuratin."""
+    services = ["snap.dcgm.dcgm-exporter.service", "snap.dcgm.nv-hostengine.service"]
+    configs = ["dcgm-exporter-listen", "nv-hostengine-port"]
+    new_values = [":9466", "5666"]
+
+    result = subprocess.run(
+        ["sudo", "snap", "get", "dcgm", "-d"], check=True, capture_output=True, text=True
+    )
+    pairs = json.loads(result.stdout.strip())
+    assert all(config in pairs for config in configs), "Missing snap configuration keys"
+
+    for config, new_value in zip(configs, new_values):
+        assert 0 == subprocess.call(
+            ["sudo", "snap", "set", "dcgm", f"{config}={new_value}"]
+        ), f"Failed to set snap configuration key '{config}'"
+
+    # restart the service to apply the new configuration
+    for service in services:
+        subprocess.run(["sudo", "systemctl", "restart", service])
+
+    for service, port in zip(services, new_values):
+        assert 0 == subprocess.call(
+            ["sudo", "lsof", "-i", f":{port.lstrip(':')}"]
+        ), f"{service} port is not listening"

From 558007e1725b51e819eec121c068c9764cfab3b9 Mon Sep 17 00:00:00 2001
From: Deezzir <deezzir@gmail.com>
Date: Wed, 11 Sep 2024 20:17:18 -0400
Subject: [PATCH 17/36] Improve dcgmi test

---
 Makefile                           | 2 +-
 tests/functional/test_snap_dcgm.py | 3 +--
 2 files changed, 2 insertions(+), 3 deletions(-)

diff --git a/Makefile b/Makefile
index 6c285f2..2b77033 100644
--- a/Makefile
+++ b/Makefile
@@ -47,4 +47,4 @@ functional: build
 	@TEST_SNAP=${SNAP_FILE} tox -e func -- ${FUNC_ARGS}
 
 # The targets below don't depend on a file
-.PHONY: help clean build lint reformat unittests functional test
+.PHONY: help clean build lint reformat functional test
diff --git a/tests/functional/test_snap_dcgm.py b/tests/functional/test_snap_dcgm.py
index eaf1f57..9f599c0 100644
--- a/tests/functional/test_snap_dcgm.py
+++ b/tests/functional/test_snap_dcgm.py
@@ -29,8 +29,7 @@ def test_dcgm_nv_hostengine():
 
 def test_dcgmi():
     """Test of the dcgmi command."""
-    result = subprocess.run(["dcgm.dcgmi", "discovery", "-l"], capture_output=True, text=True)
-    print(result)
+    result = subprocess.run(["dcgm.dcgmi", "discovery", "-l"], check=True, capture_output=True, text=True)
     assert "GPU ID" in result.stdout.strip(), "DCGMI is not working"
 
 

From 9e2494409aa329226d15675602fb231a1608b0e4 Mon Sep 17 00:00:00 2001
From: Deezzir <deezzir@gmail.com>
Date: Thu, 12 Sep 2024 11:10:12 -0400
Subject: [PATCH 18/36] Revert README.md

---
 README.md | 11 ++---------
 1 file changed, 2 insertions(+), 9 deletions(-)

diff --git a/README.md b/README.md
index 15a1f12..1e8a975 100644
--- a/README.md
+++ b/README.md
@@ -1,4 +1,4 @@
-# DCGM Snap
+# dcgm-snap
 
 This is a snap delivering NVIDIA dcgm components.
 The snap consists of [dcgm](https://developer.nvidia.com/dcgm) and [dcgm-exporter](https://github.com/NVIDIA/dcgm-exporter).
@@ -8,12 +8,5 @@ The snap consists of [dcgm](https://developer.nvidia.com/dcgm) and [dcgm-exporte
 You can build the snap locally by using the command:
 
 ```shell
-make build
-```
-
-## Install the snap
-
-```shell
-snap install --dangerous ./dcgm.snap
-sudo systemctl enable snap.dcgm.dcgm-exporter.service --now # enable and start dcgm-exporter service
+snapcraft --use-lxd
 ```

From 2577dfb5fe093c934d183527c1134737ec18bf75 Mon Sep 17 00:00:00 2001
From: Deezzir <deezzir@gmail.com>
Date: Thu, 12 Sep 2024 11:20:18 -0400
Subject: [PATCH 19/36] Switch to snap services subcommands

---
 tests/functional/conftest.py       | 2 +-
 tests/functional/test_snap_dcgm.py | 8 ++++----
 2 files changed, 5 insertions(+), 5 deletions(-)

diff --git a/tests/functional/conftest.py b/tests/functional/conftest.py
index 24e21c0..9ec8027 100644
--- a/tests/functional/conftest.py
+++ b/tests/functional/conftest.py
@@ -15,7 +15,7 @@ def install_dcgm_snap():
         0 == subprocess.run(["sudo", "snap", "install", "--dangerous", snap]).returncode
     ), f"Failed to install {snap}"
 
-    subprocess.run(["sudo", "systemctl", "enable", "--now", dcgm_exporter_service])
+    subprocess.run(["sudo", "snap", "start", "dcgm.dcgm-exporter"])
 
     dcgm_exporter_is_active = (
         lambda: subprocess.call(
diff --git a/tests/functional/test_snap_dcgm.py b/tests/functional/test_snap_dcgm.py
index 9f599c0..a8a3d43 100644
--- a/tests/functional/test_snap_dcgm.py
+++ b/tests/functional/test_snap_dcgm.py
@@ -40,15 +40,15 @@ def test_dcgmproftesters():
 
 def test_dcgm_port_configs():
     """Test snap port configuratin."""
-    services = ["snap.dcgm.dcgm-exporter.service", "snap.dcgm.nv-hostengine.service"]
+    services = ["dcgm.dcgm-exporter", "dcgm.nv-hostengine"]
     configs = ["dcgm-exporter-listen", "nv-hostengine-port"]
     new_values = [":9466", "5666"]
 
     result = subprocess.run(
         ["sudo", "snap", "get", "dcgm", "-d"], check=True, capture_output=True, text=True
     )
-    pairs = json.loads(result.stdout.strip())
-    assert all(config in pairs for config in configs), "Missing snap configuration keys"
+    dcgm_snap_config = json.loads(result.stdout.strip())
+    assert all(config in dcgm_snap_config for config in configs), "Missing snap configuration keys"
 
     for config, new_value in zip(configs, new_values):
         assert 0 == subprocess.call(
@@ -57,7 +57,7 @@ def test_dcgm_port_configs():
 
     # restart the service to apply the new configuration
     for service in services:
-        subprocess.run(["sudo", "systemctl", "restart", service])
+        subprocess.run(["sudo", "snap", "restart", service])
 
     for service, port in zip(services, new_values):
         assert 0 == subprocess.call(

From 4dfd25092752413a2143e07fc108f50e70f1aa32 Mon Sep 17 00:00:00 2001
From: Deezzir <deezzir@gmail.com>
Date: Thu, 12 Sep 2024 11:24:39 -0400
Subject: [PATCH 20/36] Remane dcgm-exporter snap config

---
 snap/hooks/configure               | 4 ++--
 snap/local/run_dcgm_exporter.sh    | 6 +++---
 tests/functional/test_snap_dcgm.py | 6 ++++--
 3 files changed, 9 insertions(+), 7 deletions(-)

diff --git a/snap/hooks/configure b/snap/hooks/configure
index 5a0d5dd..f05078a 100644
--- a/snap/hooks/configure
+++ b/snap/hooks/configure
@@ -7,6 +7,6 @@ if [ -z "$(snapctl get nv-hostengine-port)" ]; then
     snapctl set nv-hostengine-port=""
 fi
 
-if [ -z "$(snapctl get dcgm-exporter-listen)" ]; then
-    snapctl set dcgm-exporter-listen=""
+if [ -z "$(snapctl get dcgm-exporter-address)" ]; then
+    snapctl set dcgm-exporter-address=""
 fi
diff --git a/snap/local/run_dcgm_exporter.sh b/snap/local/run_dcgm_exporter.sh
index 10ec838..e852dd9 100644
--- a/snap/local/run_dcgm_exporter.sh
+++ b/snap/local/run_dcgm_exporter.sh
@@ -5,10 +5,10 @@ set -euo pipefail
 args=()
 
 # Add the dcgm-exporter-port option if it is set. Default: “:9400”
-dcgm_exporter_listen="$(snapctl get dcgm-exporter-listen)"
+dcgm_exporter_address="$(snapctl get dcgm-exporter-address)"
 
-if [ -n "$dcgm_exporter_listen" ]; then
-    args+=("-a" "$dcgm_exporter_listen")
+if [ -n "$dcgm_exporter_address" ]; then
+    args+=("-a" "$dcgm_exporter_address")
 fi
 
 exec "$SNAP/bin/dcgm-exporter" "${args[@]}"
diff --git a/tests/functional/test_snap_dcgm.py b/tests/functional/test_snap_dcgm.py
index a8a3d43..38ee4e9 100644
--- a/tests/functional/test_snap_dcgm.py
+++ b/tests/functional/test_snap_dcgm.py
@@ -29,7 +29,9 @@ def test_dcgm_nv_hostengine():
 
 def test_dcgmi():
     """Test of the dcgmi command."""
-    result = subprocess.run(["dcgm.dcgmi", "discovery", "-l"], check=True, capture_output=True, text=True)
+    result = subprocess.run(
+        ["dcgm.dcgmi", "discovery", "-l"], check=True, capture_output=True, text=True
+    )
     assert "GPU ID" in result.stdout.strip(), "DCGMI is not working"
 
 
@@ -41,7 +43,7 @@ def test_dcgmproftesters():
 def test_dcgm_port_configs():
     """Test snap port configuratin."""
     services = ["dcgm.dcgm-exporter", "dcgm.nv-hostengine"]
-    configs = ["dcgm-exporter-listen", "nv-hostengine-port"]
+    configs = ["dcgm-exporter-address", "nv-hostengine-port"]
     new_values = [":9466", "5666"]
 
     result = subprocess.run(

From bfb452af81a938ca17691f47f35704c4e233dac9 Mon Sep 17 00:00:00 2001
From: Deezzir <deezzir@gmail.com>
Date: Thu, 12 Sep 2024 12:50:50 -0400
Subject: [PATCH 21/36] Simplify dcgm-exporter test

---
 tests/functional/requirments.txt   |  1 +
 tests/functional/test_snap_dcgm.py | 25 ++++++-------------------
 tox.ini                            |  1 +
 3 files changed, 8 insertions(+), 19 deletions(-)
 create mode 100644 tests/functional/requirments.txt

diff --git a/tests/functional/requirments.txt b/tests/functional/requirments.txt
new file mode 100644
index 0000000..067f0df
--- /dev/null
+++ b/tests/functional/requirments.txt
@@ -0,0 +1 @@
+tenacity==9.0.0
diff --git a/tests/functional/test_snap_dcgm.py b/tests/functional/test_snap_dcgm.py
index 38ee4e9..ba5e38d 100644
--- a/tests/functional/test_snap_dcgm.py
+++ b/tests/functional/test_snap_dcgm.py
@@ -1,23 +1,15 @@
 import json
 import subprocess
-import time
+import urllib.request
 
+from tenacity import retry, stop_after_delay, wait_fixed
 
+
+@retry(wait=wait_fixed(5), stop=stop_after_delay(60))
 def test_dcgm_exporter_endpoint():
     """Test of the dcgm-exporter service and its endpoint."""
-    endpoint = "localhost:9400/metrics"
-    timeout = 60  # seconds
-    start_time = time.time()
-
-    def query_endpoint():
-        return subprocess.run(["curl", endpoint], text=True, capture_output=True)
-
-    while (result := query_endpoint()).returncode != 0 or not result.stdout.strip():
-        if time.time() - start_time > timeout:
-            assert False, f"Failed to reach '{endpoint}' of the dcgm-exporter service"
-        time.sleep(5)
-
-    assert "DCGM_FI_DRIVER_VERSION" in result.stdout, "No dcgm exported metrics found"
+    endpoint = "http://localhost:9400/metrics"
+    urllib.request.urlopen(endpoint)
 
 
 def test_dcgm_nv_hostengine():
@@ -35,11 +27,6 @@ def test_dcgmi():
     assert "GPU ID" in result.stdout.strip(), "DCGMI is not working"
 
 
-def test_dcgmproftesters():
-    """Test of the dcgmproftesters."""
-    pass
-
-
 def test_dcgm_port_configs():
     """Test snap port configuratin."""
     services = ["dcgm.dcgm-exporter", "dcgm.nv-hostengine"]
diff --git a/tox.ini b/tox.ini
index a0fd1a7..455178f 100644
--- a/tox.ini
+++ b/tox.ini
@@ -39,6 +39,7 @@ commands =
 [testenv:func]
 deps =
     pytest
+    -r {toxinidir}/tests/functional/requirments.txt
 passenv =
     TEST_*
 commands =

From 6c32bfc6d6945b2c375b67a8babdbcdb9cb9c3ee Mon Sep 17 00:00:00 2001
From: Deezzir <deezzir@gmail.com>
Date: Thu, 12 Sep 2024 12:51:23 -0400
Subject: [PATCH 22/36] Remove makefile and rename.sh

---
 Makefile  | 50 --------------------------------------------------
 rename.sh | 17 -----------------
 2 files changed, 67 deletions(-)
 delete mode 100644 Makefile
 delete mode 100755 rename.sh

diff --git a/Makefile b/Makefile
deleted file mode 100644
index 2b77033..0000000
--- a/Makefile
+++ /dev/null
@@ -1,50 +0,0 @@
-# This is a template `Makefile` file for snaps
-# This file is managed by bootstack-charms-spec and should not be modified
-# within individual snap repos. https://launchpad.net/bootstack-charms-spec
-
-PYTHON := /usr/bin/python3
-
-PROJECTPATH=$(dir $(realpath ${MAKEFILE_LIST}))
-SNAP_NAME=$(shell cat ${PROJECTPATH}/snap/snapcraft.yaml | grep -E '^name:' | awk '{print $$2}')
-SNAP_FILE=${PROJECTPATH}/${SNAP_NAME}.snap
-
-help:
-	@echo "This project supports the following targets"
-	@echo ""
-	@echo " make help - show this text"
-	@echo " make clean - remove unneeded files"
-	@echo " make build - build the snap"
-	@echo " make lint - run lint checkers"
-	@echo " make reformat - run lint tools to auto format code"
-	@echo " make functional - run the tests defined in the functional subdirectory"
-	@echo " make test - run lint, proof, and functional targets"
-	@echo ""
-
-lint:
-	@echo "Running lint checks"
-	@tox -e lint
-
-test: lint functional
-	@echo "Tests completed for the snap."
-
-reformat:
-	@echo "Reformat files with black and isort"
-	@tox -e reformat
-
-build:
-	@echo "Building the snap"
-	@snapcraft --use-lxd
-	@bash -c ./rename.sh
-
-clean:
-	@echo "Cleaning snap"
-	@snapcraft clean --use-lxd
-	@echo "Cleaning existing snap builds"
-	@rm -rf ${SNAP_FILE}
-
-functional: build
-	@echo "Executing functional tests using built snap"
-	@TEST_SNAP=${SNAP_FILE} tox -e func -- ${FUNC_ARGS}
-
-# The targets below don't depend on a file
-.PHONY: help clean build lint reformat functional test
diff --git a/rename.sh b/rename.sh
deleted file mode 100755
index 6d645e1..0000000
--- a/rename.sh
+++ /dev/null
@@ -1,17 +0,0 @@
-#!/bin/bash
-# This is a template `rename.sh` file for snaps
-# This file is managed by bootstack-charms-spec and should not be modified
-# within individual snap repos. https://launchpad.net/bootstack-charms-spec
-
-snap=$(grep -E "^name:" snap/snapcraft.yaml | awk '{print $2}')
-echo "renaming ${snap}_*.snap to ${snap}.snap"
-echo -n "pwd: "
-pwd
-ls -al
-echo "Removing previous snap if it exists"
-if [[ -e "${snap}.snap" ]];
-then
-    rm "${snap}.snap"
-fi
-echo "Renaming snap here."
-mv ${snap}_*.snap ${snap}.snap

From 19faa1119418d2d83221295d2538714842f7ee7d Mon Sep 17 00:00:00 2001
From: Deezzir <deezzir@gmail.com>
Date: Thu, 12 Sep 2024 15:08:36 -0400
Subject: [PATCH 23/36] Imrpove & simplify tests

---
 tests/functional/conftest.py       | 38 ++++++++---------------
 tests/functional/requirments.txt   |  1 +
 tests/functional/test_snap_dcgm.py | 48 ++++++++++++++++++++++--------
 3 files changed, 48 insertions(+), 39 deletions(-)

diff --git a/tests/functional/conftest.py b/tests/functional/conftest.py
index 9ec8027..e6b0c7a 100644
--- a/tests/functional/conftest.py
+++ b/tests/functional/conftest.py
@@ -1,37 +1,23 @@
-import os
 import subprocess
-import time
 
 import pytest
+import yaml
 
 
 @pytest.fixture(scope="session", autouse=True)
 def install_dcgm_snap():
     """Install the snap and enable dcgm-exporter service for testing."""
-    snap = os.environ["TEST_SNAP"]
-    dcgm_exporter_service = "snap.dcgm.dcgm-exporter.service"
-
-    assert (
-        0 == subprocess.run(["sudo", "snap", "install", "--dangerous", snap]).returncode
-    ), f"Failed to install {snap}"
-
-    subprocess.run(["sudo", "snap", "start", "dcgm.dcgm-exporter"])
-
-    dcgm_exporter_is_active = (
-        lambda: subprocess.call(
-            ["sudo", "systemctl", "is-active", "--quiet", dcgm_exporter_service]
+    with open("snap/snapcraft.yaml") as f:
+        snapcraft = yaml.safe_load(f)
+        snap_build_name = f"{snapcraft['name']}_*_amd64.snap"
+
+        subprocess.run(
+            f"sudo snap install --devmode {snap_build_name}",
+            check=True,
+            capture_output=True,
+            shell=True,
         )
-        == 0
-    )
-
-    timeout = 30  # seconds
-    start_time = time.time()
-
-    while not dcgm_exporter_is_active():
-        if time.time() - start_time > timeout:
-            assert False, f"Failed to start {dcgm_exporter_service} service"
-        time.sleep(5)
 
-    yield
+        yield
 
-    subprocess.run(["sudo", "snap", "remove", "--purge", "dcgm"])
+        subprocess.run("sudo snap remove --purge dcgm".split(), check=True)
diff --git a/tests/functional/requirments.txt b/tests/functional/requirments.txt
index 067f0df..771e68b 100644
--- a/tests/functional/requirments.txt
+++ b/tests/functional/requirments.txt
@@ -1 +1,2 @@
 tenacity==9.0.0
+pyyaml==6.0.2
diff --git a/tests/functional/test_snap_dcgm.py b/tests/functional/test_snap_dcgm.py
index ba5e38d..2d30f66 100644
--- a/tests/functional/test_snap_dcgm.py
+++ b/tests/functional/test_snap_dcgm.py
@@ -1,28 +1,48 @@
 import json
 import subprocess
 import urllib.request
+from time import sleep
 
 from tenacity import retry, stop_after_delay, wait_fixed
 
 
-@retry(wait=wait_fixed(5), stop=stop_after_delay(60))
-def test_dcgm_exporter_endpoint():
+@retry(wait=wait_fixed(5), stop=stop_after_delay(15))
+def test_dcgm_exporter():
     """Test of the dcgm-exporter service and its endpoint."""
     endpoint = "http://localhost:9400/metrics"
+    dcgm_exporter_service = "dcgm.dcgm-exporter"
+
+    subprocess.run(f"sudo snap start {dcgm_exporter_service}".split(), check=True)
+
+    result = subprocess.run(
+        f"sudo snap services {dcgm_exporter_service}".split(),
+        check=True,
+        capture_output=True,
+        text=True,
+    )
+    assert " active" in result.stdout.strip(), "dcgm-exporter service is not active"
+
     urllib.request.urlopen(endpoint)
 
 
 def test_dcgm_nv_hostengine():
     """Check the dcgm-nv-hostengine service."""
-    assert 0 == subprocess.call(
-        ["sudo", "systemctl", "is-active", "--quiet", "snap.dcgm.nv-hostengine.service"]
-    ), "DCGM NV Hostengine service is not running"
+    nv_hostengine_service = "dcgm.nv-hostengine"
+
+    service = subprocess.run(
+        f"snap services {nv_hostengine_service}".split(),
+        check=True,
+        capture_output=True,
+        text=True,
+    )
+
+    assert " active" in service.stdout.strip(), "nv-hostengine service is not active"
 
 
 def test_dcgmi():
     """Test of the dcgmi command."""
     result = subprocess.run(
-        ["dcgm.dcgmi", "discovery", "-l"], check=True, capture_output=True, text=True
+        "dcgm.dcgmi discovery -l".split(), check=True, capture_output=True, text=True
     )
     assert "GPU ID" in result.stdout.strip(), "DCGMI is not working"
 
@@ -34,21 +54,23 @@ def test_dcgm_port_configs():
     new_values = [":9466", "5666"]
 
     result = subprocess.run(
-        ["sudo", "snap", "get", "dcgm", "-d"], check=True, capture_output=True, text=True
+        "sudo snap get dcgm -d".split(), check=True, capture_output=True, text=True
     )
     dcgm_snap_config = json.loads(result.stdout.strip())
     assert all(config in dcgm_snap_config for config in configs), "Missing snap configuration keys"
 
     for config, new_value in zip(configs, new_values):
-        assert 0 == subprocess.call(
-            ["sudo", "snap", "set", "dcgm", f"{config}={new_value}"]
-        ), f"Failed to set snap configuration key '{config}'"
+        subprocess.run(
+            f"sudo snap set dcgm {config}={new_value}".split(), check=True
+        ), f"Failed to set {config}"
 
     # restart the service to apply the new configuration
     for service in services:
-        subprocess.run(["sudo", "snap", "restart", service])
+        subprocess.run(f"sudo snap restart {service}".split(), check=True)
+
+    sleep(5)
 
     for service, port in zip(services, new_values):
-        assert 0 == subprocess.call(
-            ["sudo", "lsof", "-i", f":{port.lstrip(':')}"]
+        subprocess.run(
+            f"sudo lsof -i :{port.lstrip(':')}".split(), check=True
         ), f"{service} port is not listening"

From 9774fdeb9d5652d590613c88d45f2deb153ab2b0 Mon Sep 17 00:00:00 2001
From: Deezzir <deezzir@gmail.com>
Date: Thu, 12 Sep 2024 15:08:45 -0400
Subject: [PATCH 24/36] Add func test to the CI

---
 .github/workflows/check.yaml | 19 +++++++++++++++++--
 1 file changed, 17 insertions(+), 2 deletions(-)

diff --git a/.github/workflows/check.yaml b/.github/workflows/check.yaml
index 5bc07fd..729c1d4 100644
--- a/.github/workflows/check.yaml
+++ b/.github/workflows/check.yaml
@@ -31,10 +31,13 @@ jobs:
         run: |
           sudo apt update
           sudo apt install -y yamllint
+          pipx install tox
 
       - name: Lint yaml files
-        run: |
-          yamllint .yamllint snap/snapcraft.yaml
+        run: yamllint .yamllint snap/snapcraft.yaml
+
+      - name: Lint tests
+        run: tox -e lint
 
   build:
     runs-on: ubuntu-22.04
@@ -45,3 +48,15 @@ jobs:
 
       - name: Verify snap builds successfully
         uses: snapcore/action-build@v1
+
+  func:
+    needs:
+      - lint
+      - build
+    runs-on: ubuntu-22.04
+    steps:
+      - uses: actions/checkout@v4
+      - name: Install dependencies
+
+      - name: Run tests
+        run: tox -e func

From 8738bc777505fdb1a9497f19bcc52021993235bd Mon Sep 17 00:00:00 2001
From: Deezzir <deezzir@gmail.com>
Date: Thu, 12 Sep 2024 15:14:45 -0400
Subject: [PATCH 25/36] Fix check.yaml

---
 .github/workflows/check.yaml | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/.github/workflows/check.yaml b/.github/workflows/check.yaml
index 729c1d4..ec3c6cb 100644
--- a/.github/workflows/check.yaml
+++ b/.github/workflows/check.yaml
@@ -57,6 +57,7 @@ jobs:
     steps:
       - uses: actions/checkout@v4
       - name: Install dependencies
-
+        run: |
+          pipx install tox
       - name: Run tests
         run: tox -e func

From bea3f7113cdd7f0e9d773209403742cafe9a7181 Mon Sep 17 00:00:00 2001
From: Deezzir <deezzir@gmail.com>
Date: Thu, 12 Sep 2024 15:26:11 -0400
Subject: [PATCH 26/36] Fix check.yaml

---
 .github/workflows/check.yaml | 14 +++++++++++---
 1 file changed, 11 insertions(+), 3 deletions(-)

diff --git a/.github/workflows/check.yaml b/.github/workflows/check.yaml
index ec3c6cb..f4d2144 100644
--- a/.github/workflows/check.yaml
+++ b/.github/workflows/check.yaml
@@ -55,9 +55,17 @@ jobs:
       - build
     runs-on: ubuntu-22.04
     steps:
-      - uses: actions/checkout@v4
+      - name: Checkout
+        uses: actions/checkout@v4
+
       - name: Install dependencies
-        run: |
-          pipx install tox
+        run: pipx install tox
+
+      - name: Download snap package(s)
+        uses: actions/download-artifact@v4
+        with:
+          pattern: ${{ needs.build.outputs.artifact-prefix }}-*
+          merge-multiple: true
+
       - name: Run tests
         run: tox -e func

From 55874a9766a9823de8c9e6c2ca7dd5de24f5d229 Mon Sep 17 00:00:00 2001
From: Deezzir <deezzir@gmail.com>
Date: Thu, 12 Sep 2024 15:45:09 -0400
Subject: [PATCH 27/36] Merge build and func CI jobs

---
 .github/workflows/check.yaml | 13 +++----------
 1 file changed, 3 insertions(+), 10 deletions(-)

diff --git a/.github/workflows/check.yaml b/.github/workflows/check.yaml
index f4d2144..b1b5d4f 100644
--- a/.github/workflows/check.yaml
+++ b/.github/workflows/check.yaml
@@ -39,7 +39,9 @@ jobs:
       - name: Lint tests
         run: tox -e lint
 
-  build:
+  func:
+    needs:
+      - lint
     runs-on: ubuntu-22.04
     steps:
       - uses: actions/checkout@v4
@@ -49,15 +51,6 @@ jobs:
       - name: Verify snap builds successfully
         uses: snapcore/action-build@v1
 
-  func:
-    needs:
-      - lint
-      - build
-    runs-on: ubuntu-22.04
-    steps:
-      - name: Checkout
-        uses: actions/checkout@v4
-
       - name: Install dependencies
         run: pipx install tox
 

From b740fa01f8b80d666bd21d65e8ed6b24d469cdeb Mon Sep 17 00:00:00 2001
From: Deezzir <deezzir@gmail.com>
Date: Thu, 12 Sep 2024 15:53:06 -0400
Subject: [PATCH 28/36] Remove redundant step for test job

---
 .github/workflows/check.yaml | 6 ------
 1 file changed, 6 deletions(-)

diff --git a/.github/workflows/check.yaml b/.github/workflows/check.yaml
index b1b5d4f..67f2cb8 100644
--- a/.github/workflows/check.yaml
+++ b/.github/workflows/check.yaml
@@ -54,11 +54,5 @@ jobs:
       - name: Install dependencies
         run: pipx install tox
 
-      - name: Download snap package(s)
-        uses: actions/download-artifact@v4
-        with:
-          pattern: ${{ needs.build.outputs.artifact-prefix }}-*
-          merge-multiple: true
-
       - name: Run tests
         run: tox -e func

From 3138f0a8cf3439a30ee2e506b1da7aa761d0f570 Mon Sep 17 00:00:00 2001
From: Deezzir <deezzir@gmail.com>
Date: Thu, 12 Sep 2024 18:38:33 -0400
Subject: [PATCH 29/36] Revert CI to separate jobs for build and func

---
 .github/workflows/check.yaml | 32 ++++++++++++++++++++++++++++++--
 tests/functional/conftest.py |  2 +-
 2 files changed, 31 insertions(+), 3 deletions(-)

diff --git a/.github/workflows/check.yaml b/.github/workflows/check.yaml
index 67f2cb8..0a6c2bd 100644
--- a/.github/workflows/check.yaml
+++ b/.github/workflows/check.yaml
@@ -27,6 +27,11 @@ jobs:
         with:
           fetch-depth: 0  # Complete git history is required to generate the version from git tags.
 
+      - name: Setup Python 3.10
+        uses: actions/setup-python@v5
+        with:
+          python-version: '3.10'
+
       - name: Install dependencies
         run: |
           sudo apt update
@@ -39,7 +44,7 @@ jobs:
       - name: Lint tests
         run: tox -e lint
 
-  func:
+  build:
     needs:
       - lint
     runs-on: ubuntu-22.04
@@ -51,8 +56,31 @@ jobs:
       - name: Verify snap builds successfully
         uses: snapcore/action-build@v1
 
+      - name: Upload the built snap
+        uses: actions/upload-artifact@v4
+        with:
+          name: SNAP_FILE
+          path: dcgm.snap
+
+  func:
+    needs:
+      - build
+    runs-on: ubuntu-22.04
+    steps:
+      - uses: actions/checkout@v4
+
+      - name: Download the built snap
+        uses: actions/download-artifact@v4
+        with:
+          name: SNAP_FILE
+
+      - name: Setup Python 3.10
+        uses: actions/setup-python@v5
+        with:
+          python-version: '3.10'
+
       - name: Install dependencies
         run: pipx install tox
 
       - name: Run tests
-        run: tox -e func
+        run: TEST_SNAP=$GITHUB_WORKSPACE/dcgm.snap tox -e func
diff --git a/tests/functional/conftest.py b/tests/functional/conftest.py
index e6b0c7a..cac57c3 100644
--- a/tests/functional/conftest.py
+++ b/tests/functional/conftest.py
@@ -9,7 +9,7 @@ def install_dcgm_snap():
     """Install the snap and enable dcgm-exporter service for testing."""
     with open("snap/snapcraft.yaml") as f:
         snapcraft = yaml.safe_load(f)
-        snap_build_name = f"{snapcraft['name']}_*_amd64.snap"
+        snap_build_name = f"{snapcraft['name']}_*.snap"
 
         subprocess.run(
             f"sudo snap install --devmode {snap_build_name}",

From c2ea073c004002a8fe459f7515b56b119adaef01 Mon Sep 17 00:00:00 2001
From: Deezzir <deezzir@gmail.com>
Date: Thu, 12 Sep 2024 18:49:09 -0400
Subject: [PATCH 30/36] Fix artifact path

---
 .github/workflows/check.yaml | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/.github/workflows/check.yaml b/.github/workflows/check.yaml
index 0a6c2bd..3091d11 100644
--- a/.github/workflows/check.yaml
+++ b/.github/workflows/check.yaml
@@ -60,7 +60,7 @@ jobs:
         uses: actions/upload-artifact@v4
         with:
           name: SNAP_FILE
-          path: dcgm.snap
+          path: dcgm*.snap
 
   func:
     needs:
@@ -83,4 +83,4 @@ jobs:
         run: pipx install tox
 
       - name: Run tests
-        run: TEST_SNAP=$GITHUB_WORKSPACE/dcgm.snap tox -e func
+        run: tox -e func

From 439376edb56460713c3480d5cdc3b498090c5c30 Mon Sep 17 00:00:00 2001
From: Deezzir <deezzir@gmail.com>
Date: Thu, 12 Sep 2024 20:42:23 -0400
Subject: [PATCH 31/36] Refine

---
 tests/functional/conftest.py       | 21 +++++++++------------
 tests/functional/test_snap_dcgm.py |  1 +
 2 files changed, 10 insertions(+), 12 deletions(-)

diff --git a/tests/functional/conftest.py b/tests/functional/conftest.py
index cac57c3..56b80a0 100644
--- a/tests/functional/conftest.py
+++ b/tests/functional/conftest.py
@@ -1,23 +1,20 @@
 import subprocess
 
 import pytest
-import yaml
 
 
 @pytest.fixture(scope="session", autouse=True)
 def install_dcgm_snap():
     """Install the snap and enable dcgm-exporter service for testing."""
-    with open("snap/snapcraft.yaml") as f:
-        snapcraft = yaml.safe_load(f)
-        snap_build_name = f"{snapcraft['name']}_*.snap"
+    snap_build_name = "dcgm_*.snap"
 
-        subprocess.run(
-            f"sudo snap install --devmode {snap_build_name}",
-            check=True,
-            capture_output=True,
-            shell=True,
-        )
+    subprocess.run(
+        f"sudo snap install --dangerous {snap_build_name}",
+        check=True,
+        capture_output=True,
+        shell=True,
+    )
 
-        yield
+    yield
 
-        subprocess.run("sudo snap remove --purge dcgm".split(), check=True)
+    subprocess.run("sudo snap remove --purge dcgm".split(), check=True)
diff --git a/tests/functional/test_snap_dcgm.py b/tests/functional/test_snap_dcgm.py
index 2d30f66..842d6df 100644
--- a/tests/functional/test_snap_dcgm.py
+++ b/tests/functional/test_snap_dcgm.py
@@ -22,6 +22,7 @@ def test_dcgm_exporter():
     )
     assert " active" in result.stdout.strip(), "dcgm-exporter service is not active"
 
+    # Check the exporter endpoint, will raise an exception if the endpoint is not reachable
     urllib.request.urlopen(endpoint)
 
 

From 60bdf8514db11166821e2a1b5f97306fa3427871 Mon Sep 17 00:00:00 2001
From: Deezzir <deezzir@gmail.com>
Date: Fri, 13 Sep 2024 13:46:52 -0400
Subject: [PATCH 32/36] Refinements

---
 snap/hooks/configure               |  2 ++
 snap/local/run_dcgm_exporter.sh    |  2 +-
 tests/functional/conftest.py       | 17 +++++++++++++++++
 tests/functional/requirments.txt   |  4 ++--
 tests/functional/test_snap_dcgm.py | 17 +++--------------
 5 files changed, 25 insertions(+), 17 deletions(-)

diff --git a/snap/hooks/configure b/snap/hooks/configure
index f05078a..a55fc8a 100644
--- a/snap/hooks/configure
+++ b/snap/hooks/configure
@@ -4,9 +4,11 @@
 # so users can see available options by running
 # `sudo snap get dcgm`.
 if [ -z "$(snapctl get nv-hostengine-port)" ]; then
+    # Setting to empty string for the nv-hostengine binary to use the default port. (5555)
     snapctl set nv-hostengine-port=""
 fi
 
 if [ -z "$(snapctl get dcgm-exporter-address)" ]; then
+    # Setting to empty string for the dcgm-exporter binary to use the default address. (:9400)
     snapctl set dcgm-exporter-address=""
 fi
diff --git a/snap/local/run_dcgm_exporter.sh b/snap/local/run_dcgm_exporter.sh
index e852dd9..a1dd641 100644
--- a/snap/local/run_dcgm_exporter.sh
+++ b/snap/local/run_dcgm_exporter.sh
@@ -4,7 +4,7 @@ set -euo pipefail
 # Build the argument list for the dcgm-exporter command
 args=()
 
-# Add the dcgm-exporter-port option if it is set. Default: “:9400”
+# Add the dcgm-exporter-address option if it is set. Default: “:9400”
 dcgm_exporter_address="$(snapctl get dcgm-exporter-address)"
 
 if [ -n "$dcgm_exporter_address" ]; then
diff --git a/tests/functional/conftest.py b/tests/functional/conftest.py
index 56b80a0..5e3e58a 100644
--- a/tests/functional/conftest.py
+++ b/tests/functional/conftest.py
@@ -1,6 +1,20 @@
 import subprocess
 
 import pytest
+from tenacity import retry, stop_after_delay, wait_fixed
+
+
+@retry(wait=wait_fixed(5), stop=stop_after_delay(20))
+def check_dcgm_exporter_service():
+    dcgm_exporter_service = "dcgm.dcgm-exporter"
+
+    result = subprocess.run(
+        f"sudo snap services {dcgm_exporter_service}".split(),
+        check=True,
+        capture_output=True,
+        text=True,
+    )
+    assert " active" in result.stdout.strip(), f"{dcgm_exporter_service} service is not active"
 
 
 @pytest.fixture(scope="session", autouse=True)
@@ -15,6 +29,9 @@ def install_dcgm_snap():
         shell=True,
     )
 
+    subprocess.run("sudo snap start dcgm.dcgm-exporter".split(), check=True)
+    check_dcgm_exporter_service()
+
     yield
 
     subprocess.run("sudo snap remove --purge dcgm".split(), check=True)
diff --git a/tests/functional/requirments.txt b/tests/functional/requirments.txt
index 771e68b..bd2d4ef 100644
--- a/tests/functional/requirments.txt
+++ b/tests/functional/requirments.txt
@@ -1,2 +1,2 @@
-tenacity==9.0.0
-pyyaml==6.0.2
+tenacity
+pyyaml
diff --git a/tests/functional/test_snap_dcgm.py b/tests/functional/test_snap_dcgm.py
index 842d6df..7d38f56 100644
--- a/tests/functional/test_snap_dcgm.py
+++ b/tests/functional/test_snap_dcgm.py
@@ -6,21 +6,10 @@
 from tenacity import retry, stop_after_delay, wait_fixed
 
 
-@retry(wait=wait_fixed(5), stop=stop_after_delay(15))
+@retry(wait=wait_fixed(5), stop=stop_after_delay(30))
 def test_dcgm_exporter():
     """Test of the dcgm-exporter service and its endpoint."""
     endpoint = "http://localhost:9400/metrics"
-    dcgm_exporter_service = "dcgm.dcgm-exporter"
-
-    subprocess.run(f"sudo snap start {dcgm_exporter_service}".split(), check=True)
-
-    result = subprocess.run(
-        f"sudo snap services {dcgm_exporter_service}".split(),
-        check=True,
-        capture_output=True,
-        text=True,
-    )
-    assert " active" in result.stdout.strip(), "dcgm-exporter service is not active"
 
     # Check the exporter endpoint, will raise an exception if the endpoint is not reachable
     urllib.request.urlopen(endpoint)
@@ -37,7 +26,7 @@ def test_dcgm_nv_hostengine():
         text=True,
     )
 
-    assert " active" in service.stdout.strip(), "nv-hostengine service is not active"
+    assert " active" in service.stdout.strip(), f"{nv_hostengine_service} service is not active"
 
 
 def test_dcgmi():
@@ -48,7 +37,7 @@ def test_dcgmi():
     assert "GPU ID" in result.stdout.strip(), "DCGMI is not working"
 
 
-def test_dcgm_port_configs():
+def test_dcgm_bind_configs():
     """Test snap port configuratin."""
     services = ["dcgm.dcgm-exporter", "dcgm.nv-hostengine"]
     configs = ["dcgm-exporter-address", "nv-hostengine-port"]

From 6987442e8a39ab461a0b7714a50bde47ea4d1904 Mon Sep 17 00:00:00 2001
From: Deezzir <deezzir@gmail.com>
Date: Fri, 13 Sep 2024 18:52:05 -0400
Subject: [PATCH 33/36] Refine tests and comments

---
 .gitignore                                    |  2 -
 pyproject.toml                                |  4 --
 snap/hooks/configure                          |  4 +-
 snap/snapcraft.yaml                           |  2 +
 tests/functional/conftest.py                  | 15 ----
 .../{requirments.txt => requirements.txt}     |  1 -
 tests/functional/test_snap_dcgm.py            | 72 +++++++++++--------
 tox.ini                                       |  6 +-
 8 files changed, 47 insertions(+), 59 deletions(-)
 rename tests/functional/{requirments.txt => requirements.txt} (56%)

diff --git a/.gitignore b/.gitignore
index b6b2573..f43a27a 100644
--- a/.gitignore
+++ b/.gitignore
@@ -1,6 +1,4 @@
 # This is a template `.gitignore` file for snaps
-# This file is managed by bootstack-charms-spec and should not be modified
-# within individual snap repos. https://launchpad.net/bootstack-charms-spec
 
 # Byte-compiled / optimized / DLL files
 __pycache__/
diff --git a/pyproject.toml b/pyproject.toml
index 049c82c..ec1e0ae 100644
--- a/pyproject.toml
+++ b/pyproject.toml
@@ -1,7 +1,3 @@
-# This is a template `pyproject.toml` file for snaps
-# This file is managed by bootstack-charms-spec and should not be modified
-# within individual snap repos. https://launchpad.net/bootstack-charms-spec
-
 [tool.flake8]
 ignore = ["C901", "D100", "D101", "D102", "D103", "W503", "W504"]
 exclude = ['.eggs', '.git', '.tox', '.venv', '.build', 'build', 'report']
diff --git a/snap/hooks/configure b/snap/hooks/configure
index a55fc8a..4478c03 100644
--- a/snap/hooks/configure
+++ b/snap/hooks/configure
@@ -5,10 +5,10 @@
 # `sudo snap get dcgm`.
 if [ -z "$(snapctl get nv-hostengine-port)" ]; then
     # Setting to empty string for the nv-hostengine binary to use the default port. (5555)
-    snapctl set nv-hostengine-port=""
+    snapctl set nv-hostengine-port="5555"
 fi
 
 if [ -z "$(snapctl get dcgm-exporter-address)" ]; then
     # Setting to empty string for the dcgm-exporter binary to use the default address. (:9400)
-    snapctl set dcgm-exporter-address=""
+    snapctl set dcgm-exporter-address=":9400"
 fi
diff --git a/snap/snapcraft.yaml b/snap/snapcraft.yaml
index 401f7f0..31ad3fc 100644
--- a/snap/snapcraft.yaml
+++ b/snap/snapcraft.yaml
@@ -26,6 +26,8 @@ apps:
       - network-bind
       - opengl
     daemon: simple
+    # As this is a dcgm snap, not the dcgm-exporter snap,
+    # user might not be interested in running dcgm-exporter, so disable it by default
     install-mode: disable
     restart-condition: on-failure
   dcgmi:
diff --git a/tests/functional/conftest.py b/tests/functional/conftest.py
index 5e3e58a..5c19053 100644
--- a/tests/functional/conftest.py
+++ b/tests/functional/conftest.py
@@ -1,20 +1,6 @@
 import subprocess
 
 import pytest
-from tenacity import retry, stop_after_delay, wait_fixed
-
-
-@retry(wait=wait_fixed(5), stop=stop_after_delay(20))
-def check_dcgm_exporter_service():
-    dcgm_exporter_service = "dcgm.dcgm-exporter"
-
-    result = subprocess.run(
-        f"sudo snap services {dcgm_exporter_service}".split(),
-        check=True,
-        capture_output=True,
-        text=True,
-    )
-    assert " active" in result.stdout.strip(), f"{dcgm_exporter_service} service is not active"
 
 
 @pytest.fixture(scope="session", autouse=True)
@@ -30,7 +16,6 @@ def install_dcgm_snap():
     )
 
     subprocess.run("sudo snap start dcgm.dcgm-exporter".split(), check=True)
-    check_dcgm_exporter_service()
 
     yield
 
diff --git a/tests/functional/requirments.txt b/tests/functional/requirements.txt
similarity index 56%
rename from tests/functional/requirments.txt
rename to tests/functional/requirements.txt
index bd2d4ef..48f8e12 100644
--- a/tests/functional/requirments.txt
+++ b/tests/functional/requirements.txt
@@ -1,2 +1 @@
 tenacity
-pyyaml
diff --git a/tests/functional/test_snap_dcgm.py b/tests/functional/test_snap_dcgm.py
index 7d38f56..ee87c51 100644
--- a/tests/functional/test_snap_dcgm.py
+++ b/tests/functional/test_snap_dcgm.py
@@ -1,32 +1,41 @@
 import json
 import subprocess
 import urllib.request
-from time import sleep
 
-from tenacity import retry, stop_after_delay, wait_fixed
+import pytest
+from tenacity import Retrying, retry, stop_after_delay, wait_fixed
 
 
 @retry(wait=wait_fixed(5), stop=stop_after_delay(30))
 def test_dcgm_exporter():
     """Test of the dcgm-exporter service and its endpoint."""
+    dcgm_exporter_service = "snap.dcgm.dcgm-exporter"
     endpoint = "http://localhost:9400/metrics"
 
+    assert 0 == subprocess.call(
+        f"sudo systemctl is-active --quiet {dcgm_exporter_service}".split()
+    ), f"{dcgm_exporter_service} is not running"
+
     # Check the exporter endpoint, will raise an exception if the endpoint is not reachable
-    urllib.request.urlopen(endpoint)
+    response = urllib.request.urlopen(endpoint)
+
+    # The output of the exporter endpoint is not tested
+    # as in a virtual environment it will not have any GPU metrics
+    assert 200 == response.getcode(), "DCGM exporter endpoint returned an error"
 
 
 def test_dcgm_nv_hostengine():
     """Check the dcgm-nv-hostengine service."""
-    nv_hostengine_service = "dcgm.nv-hostengine"
+    nv_hostengine_service = "snap.dcgm.nv-hostengine"
+    nv_hostengine_port = 5555
 
-    service = subprocess.run(
-        f"snap services {nv_hostengine_service}".split(),
-        check=True,
-        capture_output=True,
-        text=True,
-    )
+    assert 0 == subprocess.call(
+        f"sudo systemctl is-active --quiet {nv_hostengine_service}".split()
+    ), f"{nv_hostengine_service} is not running"
 
-    assert " active" in service.stdout.strip(), f"{nv_hostengine_service} service is not active"
+    assert 0 == subprocess.call(
+        f"nc -z localhost {nv_hostengine_port}".split()
+    ), f"{nv_hostengine_service} is not listening on port {nv_hostengine_port}"
 
 
 def test_dcgmi():
@@ -34,33 +43,36 @@ def test_dcgmi():
     result = subprocess.run(
         "dcgm.dcgmi discovery -l".split(), check=True, capture_output=True, text=True
     )
-    assert "GPU ID" in result.stdout.strip(), "DCGMI is not working"
 
+    # Test if the command is working and outputs a table with the GPU ID
+    # The table will be empty in a virtual environment, but the command should still work
+    assert "GPU ID" in result.stdout.strip(), "DCGMI didn't produce the expected table"
 
-def test_dcgm_bind_configs():
-    """Test snap port configuratin."""
-    services = ["dcgm.dcgm-exporter", "dcgm.nv-hostengine"]
-    configs = ["dcgm-exporter-address", "nv-hostengine-port"]
-    new_values = [":9466", "5666"]
 
+bind_test_data = [
+    ("dcgm.dcgm-exporter", "dcgm-exporter-address", ":9466"),
+    ("dcgm.nv-hostengine", "nv-hostengine-port", "5566"),
+]
+
+
+@pytest.mark.parametrize("service, config, new_value", bind_test_data)
+def test_dcgm_bind_config(service: str, config: str, new_value: str):
+    """Test snap bind configuration."""
     result = subprocess.run(
         "sudo snap get dcgm -d".split(), check=True, capture_output=True, text=True
     )
     dcgm_snap_config = json.loads(result.stdout.strip())
-    assert all(config in dcgm_snap_config for config in configs), "Missing snap configuration keys"
+    assert config in dcgm_snap_config, f"{config} is not in the snap configuration"
 
-    for config, new_value in zip(configs, new_values):
-        subprocess.run(
-            f"sudo snap set dcgm {config}={new_value}".split(), check=True
-        ), f"Failed to set {config}"
+    assert 0 == subprocess.call(
+        f"sudo snap set dcgm {config}={new_value}".split()
+    ), f"Failed to set {config} to {new_value}"
 
     # restart the service to apply the new configuration
-    for service in services:
-        subprocess.run(f"sudo snap restart {service}".split(), check=True)
-
-    sleep(5)
+    subprocess.run(f"sudo snap restart {service}".split(), check=True)
 
-    for service, port in zip(services, new_values):
-        subprocess.run(
-            f"sudo lsof -i :{port.lstrip(':')}".split(), check=True
-        ), f"{service} port is not listening"
+    for attempt in Retrying(wait=wait_fixed(2), stop=stop_after_delay(10)):
+        with attempt:
+            assert 0 == subprocess.call(
+                f"nc -z localhost {new_value.lstrip(':')}".split()
+            ), f"{service} is not listening on {new_value}"
diff --git a/tox.ini b/tox.ini
index 455178f..fd01577 100644
--- a/tox.ini
+++ b/tox.ini
@@ -1,7 +1,3 @@
-# This is a template `tox.ini` file for snaps
-# This file is managed by bootstack-charms-spec and should not be modified
-# within individual snap repos. https://launchpad.net/bootstack-charms-spec
-
 [tox]
 skipsdist=True
 envlist = lint, unit, func
@@ -39,7 +35,7 @@ commands =
 [testenv:func]
 deps =
     pytest
-    -r {toxinidir}/tests/functional/requirments.txt
+    -r {toxinidir}/tests/functional/requirements.txt
 passenv =
     TEST_*
 commands =

From 2ea70fd936a53ab9ac7f9ada386f8912d48d8ba5 Mon Sep 17 00:00:00 2001
From: Deezzir <deezzir@gmail.com>
Date: Fri, 13 Sep 2024 19:03:34 -0400
Subject: [PATCH 34/36] Align check.yaml with check workflows

---
 .github/workflows/check.yaml | 21 ++++++++++++++-------
 1 file changed, 14 insertions(+), 7 deletions(-)

diff --git a/.github/workflows/check.yaml b/.github/workflows/check.yaml
index 3091d11..08d17a4 100644
--- a/.github/workflows/check.yaml
+++ b/.github/workflows/check.yaml
@@ -27,16 +27,19 @@ jobs:
         with:
           fetch-depth: 0  # Complete git history is required to generate the version from git tags.
 
-      - name: Setup Python 3.10
+      - name: Set up Python 3.10
         uses: actions/setup-python@v5
         with:
-          python-version: '3.10'
+          python-version: "3.10"
 
       - name: Install dependencies
         run: |
           sudo apt update
           sudo apt install -y yamllint
-          pipx install tox
+          python -m pip install --upgrade pip
+          # pin tox to the current major version to avoid
+          # workflows breaking all at once when a new major version is released.
+          python -m pip install 'tox<5'
 
       - name: Lint yaml files
         run: yamllint .yamllint snap/snapcraft.yaml
@@ -68,19 +71,23 @@ jobs:
     runs-on: ubuntu-22.04
     steps:
       - uses: actions/checkout@v4
+        with:
+          fetch-depth: 0  # Complete git history is required to generate the version from git tags.
 
       - name: Download the built snap
         uses: actions/download-artifact@v4
         with:
           name: SNAP_FILE
 
-      - name: Setup Python 3.10
+      - name: Set up Python 3.10
         uses: actions/setup-python@v5
         with:
-          python-version: '3.10'
+          python-version: "3.10"
 
       - name: Install dependencies
-        run: pipx install tox
+        run: |
+          python -m pip install --upgrade pip
+          python -m pip install 'tox<5'
 
-      - name: Run tests
+      - name: Run unit tests
         run: tox -e func

From 4059a6d37d8988cfc00bd4545c02926f663b2546 Mon Sep 17 00:00:00 2001
From: Deezzir <deezzir@gmail.com>
Date: Fri, 13 Sep 2024 19:52:04 -0400
Subject: [PATCH 35/36] Update comments for default configs

---
 snap/hooks/configure | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/snap/hooks/configure b/snap/hooks/configure
index 4478c03..9d5b294 100644
--- a/snap/hooks/configure
+++ b/snap/hooks/configure
@@ -4,11 +4,11 @@
 # so users can see available options by running
 # `sudo snap get dcgm`.
 if [ -z "$(snapctl get nv-hostengine-port)" ]; then
-    # Setting to empty string for the nv-hostengine binary to use the default port. (5555)
+    # Explictly use default bind port of nv-hostengine binary
     snapctl set nv-hostengine-port="5555"
 fi
 
 if [ -z "$(snapctl get dcgm-exporter-address)" ]; then
-    # Setting to empty string for the dcgm-exporter binary to use the default address. (:9400)
+    # Explictly use default bind address of dcgm-exporter binary
     snapctl set dcgm-exporter-address=":9400"
 fi

From bd6f22e2ceaf15765a8f6771c2dad108aaa97523 Mon Sep 17 00:00:00 2001
From: Deezzir <deezzir@gmail.com>
Date: Mon, 16 Sep 2024 12:33:07 -0400
Subject: [PATCH 36/36] Revert back the bind configs after the test

---
 tests/functional/test_snap_dcgm.py | 42 ++++++++++++++++++------------
 1 file changed, 25 insertions(+), 17 deletions(-)

diff --git a/tests/functional/test_snap_dcgm.py b/tests/functional/test_snap_dcgm.py
index ee87c51..e9ee0e4 100644
--- a/tests/functional/test_snap_dcgm.py
+++ b/tests/functional/test_snap_dcgm.py
@@ -49,13 +49,13 @@ def test_dcgmi():
     assert "GPU ID" in result.stdout.strip(), "DCGMI didn't produce the expected table"
 
 
-bind_test_data = [
-    ("dcgm.dcgm-exporter", "dcgm-exporter-address", ":9466"),
-    ("dcgm.nv-hostengine", "nv-hostengine-port", "5566"),
-]
-
-
-@pytest.mark.parametrize("service, config, new_value", bind_test_data)
+@pytest.mark.parametrize(
+    "service, config, new_value",
+    [
+        ("dcgm.dcgm-exporter", "dcgm-exporter-address", ":9466"),
+        ("dcgm.nv-hostengine", "nv-hostengine-port", "5566"),
+    ],
+)
 def test_dcgm_bind_config(service: str, config: str, new_value: str):
     """Test snap bind configuration."""
     result = subprocess.run(
@@ -63,16 +63,24 @@ def test_dcgm_bind_config(service: str, config: str, new_value: str):
     )
     dcgm_snap_config = json.loads(result.stdout.strip())
     assert config in dcgm_snap_config, f"{config} is not in the snap configuration"
+    old_value = dcgm_snap_config[config]
 
-    assert 0 == subprocess.call(
-        f"sudo snap set dcgm {config}={new_value}".split()
-    ), f"Failed to set {config} to {new_value}"
+    def set_config_and_check(value: str):
+        assert 0 == subprocess.call(
+            f"sudo snap set dcgm {config}={value}".split()
+        ), f"Failed to set {config} to {new_value}"
+
+        # restart the service to apply the configuration
+        subprocess.run(f"sudo snap restart {service}".split(), check=True)
+
+        for attempt in Retrying(wait=wait_fixed(2), stop=stop_after_delay(10)):
+            with attempt:
+                assert 0 == subprocess.call(
+                    f"nc -z localhost {value.lstrip(':')}".split()
+                ), f"{service} is not listening on {value}"
 
-    # restart the service to apply the new configuration
-    subprocess.run(f"sudo snap restart {service}".split(), check=True)
+    # Check new config
+    set_config_and_check(new_value)
 
-    for attempt in Retrying(wait=wait_fixed(2), stop=stop_after_delay(10)):
-        with attempt:
-            assert 0 == subprocess.call(
-                f"nc -z localhost {new_value.lstrip(':')}".split()
-            ), f"{service} is not listening on {new_value}"
+    # Revert back
+    set_config_and_check(str(old_value))