Bump actions/checkout from 3.5.3 to 3.6.0 (#4508)

Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>

.github/workflows/builder.yml

@@ -53,7 +53,7 @@ jobs:
       requirements: ${{ steps.requirements.outputs.changed }}
     steps:
       - name: Checkout the repository
-        uses: actions/checkout@v3.5.3
+        uses: actions/checkout@v3.6.0
         with:
           fetch-depth: 0
 
@@ -92,7 +92,7 @@ jobs:
         arch: ${{ fromJson(needs.init.outputs.architectures) }}
     steps:
       - name: Checkout the repository
-        uses: actions/checkout@v3.5.3
+        uses: actions/checkout@v3.6.0
         with:
           fetch-depth: 0
 
@@ -160,7 +160,7 @@ jobs:
         run: echo "BUILD_ARGS=--test" >> $GITHUB_ENV
 
       - name: Build supervisor
-        uses: home-assistant/builder@2023.06.1
+        uses: home-assistant/builder@2023.08.0
         with:
           args: |
             $BUILD_ARGS \
@@ -178,7 +178,7 @@ jobs:
     steps:
       - name: Checkout the repository
         if: needs.init.outputs.publish == 'true'
-        uses: actions/checkout@v3.5.3
+        uses: actions/checkout@v3.6.0
 
       - name: Initialize git
         if: needs.init.outputs.publish == 'true'
@@ -203,11 +203,11 @@ jobs:
     timeout-minutes: 60
     steps:
       - name: Checkout the repository
-        uses: actions/checkout@v3.5.3
+        uses: actions/checkout@v3.6.0
 
       - name: Build the Supervisor
         if: needs.init.outputs.publish != 'true'
-        uses: home-assistant/builder@2023.06.1
+        uses: home-assistant/builder@2023.08.0
         with:
           args: |
             --test \

.github/workflows/ci.yaml

@@ -25,7 +25,7 @@ jobs:
     name: Prepare Python dependencies
     steps:
       - name: Check out code from GitHub
-        uses: actions/checkout@v3.5.3
+        uses: actions/checkout@v3.6.0
       - name: Set up Python
         id: python
         uses: actions/setup-python@v4.7.0
@@ -66,7 +66,7 @@ jobs:
     needs: prepare
     steps:
       - name: Check out code from GitHub
-        uses: actions/checkout@v3.5.3
+        uses: actions/checkout@v3.6.0
       - name: Set up Python ${{ needs.prepare.outputs.python-version }}
         uses: actions/setup-python@v4.7.0
         id: python
@@ -95,7 +95,7 @@ jobs:
     needs: prepare
     steps:
       - name: Check out code from GitHub
-        uses: actions/checkout@v3.5.3
+        uses: actions/checkout@v3.6.0
       - name: Register hadolint problem matcher
         run: |
           echo "::add-matcher::.github/workflows/matchers/hadolint.json"
@@ -110,7 +110,7 @@ jobs:
     needs: prepare
     steps:
       - name: Check out code from GitHub
-        uses: actions/checkout@v3.5.3
+        uses: actions/checkout@v3.6.0
       - name: Set up Python ${{ needs.prepare.outputs.python-version }}
         uses: actions/setup-python@v4.7.0
         id: python
@@ -154,7 +154,7 @@ jobs:
     needs: prepare
     steps:
       - name: Check out code from GitHub
-        uses: actions/checkout@v3.5.3
+        uses: actions/checkout@v3.6.0
       - name: Set up Python ${{ needs.prepare.outputs.python-version }}
         uses: actions/setup-python@v4.7.0
         id: python
@@ -186,7 +186,7 @@ jobs:
     needs: prepare
     steps:
       - name: Check out code from GitHub
-        uses: actions/checkout@v3.5.3
+        uses: actions/checkout@v3.6.0
       - name: Set up Python ${{ needs.prepare.outputs.python-version }}
         uses: actions/setup-python@v4.7.0
         id: python
@@ -227,7 +227,7 @@ jobs:
     needs: prepare
     steps:
       - name: Check out code from GitHub
-        uses: actions/checkout@v3.5.3
+        uses: actions/checkout@v3.6.0
       - name: Set up Python ${{ needs.prepare.outputs.python-version }}
         uses: actions/setup-python@v4.7.0
         id: python
@@ -271,7 +271,7 @@ jobs:
     needs: prepare
     steps:
       - name: Check out code from GitHub
-        uses: actions/checkout@v3.5.3
+        uses: actions/checkout@v3.6.0
       - name: Set up Python ${{ needs.prepare.outputs.python-version }}
         uses: actions/setup-python@v4.7.0
         id: python
@@ -303,7 +303,7 @@ jobs:
     needs: prepare
     steps:
       - name: Check out code from GitHub
-        uses: actions/checkout@v3.5.3
+        uses: actions/checkout@v3.6.0
       - name: Set up Python ${{ needs.prepare.outputs.python-version }}
         uses: actions/setup-python@v4.7.0
         id: python
@@ -344,7 +344,7 @@ jobs:
     name: Run tests Python ${{ needs.prepare.outputs.python-version }}
     steps:
       - name: Check out code from GitHub
-        uses: actions/checkout@v3.5.3
+        uses: actions/checkout@v3.6.0
       - name: Set up Python ${{ needs.prepare.outputs.python-version }}
         uses: actions/setup-python@v4.7.0
         id: python
@@ -402,7 +402,7 @@ jobs:
     needs: ["pytest", "prepare"]
     steps:
       - name: Check out code from GitHub
-        uses: actions/checkout@v3.5.3
+        uses: actions/checkout@v3.6.0
       - name: Set up Python ${{ needs.prepare.outputs.python-version }}
         uses: actions/setup-python@v4.7.0
         id: python

.github/workflows/release-drafter.yml

@@ -11,7 +11,7 @@ jobs:
     name: Release Drafter
     steps:
       - name: Checkout the repository
-        uses: actions/checkout@v3.5.3
+        uses: actions/checkout@v3.6.0
         with:
           fetch-depth: 0
 

.github/workflows/sentry.yaml

@@ -10,7 +10,7 @@ jobs:
     runs-on: ubuntu-latest
     steps:
       - name: Check out code from GitHub
-        uses: actions/checkout@v3.5.3
+        uses: actions/checkout@v3.6.0
       - name: Sentry Release
         uses: getsentry/action-release@v1.4.1
         env:

requirements.txt

@@ -1,14 +1,14 @@
 aiodns==3.0.0
-aiohttp==3.8.4
-async_timeout==4.0.2
+aiohttp==3.8.5
+async_timeout==4.0.3
 atomicwrites-homeassistant==1.4.1
 attrs==23.1.0
-awesomeversion==23.5.0
+awesomeversion==23.8.0
 brotli==1.0.9
 ciso8601==2.3.0
 colorlog==6.7.0
 cpe==1.2.1
-cryptography==41.0.2
+cryptography==41.0.3
 debugpy==1.6.7
 deepmerge==1.1.0
 dirhash==0.2.1
@@ -20,7 +20,7 @@ pulsectl==23.5.2
 pyudev==0.24.1
 ruamel.yaml==0.17.21
 securetar==2023.3.0
-sentry-sdk==1.29.0
+sentry-sdk==1.29.2
 voluptuous==0.13.1
-dbus-fast==1.86.0
+dbus-fast==1.93.0
 typing_extensions==4.7.1

requirements_tests.txt

@@ -1,5 +1,5 @@
 black==23.7.0
-coverage==7.2.7
+coverage==7.3.0
 flake8-docstrings==1.7.0
 flake8==6.1.0
 pre-commit==3.3.3
@@ -11,6 +11,6 @@ pytest-cov==4.1.0
 pytest-timeout==2.1.0
 pytest==7.4.0
 pyupgrade==3.10.1
-time-machine==2.11.0
+time-machine==2.12.0
 typing_extensions==4.7.1
 urllib3==2.0.4

supervisor/addons/__init__.py

@@ -152,11 +152,15 @@ class AddonManager(CoreSysAttributes):
                 capture_exception(err)
 
     @Job(
+        name="addon_manager_install",
         conditions=ADDON_UPDATE_CONDITIONS,
         on_condition=AddonsJobError,
     )
     async def install(self, slug: str) -> None:
         """Install an add-on."""
+        if job := self.sys_jobs.get_job():
+            job.reference = slug
+
         if slug in self.local:
             raise AddonsError(f"Add-on {slug} is already installed", _LOGGER.warning)
         store = self.store.get(slug)
@@ -247,6 +251,7 @@ class AddonManager(CoreSysAttributes):
         _LOGGER.info("Add-on '%s' successfully removed", slug)
 
     @Job(
+        name="addon_manager_update",
         conditions=ADDON_UPDATE_CONDITIONS,
         on_condition=AddonsJobError,
     )
@@ -258,6 +263,9 @@ class AddonManager(CoreSysAttributes):
         Returns a coroutine that completes when addon has state 'started' (see addon.start)
         if addon is started after update. Else nothing is returned.
         """
+        if job := self.sys_jobs.get_job():
+            job.reference = slug
+
         if slug not in self.local:
             raise AddonsError(f"Add-on {slug} is not installed", _LOGGER.error)
         addon = self.local[slug]
@@ -307,6 +315,7 @@ class AddonManager(CoreSysAttributes):
         )
 
     @Job(
+        name="addon_manager_rebuild",
         conditions=[
             JobCondition.FREE_SPACE,
             JobCondition.INTERNET_HOST,
@@ -320,6 +329,9 @@ class AddonManager(CoreSysAttributes):
         Returns a coroutine that completes when addon has state 'started' (see addon.start)
         if addon is started after rebuild. Else nothing is returned.
         """
+        if job := self.sys_jobs.get_job():
+            job.reference = slug
+
         if slug not in self.local:
             raise AddonsError(f"Add-on {slug} is not installed", _LOGGER.error)
         addon = self.local[slug]
@@ -359,6 +371,7 @@ class AddonManager(CoreSysAttributes):
         )
 
     @Job(
+        name="addon_manager_restore",
         conditions=[
             JobCondition.FREE_SPACE,
             JobCondition.INTERNET_HOST,
@@ -374,6 +387,9 @@ class AddonManager(CoreSysAttributes):
         Returns a coroutine that completes when addon has state 'started' (see addon.start)
         if addon is started after restore. Else nothing is returned.
         """
+        if job := self.sys_jobs.get_job():
+            job.reference = slug
+
         if slug not in self.local:
             _LOGGER.debug("Add-on %s is not local available for restore", slug)
             addon = Addon(self.coresys, slug)
@@ -396,7 +412,10 @@ class AddonManager(CoreSysAttributes):
 
         return wait_for_start
 
-    @Job(conditions=[JobCondition.FREE_SPACE, JobCondition.INTERNET_HOST])
+    @Job(
+        name="addon_manager_repair",
+        conditions=[JobCondition.FREE_SPACE, JobCondition.INTERNET_HOST],
+    )
     async def repair(self) -> None:
         """Repair local add-ons."""
         needs_repair: list[Addon] = []

supervisor/addons/addon.py

@@ -129,54 +129,7 @@ class Addon(AddonModel):
         )
         self._listeners: list[EventListener] = []
         self._startup_event = asyncio.Event()
-
-        @Job(
-            name=f"addon_{slug}_restart_after_problem",
-            limit=JobExecutionLimit.THROTTLE_RATE_LIMIT,
-            throttle_period=WATCHDOG_THROTTLE_PERIOD,
-            throttle_max_calls=WATCHDOG_THROTTLE_MAX_CALLS,
-            on_condition=AddonsJobError,
-        )
-        async def restart_after_problem(addon: Addon, state: ContainerState):
-            """Restart unhealthy or failed addon."""
-            attempts = 0
-            while await addon.instance.current_state() == state:
-                if not addon.in_progress:
-                    _LOGGER.warning(
-                        "Watchdog found addon %s is %s, restarting...",
-                        addon.name,
-                        state.value,
-                    )
-                    try:
-                        if state == ContainerState.FAILED:
-                            # Ensure failed container is removed before attempting reanimation
-                            if attempts == 0:
-                                with suppress(DockerError):
-                                    await addon.instance.stop(remove_container=True)
-
-                            await (await addon.start())
-                        else:
-                            await (await addon.restart())
-                    except AddonsError as err:
-                        attempts = attempts + 1
-                        _LOGGER.error(
-                            "Watchdog restart of addon %s failed!", addon.name
-                        )
-                        capture_exception(err)
-                    else:
-                        break
-
-                if attempts >= WATCHDOG_MAX_ATTEMPTS:
-                    _LOGGER.critical(
-                        "Watchdog cannot restart addon %s, failed all %s attempts",
-                        addon.name,
-                        attempts,
-                    )
-                    break
-
-                await asyncio.sleep(WATCHDOG_RETRY_SECONDS)
-
-        self._restart_after_problem = restart_after_problem
+        self._startup_task: asyncio.Task | None = None
 
     def __repr__(self) -> str:
         """Return internal representation."""
@@ -608,6 +561,12 @@ class Addon(AddonModel):
 
     async def unload(self) -> None:
         """Unload add-on and remove data."""
+        if self._startup_task:
+            # If we were waiting on startup, cancel that and let the task finish before proceeding
+            self._startup_task.cancel(f"Removing add-on {self.name} from system")
+            with suppress(asyncio.CancelledError):
+                await self._startup_task
+
         for listener in self._listeners:
             self.sys_bus.remove_listener(listener)
 
@@ -699,13 +658,18 @@ class Addon(AddonModel):
     async def _wait_for_startup(self) -> None:
         """Wait for startup event to be set with timeout."""
         try:
-            await asyncio.wait_for(self._startup_event.wait(), STARTUP_TIMEOUT)
+            self._startup_task = self.sys_create_task(self._startup_event.wait())
+            await asyncio.wait_for(self._startup_task, STARTUP_TIMEOUT)
         except asyncio.TimeoutError:
             _LOGGER.warning(
                 "Timeout while waiting for addon %s to start, took more then %s seconds",
                 self.name,
                 STARTUP_TIMEOUT,
             )
+        except asyncio.CancelledError as err:
+            _LOGGER.info("Wait for addon startup task cancelled due to: %s", err)
+        finally:
+            self._startup_task = None
 
     async def start(self) -> Awaitable[None]:
         """Set options and start add-on.
@@ -779,10 +743,7 @@ class Addon(AddonModel):
             raise AddonsError() from err
 
     async def write_stdin(self, data) -> None:
-        """Write data to add-on stdin.
-
-        Return a coroutine.
-        """
+        """Write data to add-on stdin."""
         if not self.with_stdin:
             raise AddonsNotSupportedError(
                 f"Add-on {self.slug} does not support writing to stdin!", _LOGGER.error
@@ -877,7 +838,10 @@ class Addon(AddonModel):
                 await self._backup_command(self.backup_pre)
             elif is_running and self.backup_mode == AddonBackupMode.COLD:
                 _LOGGER.info("Shutdown add-on %s for cold backup", self.slug)
-                await self.instance.stop()
+                try:
+                    await self.instance.stop()
+                except DockerError as err:
+                    raise AddonsError() from err
 
             try:
                 _LOGGER.info("Building backup for add-on %s", self.slug)
@@ -950,6 +914,11 @@ class Addon(AddonModel):
                 self.slug, data[ATTR_USER], data[ATTR_SYSTEM], restore_image
             )
 
+            # Stop it first if its running
+            if await self.instance.is_running():
+                with suppress(DockerError):
+                    await self.instance.stop()
+
             # Check version / restore image
             version = data[ATTR_VERSION]
             if not await self.instance.exists():
@@ -967,9 +936,6 @@ class Addon(AddonModel):
                 _LOGGER.info("Restore/Update of image for addon %s", self.slug)
                 with suppress(DockerError):
                     await self.instance.update(version, restore_image)
-            else:
-                with suppress(DockerError):
-                    await self.instance.stop()
 
             # Restore data
             def _restore_data():
@@ -1019,6 +985,50 @@ class Addon(AddonModel):
         """
         return self.instance.check_trust()
 
+    @Job(
+        name="addon_restart_after_problem",
+        limit=JobExecutionLimit.GROUP_THROTTLE_RATE_LIMIT,
+        throttle_period=WATCHDOG_THROTTLE_PERIOD,
+        throttle_max_calls=WATCHDOG_THROTTLE_MAX_CALLS,
+        on_condition=AddonsJobError,
+    )
+    async def _restart_after_problem(self, state: ContainerState):
+        """Restart unhealthy or failed addon."""
+        attempts = 0
+        while await self.instance.current_state() == state:
+            if not self.in_progress:
+                _LOGGER.warning(
+                    "Watchdog found addon %s is %s, restarting...",
+                    self.name,
+                    state.value,
+                )
+                try:
+                    if state == ContainerState.FAILED:
+                        # Ensure failed container is removed before attempting reanimation
+                        if attempts == 0:
+                            with suppress(DockerError):
+                                await self.instance.stop(remove_container=True)
+
+                        await (await self.start())
+                    else:
+                        await (await self.restart())
+                except AddonsError as err:
+                    attempts = attempts + 1
+                    _LOGGER.error("Watchdog restart of addon %s failed!", self.name)
+                    capture_exception(err)
+                else:
+                    break
+
+            if attempts >= WATCHDOG_MAX_ATTEMPTS:
+                _LOGGER.critical(
+                    "Watchdog cannot restart addon %s, failed all %s attempts",
+                    self.name,
+                    attempts,
+                )
+                break
+
+            await asyncio.sleep(WATCHDOG_RETRY_SECONDS)
+
     async def container_state_changed(self, event: DockerContainerStateEvent) -> None:
         """Set addon state from container state."""
         if event.name != self.instance.name:
@@ -1053,4 +1063,4 @@ class Addon(AddonModel):
             ContainerState.STOPPED,
             ContainerState.UNHEALTHY,
         ]:
-            await self._restart_after_problem(self, event.state)
+            await self._restart_after_problem(event.state)

supervisor/addons/model.py

@@ -1,5 +1,6 @@
 """Init file for Supervisor add-ons."""
 from abc import ABC, abstractmethod
+from collections import defaultdict
 from collections.abc import Awaitable, Callable
 from contextlib import suppress
 import logging
@@ -79,9 +80,11 @@ from ..const import (
     AddonStage,
     AddonStartup,
 )
-from ..coresys import CoreSys, CoreSysAttributes
+from ..coresys import CoreSys
 from ..docker.const import Capabilities
 from ..exceptions import AddonsNotSupportedError
+from ..jobs.const import JOB_GROUP_ADDON
+from ..jobs.job_group import JobGroup
 from .const import ATTR_BACKUP, ATTR_CODENOTARY, AddonBackupMode
 from .options import AddonOptions, UiOptions
 from .validate import RE_SERVICE, RE_VOLUME
@@ -91,12 +94,14 @@ _LOGGER: logging.Logger = logging.getLogger(__name__)
 Data = dict[str, Any]
 
 
-class AddonModel(CoreSysAttributes, ABC):
+class AddonModel(JobGroup, ABC):
     """Add-on Data layout."""
 
     def __init__(self, coresys: CoreSys, slug: str):
         """Initialize data holder."""
-        self.coresys: CoreSys = coresys
+        super().__init__(
+            coresys, JOB_GROUP_ADDON.format_map(defaultdict(str, slug=slug)), slug
+        )
         self.slug: str = slug
 
     @property

supervisor/addons/validate.py

@@ -143,6 +143,8 @@ RE_MACHINE = re.compile(
     r"|raspberrypi3"
     r"|raspberrypi4-64"
     r"|raspberrypi4"
+    r"|yellow"
+    r"|green"
     r"|tinker"
     r")$"
 )

supervisor/api/audio.py

@@ -1,11 +1,11 @@
 """Init file for Supervisor Audio RESTful API."""
 import asyncio
 from collections.abc import Awaitable
+from dataclasses import asdict
 import logging
 from typing import Any
 
 from aiohttp import web
-import attr
 import voluptuous as vol
 
 from ..const import (
@@ -76,15 +76,11 @@ class APIAudio(CoreSysAttributes):
             ATTR_UPDATE_AVAILABLE: self.sys_plugins.audio.need_update,
             ATTR_HOST: str(self.sys_docker.network.audio),
             ATTR_AUDIO: {
-                ATTR_CARD: [attr.asdict(card) for card in self.sys_host.sound.cards],
-                ATTR_INPUT: [
-                    attr.asdict(stream) for stream in self.sys_host.sound.inputs
-                ],
-                ATTR_OUTPUT: [
-                    attr.asdict(stream) for stream in self.sys_host.sound.outputs
-                ],
+                ATTR_CARD: [asdict(card) for card in self.sys_host.sound.cards],
+                ATTR_INPUT: [asdict(stream) for stream in self.sys_host.sound.inputs],
+                ATTR_OUTPUT: [asdict(stream) for stream in self.sys_host.sound.outputs],
                 ATTR_APPLICATION: [
-                    attr.asdict(stream) for stream in self.sys_host.sound.applications
+                    asdict(stream) for stream in self.sys_host.sound.applications
                 ],
             },
         }

supervisor/api/ingress.py

@@ -21,11 +21,18 @@ from ..const import (
     ATTR_ICON,
     ATTR_PANELS,
     ATTR_SESSION,
+    ATTR_SESSION_DATA_USER_ID,
     ATTR_TITLE,
+    HEADER_REMOTE_USER_DISPLAY_NAME,
+    HEADER_REMOTE_USER_ID,
+    HEADER_REMOTE_USER_NAME,
     HEADER_TOKEN,
     HEADER_TOKEN_OLD,
+    IngressSessionData,
+    IngressSessionDataUser,
 )
 from ..coresys import CoreSysAttributes
+from ..exceptions import HomeAssistantAPIError
 from .const import COOKIE_INGRESS
 from .utils import api_process, api_validate, require_home_assistant
 
@@ -33,10 +40,23 @@ _LOGGER: logging.Logger = logging.getLogger(__name__)
 
 VALIDATE_SESSION_DATA = vol.Schema({ATTR_SESSION: str})
 
+"""Expected optional payload of create session request"""
+SCHEMA_INGRESS_CREATE_SESSION_DATA = vol.Schema(
+    {
+        vol.Optional(ATTR_SESSION_DATA_USER_ID): str,
+    }
+)
+
 
 class APIIngress(CoreSysAttributes):
     """Ingress view to handle add-on webui routing."""
 
+    _list_of_users: list[IngressSessionDataUser]
+
+    def __init__(self) -> None:
+        """Initialize APIIngress."""
+        self._list_of_users = []
+
     def _extract_addon(self, request: web.Request) -> Addon:
         """Return addon, throw an exception it it doesn't exist."""
         token = request.match_info.get("token")
@@ -71,7 +91,19 @@ class APIIngress(CoreSysAttributes):
     @require_home_assistant
     async def create_session(self, request: web.Request) -> dict[str, Any]:
         """Create a new session."""
-        session = self.sys_ingress.create_session()
+        schema_ingress_config_session_data = await api_validate(
+            SCHEMA_INGRESS_CREATE_SESSION_DATA, request
+        )
+        data: IngressSessionData | None = None
+
+        if ATTR_SESSION_DATA_USER_ID in schema_ingress_config_session_data:
+            user = await self._find_user_by_id(
+                schema_ingress_config_session_data[ATTR_SESSION_DATA_USER_ID]
+            )
+            if user:
+                data = IngressSessionData(user)
+
+        session = self.sys_ingress.create_session(data)
         return {ATTR_SESSION: session}
 
     @api_process
@@ -99,13 +131,14 @@ class APIIngress(CoreSysAttributes):
         # Process requests
         addon = self._extract_addon(request)
         path = request.match_info.get("path")
+        session_data = self.sys_ingress.get_session_data(session)
         try:
             # Websocket
             if _is_websocket(request):
-                return await self._handle_websocket(request, addon, path)
+                return await self._handle_websocket(request, addon, path, session_data)
 
             # Request
-            return await self._handle_request(request, addon, path)
+            return await self._handle_request(request, addon, path, session_data)
 
         except aiohttp.ClientError as err:
             _LOGGER.error("Ingress error: %s", err)
@@ -113,7 +146,11 @@ class APIIngress(CoreSysAttributes):
         raise HTTPBadGateway()
 
     async def _handle_websocket(
-        self, request: web.Request, addon: Addon, path: str
+        self,
+        request: web.Request,
+        addon: Addon,
+        path: str,
+        session_data: IngressSessionData | None,
     ) -> web.WebSocketResponse:
         """Ingress route for websocket."""
         if hdrs.SEC_WEBSOCKET_PROTOCOL in request.headers:
@@ -131,7 +168,7 @@ class APIIngress(CoreSysAttributes):
 
         # Preparing
         url = self._create_url(addon, path)
-        source_header = _init_header(request, addon)
+        source_header = _init_header(request, addon, session_data)
 
         # Support GET query
         if request.query_string:
@@ -157,11 +194,15 @@ class APIIngress(CoreSysAttributes):
         return ws_server
 
     async def _handle_request(
-        self, request: web.Request, addon: Addon, path: str
+        self,
+        request: web.Request,
+        addon: Addon,
+        path: str,
+        session_data: IngressSessionData | None,
     ) -> web.Response | web.StreamResponse:
         """Ingress route for request."""
         url = self._create_url(addon, path)
-        source_header = _init_header(request, addon)
+        source_header = _init_header(request, addon, session_data)
 
         # Passing the raw stream breaks requests for some webservers
         # since we just need it for POST requests really, for all other methods
@@ -217,11 +258,33 @@ class APIIngress(CoreSysAttributes):
 
             return response
 
+    async def _find_user_by_id(self, user_id: str) -> IngressSessionDataUser | None:
+        """Find user object by the user's ID."""
+        try:
+            list_of_users = await self.sys_homeassistant.get_users()
+        except (HomeAssistantAPIError, TypeError) as err:
+            _LOGGER.error(
+                "%s error occurred while requesting list of users: %s", type(err), err
+            )
+            return None
 
-def _init_header(request: web.Request, addon: str) -> CIMultiDict | dict[str, str]:
+        if list_of_users is not None:
+            self._list_of_users = list_of_users
+
+        return next((user for user in self._list_of_users if user.id == user_id), None)
+
+
+def _init_header(
+    request: web.Request, addon: Addon, session_data: IngressSessionData | None
+) -> CIMultiDict | dict[str, str]:
     """Create initial header."""
     headers = {}
 
+    if session_data is not None:
+        headers[HEADER_REMOTE_USER_ID] = session_data.user.id
+        headers[HEADER_REMOTE_USER_NAME] = session_data.user.username
+        headers[HEADER_REMOTE_USER_DISPLAY_NAME] = session_data.user.display_name
+
     # filter flags
     for name, value in request.headers.items():
         if name in (
@@ -234,6 +297,9 @@ def _init_header(request: web.Request, addon: str) -> CIMultiDict | dict[str, st
             hdrs.SEC_WEBSOCKET_KEY,
             istr(HEADER_TOKEN),
             istr(HEADER_TOKEN_OLD),
+            istr(HEADER_REMOTE_USER_ID),
+            istr(HEADER_REMOTE_USER_NAME),
+            istr(HEADER_REMOTE_USER_DISPLAY_NAME),
         ):
             continue
         headers[name] = value

supervisor/api/network.py

@@ -277,6 +277,7 @@ class APINetwork(CoreSysAttributes):
             )
 
         vlan_interface = Interface(
+            "",
             "",
             "",
             True,

supervisor/backups/manager.py

@@ -110,6 +110,10 @@ class BackupManager(FileConfiguration, CoreSysAttributes):
         backup.store_repositories()
         backup.store_dockerconfig()
 
+        # Add backup ID to job
+        if job := self.sys_jobs.get_job():
+            job.reference = backup.slug
+
         return backup
 
     def load(self):
@@ -224,7 +228,10 @@ class BackupManager(FileConfiguration, CoreSysAttributes):
         finally:
             self.sys_core.state = CoreState.RUNNING
 
-    @Job(conditions=[JobCondition.FREE_SPACE, JobCondition.RUNNING])
+    @Job(
+        name="backup_manager_full_backup",
+        conditions=[JobCondition.FREE_SPACE, JobCondition.RUNNING],
+    )
     async def do_backup_full(
         self,
         name="",
@@ -250,7 +257,10 @@ class BackupManager(FileConfiguration, CoreSysAttributes):
                 _LOGGER.info("Creating full backup with slug %s completed", backup.slug)
             return backup
 
-    @Job(conditions=[JobCondition.FREE_SPACE, JobCondition.RUNNING])
+    @Job(
+        name="backup_manager_partial_backup",
+        conditions=[JobCondition.FREE_SPACE, JobCondition.RUNNING],
+    )
     async def do_backup_partial(
         self,
         name: str = "",
@@ -371,16 +381,21 @@ class BackupManager(FileConfiguration, CoreSysAttributes):
                 await self.sys_homeassistant.core.restart()
 
     @Job(
+        name="backup_manager_full_restore",
         conditions=[
             JobCondition.FREE_SPACE,
             JobCondition.HEALTHY,
             JobCondition.INTERNET_HOST,
             JobCondition.INTERNET_SYSTEM,
             JobCondition.RUNNING,
-        ]
+        ],
     )
     async def do_restore_full(self, backup: Backup, password=None):
         """Restore a backup."""
+        # Add backup ID to job
+        if job := self.sys_jobs.get_job():
+            job.reference = backup.slug
+
         if self.lock.locked():
             _LOGGER.error("A backup/restore process is already running")
             return False
@@ -418,13 +433,14 @@ class BackupManager(FileConfiguration, CoreSysAttributes):
                 _LOGGER.info("Full-Restore %s done", backup.slug)
 
     @Job(
+        name="backup_manager_partial_restore",
         conditions=[
             JobCondition.FREE_SPACE,
             JobCondition.HEALTHY,
             JobCondition.INTERNET_HOST,
             JobCondition.INTERNET_SYSTEM,
             JobCondition.RUNNING,
-        ]
+        ],
     )
     async def do_restore_partial(
         self,
@@ -435,6 +451,10 @@ class BackupManager(FileConfiguration, CoreSysAttributes):
         password: str | None = None,
     ):
         """Restore a backup."""
+        # Add backup ID to job
+        if job := self.sys_jobs.get_job():
+            job.reference = backup.slug
+
         if self.lock.locked():
             _LOGGER.error("A backup/restore process is already running")
             return False

supervisor/const.py

@@ -1,4 +1,5 @@
 """Constants file for Supervisor."""
+from dataclasses import dataclass
 from enum import Enum
 from ipaddress import ip_network
 from pathlib import Path
@@ -69,6 +70,9 @@ JSON_RESULT = "result"
 RESULT_ERROR = "error"
 RESULT_OK = "ok"
 
+HEADER_REMOTE_USER_ID = "X-Remote-User-Id"
+HEADER_REMOTE_USER_NAME = "X-Remote-User-Name"
+HEADER_REMOTE_USER_DISPLAY_NAME = "X-Remote-User-Display-Name"
 HEADER_TOKEN_OLD = "X-Hassio-Key"
 HEADER_TOKEN = "X-Supervisor-Token"
 
@@ -271,6 +275,9 @@ ATTR_SERVERS = "servers"
 ATTR_SERVICE = "service"
 ATTR_SERVICES = "services"
 ATTR_SESSION = "session"
+ATTR_SESSION_DATA = "session_data"
+ATTR_SESSION_DATA_USER = "user"
+ATTR_SESSION_DATA_USER_ID = "user_id"
 ATTR_SIGNAL = "signal"
 ATTR_SIZE = "size"
 ATTR_SLUG = "slug"
@@ -464,6 +471,22 @@ class CpuArch(str, Enum):
     AMD64 = "amd64"
 
 
+@dataclass
+class IngressSessionDataUser:
+    """Format of an IngressSessionDataUser object."""
+
+    id: str
+    display_name: str
+    username: str
+
+
+@dataclass
+class IngressSessionData:
+    """Format of an IngressSessionData object."""
+
+    user: IngressSessionDataUser
+
+
 STARTING_STATES = [
     CoreState.INITIALIZE,
     CoreState.STARTUP,

supervisor/core.py

@@ -70,13 +70,16 @@ class Core(CoreSysAttributes):
             )
         finally:
             self._state = new_state
-            self.sys_bus.fire_event(BusEvent.SUPERVISOR_STATE_CHANGE, new_state)
 
-            # These will be received by HA after startup has completed which won't make sense
-            if new_state not in STARTING_STATES:
-                self.sys_homeassistant.websocket.supervisor_update_event(
-                    "info", {"state": new_state}
-                )
+            # Don't attempt to notify anyone on CLOSE as we're about to stop the event loop
+            if new_state != CoreState.CLOSE:
+                self.sys_bus.fire_event(BusEvent.SUPERVISOR_STATE_CHANGE, new_state)
+
+                # These will be received by HA after startup has completed which won't make sense
+                if new_state not in STARTING_STATES:
+                    self.sys_homeassistant.websocket.supervisor_update_event(
+                        "info", {"state": new_state}
+                    )
 
     async def connect(self):
         """Connect Supervisor container."""

supervisor/data/arch.json

@@ -6,6 +6,7 @@
   "raspberrypi4": ["armv7", "armhf"],
   "raspberrypi4-64": ["aarch64", "armv7", "armhf"],
   "yellow": ["aarch64", "armv7", "armhf"],
+  "green": ["aarch64", "armv7", "armhf"],
   "tinker": ["armv7", "armhf"],
   "odroid-c2": ["aarch64", "armv7", "armhf"],
   "odroid-c4": ["aarch64", "armv7", "armhf"],

supervisor/dbus/const.py

@@ -144,6 +144,7 @@ DBUS_ATTR_OPERATION = "Operation"
 DBUS_ATTR_OPTIONS = "Options"
 DBUS_ATTR_PARSER_VERSION = "ParserVersion"
 DBUS_ATTR_PARTITIONS = "Partitions"
+DBUS_ATTR_PATH = "Path"
 DBUS_ATTR_POWER_LED = "PowerLED"
 DBUS_ATTR_PRIMARY_CONNECTION = "PrimaryConnection"
 DBUS_ATTR_READ_ONLY = "ReadOnly"

supervisor/dbus/network/configuration.py

@@ -66,7 +66,7 @@ class IpProperties:
 
 
 @dataclass(slots=True)
-class DeviceProperties:
-    """Device properties object for Network Manager."""
+class MatchProperties:
+    """Match properties object for Network Manager."""
 
-    match_device: str | None
+    path: list[str] | None = None

supervisor/dbus/network/interface.py

@@ -11,6 +11,7 @@ from ..const import (
     DBUS_ATTR_DRIVER,
     DBUS_ATTR_HWADDRESS,
     DBUS_ATTR_MANAGED,
+    DBUS_ATTR_PATH,
     DBUS_IFACE_DEVICE,
     DBUS_NAME_NM,
     DBUS_OBJECT_BASE,
@@ -74,6 +75,12 @@ class NetworkInterface(DBusInterfaceProxy):
         """Return hardware address (i.e. mac address) of device."""
         return self.properties[DBUS_ATTR_HWADDRESS]
 
+    @property
+    @dbus_property
+    def path(self) -> str:
+        """Return The path of the device as exposed by the udev property ID_PATH."""
+        return self.properties[DBUS_ATTR_PATH]
+
     @property
     def connection(self) -> NetworkConnection | None:
         """Return the connection used for this interface."""

supervisor/dbus/network/setting/__init__.py

@@ -11,9 +11,9 @@ from ...interface import DBusInterface
 from ...utils import dbus_connected
 from ..configuration import (
     ConnectionProperties,
-    DeviceProperties,
     EthernetProperties,
     IpProperties,
+    MatchProperties,
     VlanProperties,
     WirelessProperties,
     WirelessSecurityProperties,
@@ -26,7 +26,8 @@ CONF_ATTR_802_WIRELESS_SECURITY = "802-11-wireless-security"
 CONF_ATTR_VLAN = "vlan"
 CONF_ATTR_IPV4 = "ipv4"
 CONF_ATTR_IPV6 = "ipv6"
-CONF_ATTR_DEVICE = "device"
+CONF_ATTR_MATCH = "match"
+CONF_ATTR_PATH = "path"
 
 ATTR_ID = "id"
 ATTR_UUID = "uuid"
@@ -37,7 +38,7 @@ ATTR_POWERSAVE = "powersave"
 ATTR_AUTH_ALG = "auth-alg"
 ATTR_KEY_MGMT = "key-mgmt"
 ATTR_INTERFACE_NAME = "interface-name"
-ATTR_MATCH_DEVICE = "match-device"
+ATTR_PATH = "path"
 
 IPV4_6_IGNORE_FIELDS = [
     "addresses",
@@ -88,7 +89,7 @@ class NetworkSetting(DBusInterface):
         self._vlan: VlanProperties | None = None
         self._ipv4: IpProperties | None = None
         self._ipv6: IpProperties | None = None
-        self._device: DeviceProperties | None = None
+        self._match: MatchProperties | None = None
 
     @property
     def connection(self) -> ConnectionProperties | None:
@@ -126,9 +127,9 @@ class NetworkSetting(DBusInterface):
         return self._ipv6
 
     @property
-    def device(self) -> DeviceProperties | None:
-        """Return device properties if any."""
-        return self._device
+    def match(self) -> MatchProperties | None:
+        """Return match properties if any."""
+        return self._match
 
     @dbus_connected
     async def get_settings(self) -> dict[str, Any]:
@@ -166,7 +167,7 @@ class NetworkSetting(DBusInterface):
             CONF_ATTR_IPV6,
             ignore_current_value=IPV4_6_IGNORE_FIELDS,
         )
-        _merge_settings_attribute(new_settings, settings, CONF_ATTR_DEVICE)
+        _merge_settings_attribute(new_settings, settings, CONF_ATTR_MATCH)
 
         await self.dbus.Settings.Connection.call_update(new_settings)
 
@@ -233,7 +234,5 @@ class NetworkSetting(DBusInterface):
                 data[CONF_ATTR_IPV6].get(ATTR_METHOD),
             )
 
-        if CONF_ATTR_DEVICE in data:
-            self._device = DeviceProperties(
-                data[CONF_ATTR_DEVICE].get(ATTR_MATCH_DEVICE)
-            )
+        if CONF_ATTR_MATCH in data:
+            self._match = MatchProperties(data[CONF_ATTR_MATCH].get(ATTR_PATH))

supervisor/dbus/network/setting/generate.py

@@ -9,14 +9,14 @@ from dbus_fast import Variant
 
 from . import (
     ATTR_ASSIGNED_MAC,
-    ATTR_MATCH_DEVICE,
     CONF_ATTR_802_ETHERNET,
     CONF_ATTR_802_WIRELESS,
     CONF_ATTR_802_WIRELESS_SECURITY,
     CONF_ATTR_CONNECTION,
-    CONF_ATTR_DEVICE,
     CONF_ATTR_IPV4,
     CONF_ATTR_IPV6,
+    CONF_ATTR_MATCH,
+    CONF_ATTR_PATH,
     CONF_ATTR_VLAN,
 )
 from ....host.const import InterfaceMethod, InterfaceType
@@ -59,11 +59,10 @@ def get_connection_from_interface(
     }
 
     if interface.type != InterfaceType.VLAN:
-        conn[CONF_ATTR_DEVICE] = {
-            ATTR_MATCH_DEVICE: Variant(
-                "s", f"mac:{interface.mac},interface-name:{interface.name}"
-            )
-        }
+        if interface.path:
+            conn[CONF_ATTR_MATCH] = {CONF_ATTR_PATH: Variant("as", [interface.path])}
+        else:
+            conn[CONF_ATTR_CONNECTION]["interface-name"] = Variant("s", interface.name)
 
     ipv4 = {}
     if not interface.ipv4 or interface.ipv4.method == InterfaceMethod.AUTO:

supervisor/docker/addon.py

@@ -36,14 +36,15 @@ from ..exceptions import (
     CoreDNSError,
     DBusError,
     DockerError,
+    DockerJobError,
     DockerNotFound,
     HardwareNotFound,
 )
 from ..hardware.const import PolicyGroup
 from ..hardware.data import Device
-from ..jobs.decorator import Job, JobCondition, JobExecutionLimit
+from ..jobs.const import JobCondition, JobExecutionLimit
+from ..jobs.decorator import Job
 from ..resolution.const import ContextType, IssueType, SuggestionType
-from ..utils import process_lock
 from ..utils.sentry import capture_exception
 from .const import (
     ENV_TIME,
@@ -73,8 +74,8 @@ class DockerAddon(DockerInterface):
 
     def __init__(self, coresys: CoreSys, addon: Addon):
         """Initialize Docker Home Assistant wrapper."""
-        super().__init__(coresys)
         self.addon: Addon = addon
+        super().__init__(coresys)
 
         self._hw_listener: EventListener | None = None
 
@@ -493,7 +494,12 @@ class DockerAddon(DockerInterface):
 
         return mounts
 
-    async def _run(self) -> None:
+    @Job(
+        name="docker_addon_run",
+        limit=JobExecutionLimit.GROUP_ONCE,
+        on_condition=DockerJobError,
+    )
+    async def run(self) -> None:
         """Run Docker image."""
         if await self.is_running():
             return
@@ -503,7 +509,7 @@ class DockerAddon(DockerInterface):
             _LOGGER.warning("%s running with disabled protected mode!", self.addon.name)
 
         # Cleanup
-        await self._stop()
+        await self.stop()
 
         # Don't set a hostname if no separate UTS namespace is used
         hostname = None if self.uts_mode else self.addon.hostname
@@ -563,7 +569,12 @@ class DockerAddon(DockerInterface):
                 BusEvent.HARDWARE_NEW_DEVICE, self._hardware_events
             )
 
-    async def _update(
+    @Job(
+        name="docker_addon_update",
+        limit=JobExecutionLimit.GROUP_ONCE,
+        on_condition=DockerJobError,
+    )
+    async def update(
         self, version: AwesomeVersion, image: str | None = None, latest: bool = False
     ) -> None:
         """Update a docker image."""
@@ -574,15 +585,20 @@ class DockerAddon(DockerInterface):
         )
 
         # Update docker image
-        await self._install(
+        await self.install(
             version, image=image, latest=latest, need_build=self.addon.latest_need_build
         )
 
         # Stop container & cleanup
         with suppress(DockerError):
-            await self._stop()
+            await self.stop()
 
-    async def _install(
+    @Job(
+        name="docker_addon_install",
+        limit=JobExecutionLimit.GROUP_ONCE,
+        on_condition=DockerJobError,
+    )
+    async def install(
         self,
         version: AwesomeVersion,
         image: str | None = None,
@@ -595,7 +611,7 @@ class DockerAddon(DockerInterface):
         if need_build is None and self.addon.need_build or need_build:
             await self._build(version)
         else:
-            await super()._install(version, image, latest, arch)
+            await super().install(version, image, latest, arch)
 
     async def _build(self, version: AwesomeVersion) -> None:
         """Build a Docker container."""
@@ -632,14 +648,22 @@ class DockerAddon(DockerInterface):
 
         _LOGGER.info("Build %s:%s done", self.image, version)
 
-    @process_lock
+    @Job(
+        name="docker_addon_export_image",
+        limit=JobExecutionLimit.GROUP_ONCE,
+        on_condition=DockerJobError,
+    )
     def export_image(self, tar_file: Path) -> Awaitable[None]:
         """Export current images into a tar file."""
         return self.sys_run_in_executor(
             self.sys_docker.export_image, self.image, self.version, tar_file
         )
 
-    @process_lock
+    @Job(
+        name="docker_addon_import_image",
+        limit=JobExecutionLimit.GROUP_ONCE,
+        on_condition=DockerJobError,
+    )
     async def import_image(self, tar_file: Path) -> None:
         """Import a tar file as image."""
         docker_image = await self.sys_run_in_executor(
@@ -650,9 +674,13 @@ class DockerAddon(DockerInterface):
             _LOGGER.info("Importing image %s and version %s", tar_file, self.version)
 
             with suppress(DockerError):
-                await self._cleanup()
+                await self.cleanup()
 
-    @process_lock
+    @Job(
+        name="docker_addon_write_stdin",
+        limit=JobExecutionLimit.GROUP_ONCE,
+        on_condition=DockerJobError,
+    )
     async def write_stdin(self, data: bytes) -> None:
         """Write to add-on stdin."""
         if not await self.is_running():
@@ -682,7 +710,12 @@ class DockerAddon(DockerInterface):
             _LOGGER.error("Can't write to %s stdin: %s", self.name, err)
             raise DockerError() from err
 
-    async def _stop(self, remove_container: bool = True) -> None:
+    @Job(
+        name="docker_addon_stop",
+        limit=JobExecutionLimit.GROUP_ONCE,
+        on_condition=DockerJobError,
+    )
+    async def stop(self, remove_container: bool = True) -> None:
         """Stop/remove Docker container."""
         # DNS
         if self.ip_address != NO_ADDDRESS:
@@ -697,7 +730,7 @@ class DockerAddon(DockerInterface):
             self.sys_bus.remove_listener(self._hw_listener)
             self._hw_listener = None
 
-        await super()._stop(remove_container)
+        await super().stop(remove_container)
 
     async def _validate_trust(
         self, image_id: str, image: str, version: AwesomeVersion
@@ -709,7 +742,11 @@ class DockerAddon(DockerInterface):
         checksum = image_id.partition(":")[2]
         return await self.sys_security.verify_content(self.addon.codenotary, checksum)
 
-    @Job(conditions=[JobCondition.OS_AGENT], limit=JobExecutionLimit.SINGLE_WAIT)
+    @Job(
+        name="docker_addon_hardware_events",
+        conditions=[JobCondition.OS_AGENT],
+        limit=JobExecutionLimit.SINGLE_WAIT,
+    )
     async def _hardware_events(self, device: Device) -> None:
         """Process Hardware events for adjust device access."""
         if not any(

supervisor/docker/audio.py

@@ -6,7 +6,10 @@ from docker.types import Mount
 
 from ..const import DOCKER_CPU_RUNTIME_ALLOCATION, MACHINE_ID
 from ..coresys import CoreSysAttributes
+from ..exceptions import DockerJobError
 from ..hardware.const import PolicyGroup
+from ..jobs.const import JobExecutionLimit
+from ..jobs.decorator import Job
 from .const import (
     ENV_TIME,
     MOUNT_DBUS,
@@ -82,13 +85,18 @@ class DockerAudio(DockerInterface, CoreSysAttributes):
             return None
         return DOCKER_CPU_RUNTIME_ALLOCATION
 
-    async def _run(self) -> None:
+    @Job(
+        name="docker_audio_run",
+        limit=JobExecutionLimit.GROUP_ONCE,
+        on_condition=DockerJobError,
+    )
+    async def run(self) -> None:
         """Run Docker image."""
         if await self.is_running():
             return
 
         # Cleanup
-        await self._stop()
+        await self.stop()
 
         # Create & Run container
         docker_container = await self.sys_run_in_executor(

supervisor/docker/cli.py

@@ -2,6 +2,9 @@
 import logging
 
 from ..coresys import CoreSysAttributes
+from ..exceptions import DockerJobError
+from ..jobs.const import JobExecutionLimit
+from ..jobs.decorator import Job
 from .const import ENV_TIME, ENV_TOKEN
 from .interface import DockerInterface
 
@@ -23,13 +26,18 @@ class DockerCli(DockerInterface, CoreSysAttributes):
         """Return name of Docker container."""
         return CLI_DOCKER_NAME
 
-    async def _run(self) -> None:
+    @Job(
+        name="docker_cli_run",
+        limit=JobExecutionLimit.GROUP_ONCE,
+        on_condition=DockerJobError,
+    )
+    async def run(self) -> None:
         """Run Docker image."""
         if await self.is_running():
             return
 
         # Cleanup
-        await self._stop()
+        await self.stop()
 
         # Create & Run container
         docker_container = await self.sys_run_in_executor(

supervisor/docker/dns.py

@@ -4,6 +4,9 @@ import logging
 from docker.types import Mount
 
 from ..coresys import CoreSysAttributes
+from ..exceptions import DockerJobError
+from ..jobs.const import JobExecutionLimit
+from ..jobs.decorator import Job
 from .const import ENV_TIME, MOUNT_DBUS, MountType
 from .interface import DockerInterface
 
@@ -25,13 +28,18 @@ class DockerDNS(DockerInterface, CoreSysAttributes):
         """Return name of Docker container."""
         return DNS_DOCKER_NAME
 
-    async def _run(self) -> None:
+    @Job(
+        name="docker_dns_run",
+        limit=JobExecutionLimit.GROUP_ONCE,
+        on_condition=DockerJobError,
+    )
+    async def run(self) -> None:
         """Run Docker image."""
         if await self.is_running():
             return
 
         # Cleanup
-        await self._stop()
+        await self.stop()
 
         # Create & Run container
         docker_container = await self.sys_run_in_executor(

supervisor/docker/homeassistant.py

@@ -7,9 +7,11 @@ from awesomeversion import AwesomeVersion, AwesomeVersionCompareException
 from docker.types import Mount
 
 from ..const import LABEL_MACHINE, MACHINE_ID
+from ..exceptions import DockerJobError
 from ..hardware.const import PolicyGroup
 from ..homeassistant.const import LANDINGPAGE
-from ..utils import process_lock
+from ..jobs.const import JobExecutionLimit
+from ..jobs.decorator import Job
 from .const import (
     ENV_TIME,
     ENV_TOKEN,
@@ -131,13 +133,18 @@ class DockerHomeAssistant(DockerInterface):
 
         return mounts
 
-    async def _run(self) -> None:
+    @Job(
+        name="docker_home_assistant_run",
+        limit=JobExecutionLimit.GROUP_ONCE,
+        on_condition=DockerJobError,
+    )
+    async def run(self) -> None:
         """Run Docker image."""
         if await self.is_running():
             return
 
         # Cleanup
-        await self._stop()
+        await self.stop()
 
         # Create & Run container
         docker_container = await self.sys_run_in_executor(
@@ -173,7 +180,11 @@ class DockerHomeAssistant(DockerInterface):
             "Starting Home Assistant %s with version %s", self.image, self.version
         )
 
-    @process_lock
+    @Job(
+        name="docker_home_assistant_execute_command",
+        limit=JobExecutionLimit.GROUP_ONCE,
+        on_condition=DockerJobError,
+    )
     async def execute_command(self, command: str) -> CommandReturn:
         """Create a temporary container and run command."""
         return await self.sys_run_in_executor(

supervisor/docker/interface.py

@@ -2,12 +2,14 @@
 from __future__ import annotations
 
 import asyncio
+from collections import defaultdict
 from collections.abc import Awaitable
 from contextlib import suppress
 import logging
 import re
 from time import time
 from typing import Any
+from uuid import uuid4
 
 from awesomeversion import AwesomeVersion
 from awesomeversion.strategy import AwesomeVersionStrategy
@@ -24,18 +26,21 @@ from ..const import (
     BusEvent,
     CpuArch,
 )
-from ..coresys import CoreSys, CoreSysAttributes
+from ..coresys import CoreSys
 from ..exceptions import (
     CodeNotaryError,
     CodeNotaryUntrusted,
     DockerAPIError,
     DockerError,
+    DockerJobError,
     DockerNotFound,
     DockerRequestError,
     DockerTrustError,
 )
+from ..jobs.const import JOB_GROUP_DOCKER_INTERFACE, JobExecutionLimit
+from ..jobs.decorator import Job
+from ..jobs.job_group import JobGroup
 from ..resolution.const import ContextType, IssueType, SuggestionType
-from ..utils import process_lock
 from ..utils.sentry import capture_exception
 from .const import ContainerState, RestartPolicy
 from .manager import CommandReturn
@@ -73,11 +78,18 @@ def _container_state_from_model(docker_container: Container) -> ContainerState:
     return ContainerState.STOPPED
 
 
-class DockerInterface(CoreSysAttributes):
+class DockerInterface(JobGroup):
     """Docker Supervisor interface."""
 
     def __init__(self, coresys: CoreSys):
         """Initialize Docker base wrapper."""
+        super().__init__(
+            coresys,
+            JOB_GROUP_DOCKER_INTERFACE.format_map(
+                defaultdict(str, name=self.name or uuid4().hex)
+            ),
+            self.name,
+        )
         self.coresys: CoreSys = coresys
         self._meta: dict[str, Any] | None = None
         self.lock: asyncio.Lock = asyncio.Lock()
@@ -204,25 +216,19 @@ class DockerInterface(CoreSysAttributes):
 
         await self.sys_run_in_executor(self.sys_docker.docker.login, **credentials)
 
-    @process_lock
-    def install(
-        self,
-        version: AwesomeVersion,
-        image: str | None = None,
-        latest: bool = False,
-        arch: CpuArch | None = None,
-    ) -> Awaitable[None]:
-        """Pull docker image."""
-        return self._install(version, image, latest, arch)
-
-    async def _install(
+    @Job(
+        name="docker_interface_install",
+        limit=JobExecutionLimit.GROUP_ONCE,
+        on_condition=DockerJobError,
+    )
+    async def install(
         self,
         version: AwesomeVersion,
         image: str | None = None,
         latest: bool = False,
         arch: CpuArch | None = None,
     ) -> None:
-        """Pull Docker image."""
+        """Pull docker image."""
         image = image or self.image
         arch = arch or self.sys_arch.supervisor
 
@@ -328,17 +334,15 @@ class DockerInterface(CoreSysAttributes):
 
         return _container_state_from_model(docker_container)
 
-    @process_lock
-    def attach(
+    @Job(
+        name="docker_interface_attach",
+        limit=JobExecutionLimit.GROUP_ONCE,
+        on_condition=DockerJobError,
+    )
+    async def attach(
         self, version: AwesomeVersion, *, skip_state_event_if_down: bool = False
-    ) -> Awaitable[None]:
-        """Attach to running Docker container."""
-        return self._attach(version, skip_state_event_if_down)
-
-    async def _attach(
-        self, version: AwesomeVersion, skip_state_event_if_down: bool = False
     ) -> None:
-        """Attach to running docker container."""
+        """Attach to running Docker container."""
         with suppress(docker.errors.DockerException, requests.RequestException):
             docker_container = await self.sys_run_in_executor(
                 self.sys_docker.containers.get, self.name
@@ -370,21 +374,21 @@ class DockerInterface(CoreSysAttributes):
             raise DockerError()
         _LOGGER.info("Attaching to %s with version %s", self.image, self.version)
 
-    @process_lock
-    def run(self) -> Awaitable[None]:
-        """Run Docker image."""
-        return self._run()
-
-    async def _run(self) -> None:
+    @Job(
+        name="docker_interface_run",
+        limit=JobExecutionLimit.GROUP_ONCE,
+        on_condition=DockerJobError,
+    )
+    async def run(self) -> None:
         """Run Docker image."""
         raise NotImplementedError()
 
-    @process_lock
-    def stop(self, remove_container: bool = True) -> Awaitable[None]:
-        """Stop/remove Docker container."""
-        return self._stop(remove_container)
-
-    async def _stop(self, remove_container: bool = True) -> None:
+    @Job(
+        name="docker_interface_stop",
+        limit=JobExecutionLimit.GROUP_ONCE,
+        on_condition=DockerJobError,
+    )
+    async def stop(self, remove_container: bool = True) -> None:
         """Stop/remove Docker container."""
         with suppress(DockerNotFound):
             await self.sys_run_in_executor(
@@ -394,34 +398,40 @@ class DockerInterface(CoreSysAttributes):
                 remove_container,
             )
 
-    @process_lock
+    @Job(
+        name="docker_interface_start",
+        limit=JobExecutionLimit.GROUP_ONCE,
+        on_condition=DockerJobError,
+    )
     def start(self) -> Awaitable[None]:
         """Start Docker container."""
         return self.sys_run_in_executor(self.sys_docker.start_container, self.name)
 
-    @process_lock
+    @Job(
+        name="docker_interface_remove",
+        limit=JobExecutionLimit.GROUP_ONCE,
+        on_condition=DockerJobError,
+    )
     async def remove(self) -> None:
         """Remove Docker images."""
         # Cleanup container
         with suppress(DockerError):
-            await self._stop()
+            await self.stop()
 
         await self.sys_run_in_executor(
             self.sys_docker.remove_image, self.image, self.version
         )
         self._meta = None
 
-    @process_lock
-    def update(
-        self, version: AwesomeVersion, image: str | None = None, latest: bool = False
-    ) -> Awaitable[None]:
-        """Update a Docker image."""
-        return self._update(version, image, latest)
-
-    async def _update(
+    @Job(
+        name="docker_interface_update",
+        limit=JobExecutionLimit.GROUP_ONCE,
+        on_condition=DockerJobError,
+    )
+    async def update(
         self, version: AwesomeVersion, image: str | None = None, latest: bool = False
     ) -> None:
-        """Update a docker image."""
+        """Update a Docker image."""
         image = image or self.image
 
         _LOGGER.info(
@@ -429,11 +439,11 @@ class DockerInterface(CoreSysAttributes):
         )
 
         # Update docker image
-        await self._install(version, image=image, latest=latest)
+        await self.install(version, image=image, latest=latest)
 
         # Stop container & cleanup
         with suppress(DockerError):
-            await self._stop()
+            await self.stop()
 
     async def logs(self) -> bytes:
         """Return Docker logs of container."""
@@ -444,12 +454,12 @@ class DockerInterface(CoreSysAttributes):
 
         return b""
 
-    @process_lock
+    @Job(
+        name="docker_interface_cleanup",
+        limit=JobExecutionLimit.GROUP_ONCE,
+        on_condition=DockerJobError,
+    )
     def cleanup(self, old_image: str | None = None) -> Awaitable[None]:
-        """Check if old version exists and cleanup."""
-        return self._cleanup(old_image)
-
-    def _cleanup(self, old_image: str | None = None) -> Awaitable[None]:
         """Check if old version exists and cleanup."""
         return self.sys_run_in_executor(
             self.sys_docker.cleanup_old_images,
@@ -458,14 +468,22 @@ class DockerInterface(CoreSysAttributes):
             {old_image} if old_image else None,
         )
 
-    @process_lock
+    @Job(
+        name="docker_interface_restart",
+        limit=JobExecutionLimit.GROUP_ONCE,
+        on_condition=DockerJobError,
+    )
     def restart(self) -> Awaitable[None]:
         """Restart docker container."""
         return self.sys_run_in_executor(
             self.sys_docker.restart_container, self.name, self.timeout
         )
 
-    @process_lock
+    @Job(
+        name="docker_interface_execute_command",
+        limit=JobExecutionLimit.GROUP_ONCE,
+        on_condition=DockerJobError,
+    )
     async def execute_command(self, command: str) -> CommandReturn:
         """Create a temporary container and run command."""
         raise NotImplementedError()
@@ -526,7 +544,11 @@ class DockerInterface(CoreSysAttributes):
         available_version.sort(reverse=True)
         return available_version[0]
 
-    @process_lock
+    @Job(
+        name="docker_interface_run_inside",
+        limit=JobExecutionLimit.GROUP_ONCE,
+        on_condition=DockerJobError,
+    )
     def run_inside(self, command: str) -> Awaitable[CommandReturn]:
         """Execute a command inside Docker container."""
         return self.sys_run_in_executor(
@@ -540,7 +562,11 @@ class DockerInterface(CoreSysAttributes):
         checksum = image_id.partition(":")[2]
         return await self.sys_security.verify_own_content(checksum)
 
-    @process_lock
+    @Job(
+        name="docker_interface_check_trust",
+        limit=JobExecutionLimit.GROUP_ONCE,
+        on_condition=DockerJobError,
+    )
     async def check_trust(self) -> None:
         """Check trust of exists Docker image."""
         try:

supervisor/docker/multicast.py

@@ -2,6 +2,9 @@
 import logging
 
 from ..coresys import CoreSysAttributes
+from ..exceptions import DockerJobError
+from ..jobs.const import JobExecutionLimit
+from ..jobs.decorator import Job
 from .const import ENV_TIME, Capabilities
 from .interface import DockerInterface
 
@@ -28,13 +31,18 @@ class DockerMulticast(DockerInterface, CoreSysAttributes):
         """Generate needed capabilities."""
         return [Capabilities.NET_ADMIN.value]
 
-    async def _run(self) -> None:
+    @Job(
+        name="docker_multicast_run",
+        limit=JobExecutionLimit.GROUP_ONCE,
+        on_condition=DockerJobError,
+    )
+    async def run(self) -> None:
         """Run Docker image."""
         if await self.is_running():
             return
 
         # Cleanup
-        await self._stop()
+        await self.stop()
 
         # Create & Run container
         docker_container = await self.sys_run_in_executor(

supervisor/docker/observer.py

@@ -3,6 +3,9 @@ import logging
 
 from ..const import DOCKER_NETWORK_MASK
 from ..coresys import CoreSysAttributes
+from ..exceptions import DockerJobError
+from ..jobs.const import JobExecutionLimit
+from ..jobs.decorator import Job
 from .const import ENV_TIME, ENV_TOKEN, MOUNT_DOCKER, RestartPolicy
 from .interface import DockerInterface
 
@@ -25,13 +28,18 @@ class DockerObserver(DockerInterface, CoreSysAttributes):
         """Return name of Docker container."""
         return OBSERVER_DOCKER_NAME
 
-    async def _run(self) -> None:
+    @Job(
+        name="docker_observer_run",
+        limit=JobExecutionLimit.GROUP_ONCE,
+        on_condition=DockerJobError,
+    )
+    async def run(self) -> None:
         """Run Docker image."""
         if await self.is_running():
             return
 
         # Cleanup
-        await self._stop()
+        await self.stop()
 
         # Create & Run container
         docker_container = await self.sys_run_in_executor(

supervisor/docker/supervisor.py

@@ -9,7 +9,9 @@ import docker
 import requests
 
 from ..coresys import CoreSysAttributes
-from ..exceptions import DockerError
+from ..exceptions import DockerError, DockerJobError
+from ..jobs.const import JobExecutionLimit
+from ..jobs.decorator import Job
 from .const import PropagationMode
 from .interface import DockerInterface
 
@@ -43,8 +45,13 @@ class DockerSupervisor(DockerInterface, CoreSysAttributes):
             if mount.get("Destination") == "/data"
         )
 
-    async def _attach(
-        self, version: AwesomeVersion, skip_state_event_if_down: bool = False
+    @Job(
+        name="docker_supervisor_attach",
+        limit=JobExecutionLimit.GROUP_ONCE,
+        on_condition=DockerJobError,
+    )
+    async def attach(
+        self, version: AwesomeVersion, *, skip_state_event_if_down: bool = False
     ) -> None:
         """Attach to running docker container."""
         try:

supervisor/exceptions.py

@@ -36,6 +36,18 @@ class JobConditionException(JobException):
     """Exception happening for job conditions."""
 
 
+class JobStartException(JobException):
+    """Exception occurred starting a job on in current asyncio task."""
+
+
+class JobNotFound(JobException):
+    """Exception for job not found."""
+
+
+class JobGroupExecutionLimitExceeded(JobException):
+    """Exception when job group execution limit exceeded."""
+
+
 # HomeAssistant
 
 
@@ -478,6 +490,10 @@ class DockerNotFound(DockerError):
     """Docker object don't Exists."""
 
 
+class DockerJobError(DockerError, JobException):
+    """Error executing docker job."""
+
+
 # Hardware
 
 

supervisor/homeassistant/api.py

@@ -32,7 +32,10 @@ class HomeAssistantAPI(CoreSysAttributes):
         self.access_token: str | None = None
         self._access_token_expires: datetime | None = None
 
-    @Job(limit=JobExecutionLimit.SINGLE_WAIT)
+    @Job(
+        name="home_assistant_api_ensure_access_token",
+        limit=JobExecutionLimit.SINGLE_WAIT,
+    )
     async def ensure_access_token(self) -> None:
         """Ensure there is an access token."""
         if (

supervisor/homeassistant/core.py

@@ -11,7 +11,7 @@ import attr
 from awesomeversion import AwesomeVersion
 
 from ..const import ATTR_HOMEASSISTANT, BusEvent
-from ..coresys import CoreSys, CoreSysAttributes
+from ..coresys import CoreSys
 from ..docker.const import ContainerState
 from ..docker.homeassistant import DockerHomeAssistant
 from ..docker.monitor import DockerContainerStateEvent
@@ -22,11 +22,13 @@ from ..exceptions import (
     HomeAssistantError,
     HomeAssistantJobError,
     HomeAssistantUpdateError,
+    JobException,
 )
-from ..jobs.const import JobExecutionLimit
+from ..jobs.const import JOB_GROUP_HOME_ASSISTANT_CORE, JobExecutionLimit
 from ..jobs.decorator import Job, JobCondition
+from ..jobs.job_group import JobGroup
 from ..resolution.const import ContextType, IssueType
-from ..utils import convert_to_ascii, process_lock
+from ..utils import convert_to_ascii
 from ..utils.sentry import capture_exception
 from .const import (
     LANDINGPAGE,
@@ -49,12 +51,12 @@ class ConfigResult:
     log = attr.ib()
 
 
-class HomeAssistantCore(CoreSysAttributes):
+class HomeAssistantCore(JobGroup):
     """Home Assistant core object for handle it."""
 
     def __init__(self, coresys: CoreSys):
         """Initialize Home Assistant object."""
-        self.coresys: CoreSys = coresys
+        super().__init__(coresys, JOB_GROUP_HOME_ASSISTANT_CORE)
         self.instance: DockerHomeAssistant = DockerHomeAssistant(coresys)
         self.lock: asyncio.Lock = asyncio.Lock()
         self._error_state: bool = False
@@ -95,9 +97,13 @@ class HomeAssistantCore(CoreSysAttributes):
         _LOGGER.info("Starting HomeAssistant landingpage")
         if not await self.instance.is_running():
             with suppress(HomeAssistantError):
-                await self._start()
+                await self.start()
 
-    @process_lock
+    @Job(
+        name="home_assistant_core_install_landing_page",
+        limit=JobExecutionLimit.GROUP_ONCE,
+        on_condition=HomeAssistantJobError,
+    )
     async def install_landingpage(self) -> None:
         """Install a landing page."""
         # Try to use a preinstalled landingpage
@@ -127,7 +133,7 @@ class HomeAssistantCore(CoreSysAttributes):
                     LANDINGPAGE, image=self.sys_updater.image_homeassistant
                 )
                 break
-            except DockerError:
+            except (DockerError, JobException):
                 pass
             except Exception as err:  # pylint: disable=broad-except
                 capture_exception(err)
@@ -139,7 +145,11 @@ class HomeAssistantCore(CoreSysAttributes):
         self.sys_homeassistant.image = self.sys_updater.image_homeassistant
         self.sys_homeassistant.save_data()
 
-    @process_lock
+    @Job(
+        name="home_assistant_core_install",
+        limit=JobExecutionLimit.GROUP_ONCE,
+        on_condition=HomeAssistantJobError,
+    )
     async def install(self) -> None:
         """Install a landing page."""
         _LOGGER.info("Home Assistant setup")
@@ -155,7 +165,7 @@ class HomeAssistantCore(CoreSysAttributes):
                         image=self.sys_updater.image_homeassistant,
                     )
                     break
-                except DockerError:
+                except (DockerError, JobException):
                     pass
                 except Exception as err:  # pylint: disable=broad-except
                     capture_exception(err)
@@ -171,7 +181,7 @@ class HomeAssistantCore(CoreSysAttributes):
         # finishing
         try:
             _LOGGER.info("Starting Home Assistant")
-            await self._start()
+            await self.start()
         except HomeAssistantError:
             _LOGGER.error("Can't start Home Assistant!")
 
@@ -179,8 +189,8 @@ class HomeAssistantCore(CoreSysAttributes):
         with suppress(DockerError):
             await self.instance.cleanup()
 
-    @process_lock
     @Job(
+        name="home_assistant_core_update",
         conditions=[
             JobCondition.FREE_SPACE,
             JobCondition.HEALTHY,
@@ -188,6 +198,7 @@ class HomeAssistantCore(CoreSysAttributes):
             JobCondition.PLUGINS_UPDATED,
             JobCondition.SUPERVISOR_UPDATED,
         ],
+        limit=JobExecutionLimit.GROUP_ONCE,
         on_condition=HomeAssistantJobError,
     )
     async def update(
@@ -231,7 +242,7 @@ class HomeAssistantCore(CoreSysAttributes):
             self.sys_homeassistant.image = self.sys_updater.image_homeassistant
 
             if running:
-                await self._start()
+                await self.start()
             _LOGGER.info("Successfully started Home Assistant %s", to_version)
 
             # Successfull - last step
@@ -281,23 +292,11 @@ class HomeAssistantCore(CoreSysAttributes):
             self.sys_resolution.create_issue(IssueType.UPDATE_FAILED, ContextType.CORE)
             raise HomeAssistantUpdateError()
 
-    async def _start(self) -> None:
-        """Start Home Assistant Docker & wait."""
-        # Create new API token
-        self.sys_homeassistant.supervisor_token = secrets.token_hex(56)
-        self.sys_homeassistant.save_data()
-
-        # Write audio settings
-        self.sys_homeassistant.write_pulse()
-
-        try:
-            await self.instance.run()
-        except DockerError as err:
-            raise HomeAssistantError() from err
-
-        await self._block_till_run(self.sys_homeassistant.version)
-
-    @process_lock
+    @Job(
+        name="home_assistant_core_start",
+        limit=JobExecutionLimit.GROUP_ONCE,
+        on_condition=HomeAssistantJobError,
+    )
     async def start(self) -> None:
         """Run Home Assistant docker."""
         if await self.instance.is_running():
@@ -314,9 +313,25 @@ class HomeAssistantCore(CoreSysAttributes):
             await self._block_till_run(self.sys_homeassistant.version)
         # No Instance/Container found, extended start
         else:
-            await self._start()
+            # Create new API token
+            self.sys_homeassistant.supervisor_token = secrets.token_hex(56)
+            self.sys_homeassistant.save_data()
 
-    @process_lock
+            # Write audio settings
+            self.sys_homeassistant.write_pulse()
+
+            try:
+                await self.instance.run()
+            except DockerError as err:
+                raise HomeAssistantError() from err
+
+            await self._block_till_run(self.sys_homeassistant.version)
+
+    @Job(
+        name="home_assistant_core_stop",
+        limit=JobExecutionLimit.GROUP_ONCE,
+        on_condition=HomeAssistantJobError,
+    )
     async def stop(self) -> None:
         """Stop Home Assistant Docker."""
         try:
@@ -324,7 +339,11 @@ class HomeAssistantCore(CoreSysAttributes):
         except DockerError as err:
             raise HomeAssistantError() from err
 
-    @process_lock
+    @Job(
+        name="home_assistant_core_restart",
+        limit=JobExecutionLimit.GROUP_ONCE,
+        on_condition=HomeAssistantJobError,
+    )
     async def restart(self) -> None:
         """Restart Home Assistant Docker."""
         try:
@@ -334,12 +353,16 @@ class HomeAssistantCore(CoreSysAttributes):
 
         await self._block_till_run(self.sys_homeassistant.version)
 
-    @process_lock
+    @Job(
+        name="home_assistant_core_rebuild",
+        limit=JobExecutionLimit.GROUP_ONCE,
+        on_condition=HomeAssistantJobError,
+    )
     async def rebuild(self) -> None:
         """Rebuild Home Assistant Docker container."""
         with suppress(DockerError):
             await self.instance.stop()
-        await self._start()
+        await self.start()
 
     def logs(self) -> Awaitable[bytes]:
         """Get HomeAssistant docker logs.
@@ -356,10 +379,7 @@ class HomeAssistantCore(CoreSysAttributes):
         return self.instance.check_trust()
 
     async def stats(self) -> DockerStats:
-        """Return stats of Home Assistant.
-
-        Return a coroutine.
-        """
+        """Return stats of Home Assistant."""
         try:
             return await self.instance.stats()
         except DockerError as err:
@@ -386,9 +406,12 @@ class HomeAssistantCore(CoreSysAttributes):
 
     async def check_config(self) -> ConfigResult:
         """Run Home Assistant config check."""
-        result = await self.instance.execute_command(
-            "python3 -m homeassistant -c /config --script check_config"
-        )
+        try:
+            result = await self.instance.execute_command(
+                "python3 -m homeassistant -c /config --script check_config"
+            )
+        except DockerError as err:
+            raise HomeAssistantError() from err
 
         # If not valid
         if result.exit_code is None:
@@ -431,10 +454,11 @@ class HomeAssistantCore(CoreSysAttributes):
         raise HomeAssistantCrashError()
 
     @Job(
+        name="home_assistant_core_repair",
         conditions=[
             JobCondition.FREE_SPACE,
             JobCondition.INTERNET_HOST,
-        ]
+        ],
     )
     async def repair(self):
         """Repair local Home Assistant data."""
@@ -456,6 +480,7 @@ class HomeAssistantCore(CoreSysAttributes):
             await self._restart_after_problem(event.state)
 
     @Job(
+        name="home_assistant_core_restart_after_problem",
         limit=JobExecutionLimit.THROTTLE_RATE_LIMIT,
         throttle_period=WATCHDOG_THROTTLE_PERIOD,
         throttle_max_calls=WATCHDOG_THROTTLE_MAX_CALLS,

supervisor/homeassistant/module.py

@@ -1,5 +1,6 @@
 """Home Assistant control object."""
 import asyncio
+from datetime import timedelta
 from ipaddress import IPv4Address
 import logging
 from pathlib import Path, PurePath
@@ -28,6 +29,7 @@ from ..const import (
     ATTR_WATCHDOG,
     FILE_HASSIO_HOMEASSISTANT,
     BusEvent,
+    IngressSessionDataUser,
 )
 from ..coresys import CoreSys, CoreSysAttributes
 from ..exceptions import (
@@ -38,7 +40,7 @@ from ..exceptions import (
 )
 from ..hardware.const import PolicyGroup
 from ..hardware.data import Device
-from ..jobs.decorator import Job
+from ..jobs.decorator import Job, JobExecutionLimit
 from ..utils import remove_folder
 from ..utils.common import FileConfiguration
 from ..utils.json import read_json_file, write_json_file
@@ -304,7 +306,7 @@ class HomeAssistant(FileConfiguration, CoreSysAttributes):
 
         self.sys_homeassistant.websocket.send_message({ATTR_TYPE: "usb/scan"})
 
-    @Job()
+    @Job(name="home_assistant_module_backup")
     async def backup(self, tar_file: tarfile.TarFile) -> None:
         """Backup Home Assistant Core config/ directory."""
 
@@ -432,3 +434,21 @@ class HomeAssistant(FileConfiguration, CoreSysAttributes):
                 ATTR_WATCHDOG,
             ):
                 self._data[attr] = data[attr]
+
+    @Job(
+        name="home_assistant_get_users",
+        limit=JobExecutionLimit.THROTTLE_WAIT,
+        throttle_period=timedelta(minutes=5),
+    )
+    async def get_users(self) -> list[IngressSessionDataUser]:
+        """Get list of all configured users."""
+        list_of_users = await self.sys_homeassistant.websocket.async_send_command(
+            {ATTR_TYPE: "config/auth/list"}
+        )
+
+        return [
+            IngressSessionDataUser(
+                id=data["id"], username=data["username"], display_name=data["name"]
+            )
+            for data in list_of_users
+        ]

supervisor/homeassistant/secrets.py

@@ -40,7 +40,11 @@ class HomeAssistantSecrets(CoreSysAttributes):
         """Reload secrets."""
         await self._read_secrets()
 
-    @Job(limit=JobExecutionLimit.THROTTLE_WAIT, throttle_period=timedelta(seconds=60))
+    @Job(
+        name="home_assistant_secrets_read",
+        limit=JobExecutionLimit.THROTTLE_WAIT,
+        throttle_period=timedelta(seconds=60),
+    )
     async def _read_secrets(self):
         """Read secrets.yaml into memory."""
         if not self.path_secrets.exists():

supervisor/host/configuration.py

@@ -61,6 +61,7 @@ class Interface:
 
     name: str
     mac: str
+    path: str
     enabled: bool
     connected: bool
     primary: bool
@@ -75,12 +76,8 @@ class Interface:
         if not inet.settings:
             return False
 
-        if inet.settings.device and inet.settings.device.match_device:
-            matchers = inet.settings.device.match_device.split(",", 1)
-            return (
-                f"mac:{self.mac}" in matchers
-                or f"interface-name:{self.name}" in matchers
-            )
+        if inet.settings.match and inet.settings.match.path:
+            return inet.settings.match.path == [self.path]
 
         return inet.settings.connection.interface_name == self.name
 
@@ -108,6 +105,7 @@ class Interface:
         return Interface(
             inet.name,
             inet.hw_address,
+            inet.path,
             inet.settings is not None,
             Interface._map_nm_connected(inet.connection),
             inet.primary,

supervisor/host/network.py

@@ -107,7 +107,7 @@ class NetworkManager(CoreSysAttributes):
 
         return Interface.from_dbus_interface(self.sys_dbus.network.get(inet_name))
 
-    @Job(conditions=[JobCondition.HOST_NETWORK])
+    @Job(name="network_manager_load", conditions=[JobCondition.HOST_NETWORK])
     async def load(self):
         """Load network information and reapply defaults over dbus."""
         # Apply current settings on each interface so OS can update any out of date defaults

supervisor/host/sound.py

@@ -232,7 +232,11 @@ class SoundControl(CoreSysAttributes):
         await self.sys_run_in_executor(_activate_profile)
         await self.update()
 
-    @Job(limit=JobExecutionLimit.THROTTLE_WAIT, throttle_period=timedelta(seconds=10))
+    @Job(
+        name="sound_control_update",
+        limit=JobExecutionLimit.THROTTLE_WAIT,
+        throttle_period=timedelta(seconds=10),
+    )
     async def update(self):
         """Update properties over dbus."""
         _LOGGER.info("Updating PulseAudio information")

supervisor/ingress.py

@@ -5,7 +5,13 @@ import random
 import secrets
 
 from .addons.addon import Addon
-from .const import ATTR_PORTS, ATTR_SESSION, FILE_HASSIO_INGRESS
+from .const import (
+    ATTR_PORTS,
+    ATTR_SESSION,
+    ATTR_SESSION_DATA,
+    FILE_HASSIO_INGRESS,
+    IngressSessionData,
+)
 from .coresys import CoreSys, CoreSysAttributes
 from .utils import check_port
 from .utils.common import FileConfiguration
@@ -30,11 +36,20 @@ class Ingress(FileConfiguration, CoreSysAttributes):
             return None
         return self.sys_addons.get(self.tokens[token], local_only=True)
 
+    def get_session_data(self, session_id: str) -> IngressSessionData | None:
+        """Return complementary data of current session or None."""
+        return self.sessions_data.get(session_id)
+
     @property
     def sessions(self) -> dict[str, float]:
         """Return sessions."""
         return self._data[ATTR_SESSION]
 
+    @property
+    def sessions_data(self) -> dict[str, IngressSessionData]:
+        """Return sessions_data."""
+        return self._data[ATTR_SESSION_DATA]
+
     @property
     def ports(self) -> dict[str, int]:
         """Return list of dynamic ports."""
@@ -71,6 +86,7 @@ class Ingress(FileConfiguration, CoreSysAttributes):
         now = utcnow()
 
         sessions = {}
+        sessions_data: dict[str, IngressSessionData] = {}
         for session, valid in self.sessions.items():
             # check if timestamp valid, to avoid crash on malformed timestamp
             try:
@@ -84,10 +100,13 @@ class Ingress(FileConfiguration, CoreSysAttributes):
 
             # Is valid
             sessions[session] = valid
+            sessions_data[session] = self.get_session_data(session)
 
         # Write back
         self.sessions.clear()
         self.sessions.update(sessions)
+        self.sessions_data.clear()
+        self.sessions_data.update(sessions_data)
 
     def _update_token_list(self) -> None:
         """Regenerate token <-> Add-on map."""
@@ -97,12 +116,15 @@ class Ingress(FileConfiguration, CoreSysAttributes):
         for addon in self.addons:
             self.tokens[addon.ingress_token] = addon.slug
 
-    def create_session(self) -> str:
+    def create_session(self, data: IngressSessionData | None = None) -> str:
         """Create new session."""
         session = secrets.token_hex(64)
         valid = utcnow() + timedelta(minutes=15)
 
         self.sessions[session] = valid.timestamp()
+        if data is not None:
+            self.sessions_data[session] = data
+
         return session
 
     def validate_session(self, session: str) -> bool:

supervisor/jobs/__init__.py

@@ -1,53 +1,70 @@
 """Supervisor job manager."""
+from collections.abc import Callable
+from contextlib import contextmanager
+from contextvars import ContextVar, Token
 import logging
+from uuid import UUID, uuid4
+
+from attrs import define, field
+from attrs.setters import frozen
+from attrs.validators import ge, le
 
 from ..coresys import CoreSys, CoreSysAttributes
+from ..exceptions import JobNotFound, JobStartException
 from ..utils.common import FileConfiguration
 from .const import ATTR_IGNORE_CONDITIONS, FILE_CONFIG_JOBS, JobCondition
 from .validate import SCHEMA_JOBS_CONFIG
 
-_LOGGER: logging.Logger = logging.getLogger(__package__)
+# Context vars only act as a global within the same asyncio task
+# When a new asyncio task is started the current context is copied over.
+# Modifications to it in one task are not visible to others though.
+# This allows us to track what job is currently in progress in each task.
+_CURRENT_JOB: ContextVar[UUID] = ContextVar("current_job")
+
+_LOGGER: logging.Logger = logging.getLogger(__name__)
 
 
-class SupervisorJob(CoreSysAttributes):
-    """Supervisor running job class."""
+@define
+class SupervisorJob:
+    """Representation of a job running in supervisor."""
 
-    def __init__(self, coresys: CoreSys, name: str):
-        """Initialize the JobManager class."""
-        self.coresys: CoreSys = coresys
-        self.name: str = name
-        self._progress: int = 0
-        self._stage: str | None = None
+    name: str = field(on_setattr=frozen)
+    reference: str | None = None
+    progress: int = field(default=0, validator=[ge(0), le(100)])
+    stage: str | None = None
+    uuid: UUID = field(init=False, factory=lambda: uuid4().hex, on_setattr=frozen)
+    parent_id: UUID = field(
+        init=False, factory=lambda: _CURRENT_JOB.get(None), on_setattr=frozen
+    )
+    done: bool = field(init=False, default=False)
 
-    @property
-    def progress(self) -> int:
-        """Return the current progress."""
-        return self._progress
+    @contextmanager
+    def start(self, *, on_done: Callable[["SupervisorJob"], None] | None = None):
+        """Start the job in the current task.
 
-    @property
-    def stage(self) -> str | None:
-        """Return the current stage."""
-        return self._stage
+        This can only be called if the parent ID matches the job running in the current task.
+        This is to ensure that each asyncio task can only be doing one job at a time as that
+        determines what resources it can and cannot access.
+        """
+        if self.done:
+            raise JobStartException("Job is already complete")
+        if _CURRENT_JOB.get(None) != self.parent_id:
+            raise JobStartException("Job has a different parent from current job")
 
-    def update(self, progress: int | None = None, stage: str | None = None) -> None:
-        """Update the job object."""
-        if progress is not None:
-            if progress >= round(100):
-                self.sys_jobs.remove_job(self)
-                return
-            self._progress = round(progress)
-        if stage is not None:
-            self._stage = stage
-        _LOGGER.debug(
-            "Job updated; name: %s, progress: %s, stage: %s",
-            self.name,
-            self.progress,
-            self.stage,
-        )
+        token: Token[UUID] | None = None
+        try:
+            token = _CURRENT_JOB.set(self.uuid)
+            yield self
+        finally:
+            self.done = True
+            if token:
+                _CURRENT_JOB.reset(token)
+            if on_done:
+                on_done(self)
 
 
 class JobManager(FileConfiguration, CoreSysAttributes):
-    """Job class."""
+    """Job Manager class."""
 
     def __init__(self, coresys: CoreSys):
         """Initialize the JobManager class."""
@@ -58,7 +75,7 @@ class JobManager(FileConfiguration, CoreSysAttributes):
     @property
     def jobs(self) -> list[SupervisorJob]:
         """Return a list of current jobs."""
-        return self._jobs
+        return list(self._jobs.values())
 
     @property
     def ignore_conditions(self) -> list[JobCondition]:
@@ -70,18 +87,30 @@ class JobManager(FileConfiguration, CoreSysAttributes):
         """Set a list of ignored condition."""
         self._data[ATTR_IGNORE_CONDITIONS] = value
 
-    def get_job(self, name: str) -> SupervisorJob:
-        """Return a job, create one if it does not exist."""
-        if name not in self._jobs:
-            self._jobs[name] = SupervisorJob(self.coresys, name)
+    def new_job(
+        self, name: str, reference: str | None = None, initial_stage: str | None = None
+    ) -> SupervisorJob:
+        """Create a new job."""
+        job = SupervisorJob(name, reference=reference, stage=initial_stage)
+        self._jobs[job.uuid] = job
+        return job
 
-        return self._jobs[name]
+    def get_job(self, uuid: UUID | None = None) -> SupervisorJob | None:
+        """Return a job by uuid if it exists. Returns the current job of the asyncio task if uuid omitted."""
+        if uuid:
+            return self._jobs.get(uuid)
+
+        if uuid := _CURRENT_JOB.get(None):
+            return self._jobs.get(uuid)
+
+        return None
 
     def remove_job(self, job: SupervisorJob) -> None:
-        """Remove a job."""
-        if job.name in self._jobs:
-            del self._jobs[job.name]
+        """Remove a job by UUID."""
+        if job.uuid not in self._jobs:
+            raise JobNotFound(f"Could not find job {job.name}", _LOGGER.error)
 
-    def clear(self) -> None:
-        """Clear all jobs."""
-        self._jobs.clear()
+        if not job.done:
+            _LOGGER.warning("Removing incomplete job %s from job manager", job.name)
+
+        del self._jobs[job.uuid]

supervisor/jobs/const.py

@@ -8,6 +8,10 @@ FILE_CONFIG_JOBS = Path(SUPERVISOR_DATA, "jobs.json")
 
 ATTR_IGNORE_CONDITIONS = "ignore_conditions"
 
+JOB_GROUP_ADDON = "addon_{slug}"
+JOB_GROUP_DOCKER_INTERFACE = "container_{name}"
+JOB_GROUP_HOME_ASSISTANT_CORE = "home_assistant_core"
+
 
 class JobCondition(str, Enum):
     """Job condition enum."""
@@ -34,3 +38,8 @@ class JobExecutionLimit(str, Enum):
     THROTTLE = "throttle"
     THROTTLE_WAIT = "throttle_wait"
     THROTTLE_RATE_LIMIT = "throttle_rate_limit"
+    GROUP_ONCE = "group_once"
+    GROUP_WAIT = "group_wait"
+    GROUP_THROTTLE = "group_throttle"
+    GROUP_THROTTLE_WAIT = "group_throttle_wait"
+    GROUP_THROTTLE_RATE_LIMIT = "group_throttle_rate_limit"

supervisor/jobs/decorator.py

@@ -8,13 +8,20 @@ from typing import Any
 
 from ..const import CoreState
 from ..coresys import CoreSys, CoreSysAttributes
-from ..exceptions import HassioError, JobConditionException, JobException
+from ..exceptions import (
+    HassioError,
+    JobConditionException,
+    JobException,
+    JobGroupExecutionLimitExceeded,
+)
 from ..host.const import HostFeature
 from ..resolution.const import MINIMUM_FREE_SPACE_THRESHOLD, ContextType, IssueType
 from ..utils.sentry import capture_exception
 from .const import JobCondition, JobExecutionLimit
+from .job_group import JobGroup
 
 _LOGGER: logging.Logger = logging.getLogger(__package__)
+_JOB_NAMES: set[str] = set()
 
 
 class Job(CoreSysAttributes):
@@ -22,7 +29,7 @@ class Job(CoreSysAttributes):
 
     def __init__(
         self,
-        name: str | None = None,
+        name: str,
         conditions: list[JobCondition] | None = None,
         cleanup: bool = True,
         on_condition: JobException | None = None,
@@ -33,6 +40,10 @@ class Job(CoreSysAttributes):
         throttle_max_calls: int | None = None,
     ):
         """Initialize the Job class."""
+        if name in _JOB_NAMES:
+            raise RuntimeError(f"A job already exists with name {name}!")
+
+        _JOB_NAMES.add(name)
         self.name = name
         self.conditions = conditions
         self.cleanup = cleanup
@@ -42,8 +53,8 @@ class Job(CoreSysAttributes):
         self.throttle_max_calls = throttle_max_calls
         self._lock: asyncio.Semaphore | None = None
         self._method = None
-        self._last_call = datetime.min
-        self._rate_limited_calls: list[datetime] | None = None
+        self._last_call: dict[str | None, datetime] = {}
+        self._rate_limited_calls: dict[str, list[datetime]] | None = None
 
         # Validate Options
         if (
@@ -52,19 +63,70 @@ class Job(CoreSysAttributes):
                 JobExecutionLimit.THROTTLE,
                 JobExecutionLimit.THROTTLE_WAIT,
                 JobExecutionLimit.THROTTLE_RATE_LIMIT,
+                JobExecutionLimit.GROUP_THROTTLE,
+                JobExecutionLimit.GROUP_THROTTLE_WAIT,
+                JobExecutionLimit.GROUP_THROTTLE_RATE_LIMIT,
             )
             and self._throttle_period is None
         ):
-            raise RuntimeError("Using Job without a Throttle period!")
+            raise RuntimeError(
+                f"Job {name} is using execution limit {limit.value} without a throttle period!"
+            )
 
-        if self.limit == JobExecutionLimit.THROTTLE_RATE_LIMIT:
+        if self.limit in (
+            JobExecutionLimit.THROTTLE_RATE_LIMIT,
+            JobExecutionLimit.GROUP_THROTTLE_RATE_LIMIT,
+        ):
             if self.throttle_max_calls is None:
-                raise RuntimeError("Using rate limit without throttle max calls!")
+                raise RuntimeError(
+                    f"Job {name} is using execution limit {limit.value} without throttle max calls!"
+                )
 
-            self._rate_limited_calls = []
+            self._rate_limited_calls = {}
 
-    @property
-    def throttle_period(self) -> timedelta | None:
+    def last_call(self, group_name: str | None = None) -> datetime:
+        """Return last call datetime."""
+        return self._last_call.get(group_name, datetime.min)
+
+    def set_last_call(self, value: datetime, group_name: str | None = None) -> None:
+        """Set last call datetime."""
+        self._last_call[group_name] = value
+
+    def rate_limited_calls(
+        self, group_name: str | None = None
+    ) -> list[datetime] | None:
+        """Return rate limited calls if used."""
+        if self._rate_limited_calls is None:
+            return None
+
+        return self._rate_limited_calls.get(group_name, [])
+
+    def add_rate_limited_call(
+        self, value: datetime, group_name: str | None = None
+    ) -> None:
+        """Add a rate limited call to list if used."""
+        if self._rate_limited_calls is None:
+            raise RuntimeError(
+                f"Rate limited calls not available for limit type {self.limit}"
+            )
+
+        if group_name in self._rate_limited_calls:
+            self._rate_limited_calls[group_name].append(value)
+        else:
+            self._rate_limited_calls[group_name] = [value]
+
+    def set_rate_limited_calls(
+        self, value: list[datetime], group_name: str | None = None
+    ) -> None:
+        """Set rate limited calls if used."""
+        if self._rate_limited_calls is None:
+            raise RuntimeError(
+                f"Rate limited calls not available for limit type {self.limit}"
+            )
+
+        self._rate_limited_calls[group_name] = value
+
+    def throttle_period(self, group_name: str | None = None) -> timedelta | None:
         """Return throttle period."""
         if self._throttle_period is None:
             return None
@@ -73,36 +135,59 @@ class Job(CoreSysAttributes):
             return self._throttle_period
 
         return self._throttle_period(
-            self.coresys, self._last_call, self._rate_limited_calls
+            self.coresys,
+            self.last_call(group_name),
+            self.rate_limited_calls(group_name),
         )
 
-    def _post_init(self, args: tuple[Any]) -> None:
+    def _post_init(self, obj: JobGroup | CoreSysAttributes) -> JobGroup | None:
         """Runtime init."""
-        if self.name is None:
-            self.name = str(self._method.__qualname__).lower().replace(".", "_")
-
         # Coresys
         try:
-            self.coresys = args[0].coresys
+            self.coresys = obj.coresys
         except AttributeError:
             pass
         if not self.coresys:
             raise RuntimeError(f"Job on {self.name} need to be an coresys object!")
 
-        # Others
+        # Setup lock for limits
         if self._lock is None:
             self._lock = asyncio.Semaphore()
 
+        # Job groups
+        if self.limit in (
+            JobExecutionLimit.GROUP_ONCE,
+            JobExecutionLimit.GROUP_WAIT,
+            JobExecutionLimit.GROUP_THROTTLE,
+            JobExecutionLimit.GROUP_THROTTLE_WAIT,
+            JobExecutionLimit.GROUP_THROTTLE_RATE_LIMIT,
+        ):
+            try:
+                _ = obj.acquire and obj.release
+            except AttributeError:
+                raise RuntimeError(
+                    f"Job on {self.name} need to be a JobGroup to use group based limits!"
+                ) from None
+
+            return obj
+        return None
+
     def __call__(self, method):
         """Call the wrapper logic."""
         self._method = method
 
         @wraps(method)
-        async def wrapper(*args, **kwargs) -> Any:
-            """Wrap the method."""
-            self._post_init(args)
+        async def wrapper(obj: JobGroup | CoreSysAttributes, *args, **kwargs) -> Any:
+            """Wrap the method.
 
-            job = self.sys_jobs.get_job(self.name)
+            This method must be on an instance of CoreSysAttributes. If a JOB_GROUP limit
+            is used, then it must be on an instance of JobGroup.
+            """
+            job_group = self._post_init(obj)
+            group_name: str | None = job_group.group_name if job_group else None
+            job = self.sys_jobs.new_job(
+                self.name, job_group.job_reference if job_group else None
+            )
 
             # Handle condition
             if self.conditions:
@@ -118,50 +203,78 @@ class Job(CoreSysAttributes):
             # Handle exection limits
             if self.limit in (JobExecutionLimit.SINGLE_WAIT, JobExecutionLimit.ONCE):
                 await self._acquire_exection_limit()
-            elif self.limit == JobExecutionLimit.THROTTLE:
-                time_since_last_call = datetime.now() - self._last_call
-                if time_since_last_call < self.throttle_period:
+            elif self.limit in (
+                JobExecutionLimit.GROUP_ONCE,
+                JobExecutionLimit.GROUP_WAIT,
+            ):
+                try:
+                    await obj.acquire(job, self.limit == JobExecutionLimit.GROUP_WAIT)
+                except JobGroupExecutionLimitExceeded as err:
+                    if self.on_condition:
+                        raise self.on_condition(str(err)) from err
+                    raise err
+            elif self.limit in (
+                JobExecutionLimit.THROTTLE,
+                JobExecutionLimit.GROUP_THROTTLE,
+            ):
+                time_since_last_call = datetime.now() - self.last_call(group_name)
+                if time_since_last_call < self.throttle_period(group_name):
                     return
-            elif self.limit == JobExecutionLimit.THROTTLE_WAIT:
+            elif self.limit in (
+                JobExecutionLimit.THROTTLE_WAIT,
+                JobExecutionLimit.GROUP_THROTTLE_WAIT,
+            ):
                 await self._acquire_exection_limit()
-                time_since_last_call = datetime.now() - self._last_call
-                if time_since_last_call < self.throttle_period:
+                time_since_last_call = datetime.now() - self.last_call(group_name)
+                if time_since_last_call < self.throttle_period(group_name):
                     self._release_exception_limits()
                     return
-            elif self.limit == JobExecutionLimit.THROTTLE_RATE_LIMIT:
+            elif self.limit in (
+                JobExecutionLimit.THROTTLE_RATE_LIMIT,
+                JobExecutionLimit.GROUP_THROTTLE_RATE_LIMIT,
+            ):
                 # Only reprocess array when necessary (at limit)
-                if len(self._rate_limited_calls) >= self.throttle_max_calls:
-                    self._rate_limited_calls = [
-                        call
-                        for call in self._rate_limited_calls
-                        if call > datetime.now() - self.throttle_period
-                    ]
+                if len(self.rate_limited_calls(group_name)) >= self.throttle_max_calls:
+                    self.set_rate_limited_calls(
+                        [
+                            call
+                            for call in self.rate_limited_calls(group_name)
+                            if call > datetime.now() - self.throttle_period(group_name)
+                        ],
+                        group_name,
+                    )
 
-                if len(self._rate_limited_calls) >= self.throttle_max_calls:
+                if len(self.rate_limited_calls(group_name)) >= self.throttle_max_calls:
                     on_condition = (
                         JobException if self.on_condition is None else self.on_condition
                     )
                     raise on_condition(
-                        f"Rate limit exceeded, more then {self.throttle_max_calls} calls in {self.throttle_period}",
+                        f"Rate limit exceeded, more then {self.throttle_max_calls} calls in {self.throttle_period(group_name)}",
                     )
 
             # Execute Job
-            try:
-                self._last_call = datetime.now()
-                if self._rate_limited_calls is not None:
-                    self._rate_limited_calls.append(self._last_call)
+            with job.start(on_done=self.sys_jobs.remove_job if self.cleanup else None):
+                try:
+                    self.set_last_call(datetime.now(), group_name)
+                    if self.rate_limited_calls(group_name) is not None:
+                        self.add_rate_limited_call(
+                            self.last_call(group_name), group_name
+                        )
 
-                return await self._method(*args, **kwargs)
-            except HassioError as err:
-                raise err
-            except Exception as err:
-                _LOGGER.exception("Unhandled exception: %s", err)
-                capture_exception(err)
-                raise JobException() from err
-            finally:
-                if self.cleanup:
-                    self.sys_jobs.remove_job(job)
-                self._release_exception_limits()
+                    return await self._method(obj, *args, **kwargs)
+                except HassioError as err:
+                    raise err
+                except Exception as err:
+                    _LOGGER.exception("Unhandled exception: %s", err)
+                    capture_exception(err)
+                    raise JobException() from err
+                finally:
+                    self._release_exception_limits()
+                    if self.limit in (
+                        JobExecutionLimit.GROUP_ONCE,
+                        JobExecutionLimit.GROUP_WAIT,
+                    ):
+                        obj.release()
 
         return wrapper
 
@@ -283,6 +396,7 @@ class Job(CoreSysAttributes):
             JobExecutionLimit.SINGLE_WAIT,
             JobExecutionLimit.ONCE,
             JobExecutionLimit.THROTTLE_WAIT,
+            JobExecutionLimit.GROUP_THROTTLE_WAIT,
         ):
             return
 
@@ -300,6 +414,7 @@ class Job(CoreSysAttributes):
             JobExecutionLimit.SINGLE_WAIT,
             JobExecutionLimit.ONCE,
             JobExecutionLimit.THROTTLE_WAIT,
+            JobExecutionLimit.GROUP_THROTTLE_WAIT,
         ):
             return
         self._lock.release()

supervisor/jobs/job_group.py

@@ -0,0 +1,81 @@
+"""Job group object."""
+
+from asyncio import Lock
+
+from . import SupervisorJob
+from ..coresys import CoreSys, CoreSysAttributes
+from ..exceptions import JobException, JobGroupExecutionLimitExceeded
+
+
+class JobGroup(CoreSysAttributes):
+    """Object with methods that require a common lock.
+
+    This is used in classes like our DockerInterface class. Where each method
+    requires a lock as it involves some extensive I/O with Docker. But some
+    methods may need to call others as a part of processing to complete a
+    higher-level task and should not need to relinquish the lock in between.
+    """
+
+    def __init__(
+        self, coresys: CoreSys, group_name: str, job_reference: str | None = None
+    ) -> None:
+        """Initialize object."""
+        self.coresys: CoreSys = coresys
+        self._group_name: str = group_name
+        self._lock: Lock = Lock()
+        self._active_job: SupervisorJob | None = None
+        self._parent_jobs: list[SupervisorJob] = []
+        self._job_reference: str | None = job_reference
+
+    @property
+    def active_job(self) -> SupervisorJob | None:
+        """Get active job ID."""
+        return self._active_job
+
+    @property
+    def group_name(self) -> str:
+        """Return group name."""
+        return self._group_name
+
+    @property
+    def has_lock(self) -> bool:
+        """Return true if current task has the lock on this job group."""
+        return (
+            self.active_job
+            and (task_job := self.sys_jobs.get_job())
+            and self.active_job == task_job
+        )
+
+    @property
+    def job_reference(self) -> str | None:
+        """Return value to use as reference for all jobs created for this job group."""
+        return self._job_reference
+
+    async def acquire(self, job: SupervisorJob, wait: bool = False) -> None:
+        """Acquire the lock for the group for the specified job."""
+        # If there's another job running and we're not waiting, raise
+        if self.active_job and not self.has_lock and not wait:
+            raise JobGroupExecutionLimitExceeded(
+                f"Another job is running for job group {self.group_name}"
+            )
+
+        # Else if we don't have the lock, acquire it
+        if not self.has_lock:
+            await self._lock.acquire()
+
+        # Store the job ID we acquired the lock for
+        if self.active_job:
+            self._parent_jobs.append(self.active_job)
+
+        self._active_job = job
+
+    def release(self) -> None:
+        """Release the lock for the group or return it to parent."""
+        if not self.has_lock:
+            raise JobException("Cannot release as caller does not own lock")
+
+        if self._parent_jobs:
+            self._active_job = self._parent_jobs.pop()
+        else:
+            self._active_job = None
+            self._lock.release()

supervisor/misc/tasks.py

@@ -28,15 +28,13 @@ RUN_RELOAD_BACKUPS = 72000
 RUN_RELOAD_HOST = 7600
 RUN_RELOAD_UPDATER = 7200
 RUN_RELOAD_INGRESS = 930
+RUN_RELOAD_MOUNTS = 900
 
 RUN_WATCHDOG_HOMEASSISTANT_API = 120
 
 RUN_WATCHDOG_ADDON_APPLICATON = 120
 RUN_WATCHDOG_OBSERVER_APPLICATION = 180
 
-RUN_REFRESH_ADDON = 15
-RUN_REFRESH_MOUNTS = 900
-
 PLUGIN_AUTO_UPDATE_CONDITIONS = PLUGIN_UPDATE_CONDITIONS + [JobCondition.RUNNING]
 
 
@@ -65,7 +63,7 @@ class Tasks(CoreSysAttributes):
         self.sys_scheduler.register_task(self.sys_backups.reload, RUN_RELOAD_BACKUPS)
         self.sys_scheduler.register_task(self.sys_host.reload, RUN_RELOAD_HOST)
         self.sys_scheduler.register_task(self.sys_ingress.reload, RUN_RELOAD_INGRESS)
-        self.sys_scheduler.register_task(self.sys_mounts.reload, RUN_REFRESH_MOUNTS)
+        self.sys_scheduler.register_task(self.sys_mounts.reload, RUN_RELOAD_MOUNTS)
 
         # Watchdog
         self.sys_scheduler.register_task(
@@ -78,12 +76,12 @@ class Tasks(CoreSysAttributes):
             self._watchdog_addon_application, RUN_WATCHDOG_ADDON_APPLICATON
         )
 
-        # Refresh
-        self.sys_scheduler.register_task(self._refresh_addon, RUN_REFRESH_ADDON)
-
         _LOGGER.info("All core tasks are scheduled")
 
-    @Job(conditions=ADDON_UPDATE_CONDITIONS + [JobCondition.RUNNING])
+    @Job(
+        name="tasks_update_addons",
+        conditions=ADDON_UPDATE_CONDITIONS + [JobCondition.RUNNING],
+    )
     async def _update_addons(self):
         """Check if an update is available for an Add-on and update it."""
         start_tasks: list[Awaitable[None]] = []
@@ -112,13 +110,14 @@ class Tasks(CoreSysAttributes):
         await asyncio.gather(*start_tasks)
 
     @Job(
+        name="tasks_update_supervisor",
         conditions=[
             JobCondition.AUTO_UPDATE,
             JobCondition.FREE_SPACE,
             JobCondition.HEALTHY,
             JobCondition.INTERNET_HOST,
             JobCondition.RUNNING,
-        ]
+        ],
     )
     async def _update_supervisor(self):
         """Check and run update of Supervisor Supervisor."""
@@ -172,7 +171,7 @@ class Tasks(CoreSysAttributes):
         finally:
             self._cache[HASS_WATCHDOG_API] = 0
 
-    @Job(conditions=PLUGIN_AUTO_UPDATE_CONDITIONS)
+    @Job(name="tasks_update_cli", conditions=PLUGIN_AUTO_UPDATE_CONDITIONS)
     async def _update_cli(self):
         """Check and run update of cli."""
         if not self.sys_plugins.cli.need_update:
@@ -183,7 +182,7 @@ class Tasks(CoreSysAttributes):
         )
         await self.sys_plugins.cli.update()
 
-    @Job(conditions=PLUGIN_AUTO_UPDATE_CONDITIONS)
+    @Job(name="tasks_update_dns", conditions=PLUGIN_AUTO_UPDATE_CONDITIONS)
     async def _update_dns(self):
         """Check and run update of CoreDNS plugin."""
         if not self.sys_plugins.dns.need_update:
@@ -195,7 +194,7 @@ class Tasks(CoreSysAttributes):
         )
         await self.sys_plugins.dns.update()
 
-    @Job(conditions=PLUGIN_AUTO_UPDATE_CONDITIONS)
+    @Job(name="tasks_update_audio", conditions=PLUGIN_AUTO_UPDATE_CONDITIONS)
     async def _update_audio(self):
         """Check and run update of PulseAudio plugin."""
         if not self.sys_plugins.audio.need_update:
@@ -207,7 +206,7 @@ class Tasks(CoreSysAttributes):
         )
         await self.sys_plugins.audio.update()
 
-    @Job(conditions=PLUGIN_AUTO_UPDATE_CONDITIONS)
+    @Job(name="tasks_update_observer", conditions=PLUGIN_AUTO_UPDATE_CONDITIONS)
     async def _update_observer(self):
         """Check and run update of Observer plugin."""
         if not self.sys_plugins.observer.need_update:
@@ -219,7 +218,7 @@ class Tasks(CoreSysAttributes):
         )
         await self.sys_plugins.observer.update()
 
-    @Job(conditions=PLUGIN_AUTO_UPDATE_CONDITIONS)
+    @Job(name="tasks_update_multicast", conditions=PLUGIN_AUTO_UPDATE_CONDITIONS)
     async def _update_multicast(self):
         """Check and run update of multicast."""
         if not self.sys_plugins.multicast.need_update:
@@ -278,21 +277,7 @@ class Tasks(CoreSysAttributes):
             finally:
                 self._cache[addon.slug] = 0
 
-    async def _refresh_addon(self) -> None:
-        """Refresh addon state."""
-        for addon in self.sys_addons.installed:
-            # if watchdog need looking for
-            if addon.watchdog or addon.state != AddonState.STARTED:
-                continue
-
-            # if Addon have running actions
-            if addon.in_progress or await addon.is_running():
-                continue
-
-            # Adjust state
-            addon.state = AddonState.STOPPED
-
-    @Job(conditions=[JobCondition.SUPERVISOR_UPDATED])
+    @Job(name="tasks_reload_store", conditions=[JobCondition.SUPERVISOR_UPDATED])
     async def _reload_store(self) -> None:
         """Reload store and check for addon updates."""
         await self.sys_store.reload()

supervisor/mounts/manager.py

@@ -139,7 +139,7 @@ class MountManager(FileConfiguration, CoreSysAttributes):
                 ]
             )
 
-    @Job(conditions=[JobCondition.MOUNT_AVAILABLE])
+    @Job(name="mount_manager_reload", conditions=[JobCondition.MOUNT_AVAILABLE])
     async def reload(self) -> None:
         """Update mounts info via dbus and reload failed mounts."""
         if not self.mounts:
@@ -180,9 +180,17 @@ class MountManager(FileConfiguration, CoreSysAttributes):
                 ],
             )
 
-    @Job(conditions=[JobCondition.MOUNT_AVAILABLE], on_condition=MountJobError)
+    @Job(
+        name="mount_manager_create_mount",
+        conditions=[JobCondition.MOUNT_AVAILABLE],
+        on_condition=MountJobError,
+    )
     async def create_mount(self, mount: Mount) -> None:
         """Add/update a mount."""
+        # Add mount name to job
+        if job := self.sys_jobs.get_job():
+            job.reference = mount.name
+
         if mount.name in self._mounts:
             _LOGGER.debug("Mount '%s' exists, unmounting then mounting from new config")
             await self.remove_mount(mount.name, retain_entry=True)
@@ -200,9 +208,17 @@ class MountManager(FileConfiguration, CoreSysAttributes):
         elif mount.usage == MountUsage.SHARE:
             await self._bind_share(mount)
 
-    @Job(conditions=[JobCondition.MOUNT_AVAILABLE], on_condition=MountJobError)
+    @Job(
+        name="mount_manager_remove_mount",
+        conditions=[JobCondition.MOUNT_AVAILABLE],
+        on_condition=MountJobError,
+    )
     async def remove_mount(self, name: str, *, retain_entry: bool = False) -> None:
         """Remove a mount."""
+        # Add mount name to job
+        if job := self.sys_jobs.get_job():
+            job.reference = name
+
         if name not in self._mounts:
             raise MountNotFound(
                 f"Cannot remove '{name}', no mount exists with that name"
@@ -223,9 +239,17 @@ class MountManager(FileConfiguration, CoreSysAttributes):
 
         return mount
 
-    @Job(conditions=[JobCondition.MOUNT_AVAILABLE], on_condition=MountJobError)
+    @Job(
+        name="mount_manager_reload_mount",
+        conditions=[JobCondition.MOUNT_AVAILABLE],
+        on_condition=MountJobError,
+    )
     async def reload_mount(self, name: str) -> None:
         """Reload a mount to retry mounting with same config."""
+        # Add mount name to job
+        if job := self.sys_jobs.get_job():
+            job.reference = name
+
         if name not in self._mounts:
             raise MountNotFound(
                 f"Cannot reload '{name}', no mount exists with that name"

supervisor/os/data_disk.py

@@ -165,7 +165,7 @@ class DataDisk(CoreSysAttributes):
             if block.drive == drive.object_path
         ]
 
-    @Job(conditions=[JobCondition.OS_AGENT])
+    @Job(name="data_disk_load", conditions=[JobCondition.OS_AGENT])
     async def load(self) -> None:
         """Load DataDisk feature."""
         # Update datadisk details on OS-Agent
@@ -173,6 +173,7 @@ class DataDisk(CoreSysAttributes):
             await self.sys_dbus.agent.datadisk.reload_device()
 
     @Job(
+        name="data_disk_migrate",
         conditions=[JobCondition.HAOS, JobCondition.OS_AGENT, JobCondition.HEALTHY],
         limit=JobExecutionLimit.ONCE,
         on_condition=HassOSJobError,

supervisor/os/manager.py

@@ -156,6 +156,7 @@ class OSManager(CoreSysAttributes):
         )
 
     @Job(
+        name="os_manager_config_sync",
         conditions=[JobCondition.HAOS],
         on_condition=HassOSJobError,
     )
@@ -170,6 +171,7 @@ class OSManager(CoreSysAttributes):
         await self.sys_host.services.restart("hassos-config.service")
 
     @Job(
+        name="os_manager_update",
         conditions=[
             JobCondition.HAOS,
             JobCondition.INTERNET_SYSTEM,
@@ -225,7 +227,7 @@ class OSManager(CoreSysAttributes):
         )
         raise HassOSUpdateError()
 
-    @Job(conditions=[JobCondition.HAOS])
+    @Job(name="os_manager_mark_healthy", conditions=[JobCondition.HAOS])
     async def mark_healthy(self) -> None:
         """Set booted partition as good for rauc."""
         try:

supervisor/plugins/audio.py

@@ -118,6 +118,7 @@ class PluginAudio(PluginBase):
         self.save_data()
 
     @Job(
+        name="plugin_audio_update",
         conditions=PLUGIN_UPDATE_CONDITIONS,
         on_condition=AudioJobError,
     )
@@ -218,6 +219,7 @@ class PluginAudio(PluginBase):
             ) from err
 
     @Job(
+        name="plugin_audio_restart_after_problem",
         limit=JobExecutionLimit.THROTTLE_RATE_LIMIT,
         throttle_period=WATCHDOG_THROTTLE_PERIOD,
         throttle_max_calls=WATCHDOG_THROTTLE_MAX_CALLS,

supervisor/plugins/cli.py

@@ -75,6 +75,7 @@ class PluginCli(PluginBase):
         self.save_data()
 
     @Job(
+        name="plugin_cli_update",
         conditions=PLUGIN_UPDATE_CONDITIONS,
         on_condition=CliJobError,
     )
@@ -151,6 +152,7 @@ class PluginCli(PluginBase):
             capture_exception(err)
 
     @Job(
+        name="plugin_cli_restart_after_problem",
         limit=JobExecutionLimit.THROTTLE_RATE_LIMIT,
         throttle_period=WATCHDOG_THROTTLE_PERIOD,
         throttle_max_calls=WATCHDOG_THROTTLE_MAX_CALLS,

supervisor/plugins/dns.py

@@ -187,6 +187,7 @@ class PluginDns(PluginBase):
         await self.write_hosts()
 
     @Job(
+        name="plugin_dns_update",
         conditions=PLUGIN_UPDATE_CONDITIONS,
         on_condition=CoreDNSJobError,
     )
@@ -269,6 +270,7 @@ class PluginDns(PluginBase):
         return await super().watchdog_container(event)
 
     @Job(
+        name="plugin_dns_restart_after_problem",
         limit=JobExecutionLimit.THROTTLE_RATE_LIMIT,
         throttle_period=WATCHDOG_THROTTLE_PERIOD,
         throttle_max_calls=WATCHDOG_THROTTLE_MAX_CALLS,

supervisor/plugins/multicast.py

@@ -71,6 +71,7 @@ class PluginMulticast(PluginBase):
         self.save_data()
 
     @Job(
+        name="plugin_multicast_update",
         conditions=PLUGIN_UPDATE_CONDITIONS,
         on_condition=MulticastJobError,
     )
@@ -146,6 +147,7 @@ class PluginMulticast(PluginBase):
             capture_exception(err)
 
     @Job(
+        name="plugin_multicast_restart_after_problem",
         limit=JobExecutionLimit.THROTTLE_RATE_LIMIT,
         throttle_period=WATCHDOG_THROTTLE_PERIOD,
         throttle_max_calls=WATCHDOG_THROTTLE_MAX_CALLS,

supervisor/plugins/observer.py

@@ -79,6 +79,7 @@ class PluginObserver(PluginBase):
         self.save_data()
 
     @Job(
+        name="plugin_observer_update",
         conditions=PLUGIN_UPDATE_CONDITIONS,
         on_condition=ObserverJobError,
     )
@@ -156,6 +157,7 @@ class PluginObserver(PluginBase):
             capture_exception(err)
 
     @Job(
+        name="plugin_observer_restart_after_problem",
         limit=JobExecutionLimit.THROTTLE_RATE_LIMIT,
         throttle_period=WATCHDOG_THROTTLE_PERIOD,
         throttle_max_calls=WATCHDOG_THROTTLE_MAX_CALLS,

supervisor/resolution/checks/addon_pwned.py

@@ -22,6 +22,7 @@ class CheckAddonPwned(CheckBase):
     """CheckAddonPwned class for check."""
 
     @Job(
+        name="check_addon_pwned_run",
         conditions=[JobCondition.INTERNET_SYSTEM],
         limit=JobExecutionLimit.THROTTLE,
         throttle_period=timedelta(hours=24),
@@ -62,7 +63,7 @@ class CheckAddonPwned(CheckBase):
                 except PwnedError:
                     pass
 
-    @Job(conditions=[JobCondition.INTERNET_SYSTEM])
+    @Job(name="check_addon_pwned_approve", conditions=[JobCondition.INTERNET_SYSTEM])
     async def approve_check(self, reference: str | None = None) -> bool:
         """Approve check if it is affected by issue."""
         addon = self.sys_addons.get(reference)

supervisor/resolution/checks/dns_server.py

@@ -23,6 +23,7 @@ class CheckDNSServer(CheckBase):
     """CheckDNSServer class for check."""
 
     @Job(
+        name="check_dns_server_run",
         conditions=[JobCondition.INTERNET_SYSTEM],
         limit=JobExecutionLimit.THROTTLE,
         throttle_period=timedelta(hours=24),
@@ -42,7 +43,7 @@ class CheckDNSServer(CheckBase):
             )
             capture_exception(results[i])
 
-    @Job(conditions=[JobCondition.INTERNET_SYSTEM])
+    @Job(name="check_dns_server_approve", conditions=[JobCondition.INTERNET_SYSTEM])
     async def approve_check(self, reference: str | None = None) -> bool:
         """Approve check if it is affected by issue."""
         if reference not in self.dns_servers:

supervisor/resolution/checks/dns_server_ipv6.py

@@ -23,6 +23,7 @@ class CheckDNSServerIPv6(CheckBase):
     """CheckDNSServerIPv6 class for check."""
 
     @Job(
+        name="check_dns_server_ipv6_run",
         conditions=[JobCondition.INTERNET_SYSTEM],
         limit=JobExecutionLimit.THROTTLE,
         throttle_period=timedelta(hours=24),
@@ -47,7 +48,9 @@ class CheckDNSServerIPv6(CheckBase):
             )
             capture_exception(results[i])
 
-    @Job(conditions=[JobCondition.INTERNET_SYSTEM])
+    @Job(
+        name="check_dns_server_ipv6_approve", conditions=[JobCondition.INTERNET_SYSTEM]
+    )
     async def approve_check(self, reference: str | None = None) -> bool:
         """Approve check if it is affected by issue."""
         if reference not in self.dns_servers:

supervisor/resolution/fixup.py

@@ -36,7 +36,10 @@ class ResolutionFixup(CoreSysAttributes):
         """Return a list of all fixups."""
         return list(self._fixups.values())
 
-    @Job(conditions=[JobCondition.HEALTHY, JobCondition.RUNNING])
+    @Job(
+        name="resolution_fixup_run_autofix",
+        conditions=[JobCondition.HEALTHY, JobCondition.RUNNING],
+    )
     async def run_autofix(self) -> None:
         """Run all startup fixes."""
         _LOGGER.info("Starting system autofix at state %s", self.sys_core.state)

supervisor/resolution/fixups/store_execute_reload.py

@@ -25,6 +25,7 @@ class FixupStoreExecuteReload(FixupBase):
     """Storage class for fixup."""
 
     @Job(
+        name="fixup_store_execute_reload_process",
         conditions=[JobCondition.INTERNET_SYSTEM, JobCondition.FREE_SPACE],
         on_condition=ResolutionFixupJobError,
     )

supervisor/resolution/fixups/store_execute_reset.py

@@ -26,6 +26,7 @@ class FixupStoreExecuteReset(FixupBase):
     """Storage class for fixup."""
 
     @Job(
+        name="fixup_store_execute_reset_process",
         conditions=[JobCondition.INTERNET_SYSTEM, JobCondition.FREE_SPACE],
         on_condition=ResolutionFixupJobError,
     )

supervisor/resolution/fixups/system_execute_integrity.py

@@ -22,6 +22,7 @@ class FixupSystemExecuteIntegrity(FixupBase):
     """Storage class for fixup."""
 
     @Job(
+        name="fixup_system_execute_integrity_process",
         conditions=[JobCondition.INTERNET_SYSTEM],
         on_condition=ResolutionFixupJobError,
         limit=JobExecutionLimit.THROTTLE,

supervisor/security/module.py

@@ -103,6 +103,7 @@ class Security(FileConfiguration, CoreSysAttributes):
             return
 
     @Job(
+        name="security_manager_integrity_check",
         conditions=[JobCondition.INTERNET_SYSTEM],
         on_condition=SecurityJobError,
         limit=JobExecutionLimit.ONCE,

supervisor/store/__init__.py

@@ -80,7 +80,11 @@ class StoreManager(CoreSysAttributes, FileConfiguration):
             self._data[ATTR_REPOSITORIES], add_with_errors=True
         )
 
-    @Job(conditions=[JobCondition.SUPERVISOR_UPDATED], on_condition=StoreJobError)
+    @Job(
+        name="store_manager_reload",
+        conditions=[JobCondition.SUPERVISOR_UPDATED],
+        on_condition=StoreJobError,
+    )
     async def reload(self) -> None:
         """Update add-ons from repository and reload list."""
         tasks = [self.sys_create_task(repository.update()) for repository in self.all]
@@ -92,6 +96,7 @@ class StoreManager(CoreSysAttributes, FileConfiguration):
         self._read_addons()
 
     @Job(
+        name="store_manager_add_repository",
         conditions=[JobCondition.INTERNET_SYSTEM, JobCondition.SUPERVISOR_UPDATED],
         on_condition=StoreJobError,
     )

supervisor/store/data.py

@@ -1,5 +1,5 @@
 """Init file for Supervisor add-on data."""
-from collections.abc import Awaitable
+from dataclasses import dataclass
 import logging
 from pathlib import Path
 from typing import Any
@@ -29,6 +29,72 @@ from .validate import SCHEMA_REPOSITORY_CONFIG
 _LOGGER: logging.Logger = logging.getLogger(__name__)
 
 
+@dataclass(slots=True)
+class ProcessedRepository:
+    """Representation of a repository processed from its git folder."""
+
+    slug: str
+    path: Path
+    config: dict[str, Any]
+
+
+def _read_addon_translations(addon_path: Path) -> dict:
+    """Read translations from add-ons folder.
+
+    Should be run in the executor.
+    """
+    translations_dir = addon_path / "translations"
+    translations = {}
+
+    if not translations_dir.exists():
+        return translations
+
+    translation_files = [
+        translation
+        for translation in translations_dir.glob("*")
+        if translation.suffix in FILE_SUFFIX_CONFIGURATION
+    ]
+
+    for translation in translation_files:
+        try:
+            translations[translation.stem] = SCHEMA_ADDON_TRANSLATIONS(
+                read_json_or_yaml_file(translation)
+            )
+
+        except (ConfigurationFileError, vol.Invalid) as err:
+            _LOGGER.warning("Can't read translations from %s - %s", translation, err)
+            continue
+
+    return translations
+
+
+def _read_git_repository(path: Path) -> ProcessedRepository | None:
+    """Process a custom repository folder."""
+    slug = extract_hash_from_path(path)
+
+    # exists repository json
+    try:
+        repository_file = find_one_filetype(
+            path, "repository", FILE_SUFFIX_CONFIGURATION
+        )
+    except ConfigurationFileError:
+        _LOGGER.warning("No repository information exists at %s", path)
+        return None
+
+    try:
+        return ProcessedRepository(
+            slug,
+            path,
+            SCHEMA_REPOSITORY_CONFIG(read_json_or_yaml_file(repository_file)),
+        )
+    except ConfigurationFileError:
+        _LOGGER.warning("Can't read repository information from %s", repository_file)
+        return None
+    except vol.Invalid:
+        _LOGGER.warning("Repository parse error %s", repository_file)
+        return None
+
+
 class StoreData(CoreSysAttributes):
     """Hold data for Add-ons inside Supervisor."""
 
@@ -38,63 +104,43 @@ class StoreData(CoreSysAttributes):
         self.repositories: dict[str, Any] = {}
         self.addons: dict[str, Any] = {}
 
-    async def update(self) -> Awaitable[None]:
+    async def update(self) -> None:
         """Read data from add-on repository."""
-        return await self.sys_run_in_executor(self._update)
-
-    def _update(self) -> None:
         self.repositories.clear()
         self.addons.clear()
 
         # read core repository
-        self._read_addons_folder(self.sys_config.path_addons_core, REPOSITORY_CORE)
+        await self._read_addons_folder(
+            self.sys_config.path_addons_core, REPOSITORY_CORE
+        )
 
         # read local repository
-        self._read_addons_folder(self.sys_config.path_addons_local, REPOSITORY_LOCAL)
+        await self._read_addons_folder(
+            self.sys_config.path_addons_local, REPOSITORY_LOCAL
+        )
 
         # add built-in repositories information
-        self._set_builtin_repositories()
+        await self._set_builtin_repositories()
 
         # read custom git repositories
-        for repository_element in self.sys_config.path_addons_git.iterdir():
-            if repository_element.is_dir():
-                self._read_git_repository(repository_element)
+        def _read_git_repositories() -> list[ProcessedRepository]:
+            return [
+                repo
+                for repository_element in self.sys_config.path_addons_git.iterdir()
+                if repository_element.is_dir()
+                and (repo := _read_git_repository(repository_element))
+            ]
 
-    def _read_git_repository(self, path: Path) -> None:
-        """Process a custom repository folder."""
-        slug = extract_hash_from_path(path)
+        for repo in await self.sys_run_in_executor(_read_git_repositories):
+            self.repositories[repo.slug] = repo.config
+            await self._read_addons_folder(repo.path, repo.slug)
 
-        # exists repository json
-        try:
-            repository_file = find_one_filetype(
-                path, "repository", FILE_SUFFIX_CONFIGURATION
-            )
-        except ConfigurationFileError:
-            _LOGGER.warning("No repository information exists at %s", path)
-            return
-
-        try:
-            repository_info = SCHEMA_REPOSITORY_CONFIG(
-                read_json_or_yaml_file(repository_file)
-            )
-        except ConfigurationFileError:
-            _LOGGER.warning(
-                "Can't read repository information from %s", repository_file
-            )
-            return
-        except vol.Invalid:
-            _LOGGER.warning("Repository parse error %s", repository_file)
-            return
-
-        # process data
-        self.repositories[slug] = repository_info
-        self._read_addons_folder(path, slug)
-
-    def _find_addons(self, path: Path, repository: dict) -> list[Path] | None:
+    async def _find_addons(self, path: Path, repository: dict) -> list[Path] | None:
         """Find add-ons in the path."""
-        try:
+
+        def _get_addons_list() -> list[Path]:
             # Generate a list without artefact, safe for corruptions
-            addon_list = [
+            return [
                 addon
                 for addon in path.glob("**/config.*")
                 if not [
@@ -104,6 +150,9 @@ class StoreData(CoreSysAttributes):
                 ]
                 and addon.suffix in FILE_SUFFIX_CONFIGURATION
             ]
+
+        try:
+            addon_list = await self.sys_run_in_executor(_get_addons_list)
         except OSError as err:
             suggestion = None
             if path.stem != StoreType.LOCAL:
@@ -120,77 +169,59 @@ class StoreData(CoreSysAttributes):
             return None
         return addon_list
 
-    def _read_addons_folder(self, path: Path, repository: dict) -> None:
+    async def _read_addons_folder(self, path: Path, repository: str) -> None:
         """Read data from add-ons folder."""
-        if not (addon_list := self._find_addons(path, repository)):
+        if not (addon_list := await self._find_addons(path, repository)):
             return
 
-        for addon in addon_list:
-            try:
-                addon_config = read_json_or_yaml_file(addon)
-            except ConfigurationFileError:
-                _LOGGER.warning("Can't read %s from repository %s", addon, repository)
-                continue
+        def _process_addons_config() -> dict[str, dict[str, Any]]:
+            addons_config: dict[str, dict[str, Any]] = {}
+            for addon in addon_list:
+                try:
+                    addon_config = read_json_or_yaml_file(addon)
+                except ConfigurationFileError:
+                    _LOGGER.warning(
+                        "Can't read %s from repository %s", addon, repository
+                    )
+                    continue
 
-            # validate
-            try:
-                addon_config = SCHEMA_ADDON_CONFIG(addon_config)
-            except vol.Invalid as ex:
-                _LOGGER.warning(
-                    "Can't read %s: %s", addon, humanize_error(addon_config, ex)
-                )
-                continue
+                # validate
+                try:
+                    addon_config = SCHEMA_ADDON_CONFIG(addon_config)
+                except vol.Invalid as ex:
+                    _LOGGER.warning(
+                        "Can't read %s: %s", addon, humanize_error(addon_config, ex)
+                    )
+                    continue
 
-            # Generate slug
-            addon_slug = f"{repository}_{addon_config[ATTR_SLUG]}"
+                # Generate slug
+                addon_slug = f"{repository}_{addon_config[ATTR_SLUG]}"
 
-            # store
-            addon_config[ATTR_REPOSITORY] = repository
-            addon_config[ATTR_LOCATON] = str(addon.parent)
-            addon_config[ATTR_TRANSLATIONS] = self._read_addon_translations(
-                addon.parent
-            )
-            self.addons[addon_slug] = addon_config
+                # store
+                addon_config[ATTR_REPOSITORY] = repository
+                addon_config[ATTR_LOCATON] = str(addon.parent)
+                addon_config[ATTR_TRANSLATIONS] = _read_addon_translations(addon.parent)
+                addons_config[addon_slug] = addon_config
 
-    def _set_builtin_repositories(self):
+            return addons_config
+
+        self.addons.update(await self.sys_run_in_executor(_process_addons_config))
+
+    async def _set_builtin_repositories(self):
         """Add local built-in repository into dataset."""
-        try:
-            builtin_file = Path(__file__).parent.joinpath("built-in.json")
-            builtin_data = read_json_file(builtin_file)
-        except ConfigurationFileError:
-            _LOGGER.warning("Can't read built-in json")
-            return
 
-        # core repository
-        self.repositories[REPOSITORY_CORE] = builtin_data[REPOSITORY_CORE]
-
-        # local repository
-        self.repositories[REPOSITORY_LOCAL] = builtin_data[REPOSITORY_LOCAL]
-
-    def _read_addon_translations(self, addon_path: Path) -> dict:
-        """Read translations from add-ons folder."""
-        translations_dir = addon_path / "translations"
-        translations = {}
-
-        if not translations_dir.exists():
-            return translations
-
-        translation_files = [
-            translation
-            for translation in translations_dir.glob("*")
-            if translation.suffix in FILE_SUFFIX_CONFIGURATION
-        ]
-
-        for translation in translation_files:
+        def _get_builtins() -> dict[str, dict[str, str]] | None:
             try:
-                translations[translation.stem] = SCHEMA_ADDON_TRANSLATIONS(
-                    read_json_or_yaml_file(translation)
-                )
+                builtin_file = Path(__file__).parent.joinpath("built-in.json")
+                return read_json_file(builtin_file)
+            except ConfigurationFileError:
+                _LOGGER.warning("Can't read built-in json")
+                return None
 
-            except (ConfigurationFileError, vol.Invalid) as err:
-                _LOGGER.warning(
-                    "Can't read translations from %s - %s", translation, err
-                )
-                continue
+        builtin_data = await self.sys_run_in_executor(_get_builtins)
+        if builtin_data:
+            # core repository
+            self.repositories[REPOSITORY_CORE] = builtin_data[REPOSITORY_CORE]
 
-        return translations
+            # local repository
+            self.repositories[REPOSITORY_LOCAL] = builtin_data[REPOSITORY_LOCAL]

supervisor/store/git.py

@@ -77,6 +77,7 @@ class GitRepo(CoreSysAttributes):
                 raise StoreGitError() from err
 
     @Job(
+        name="git_repo_clone",
         conditions=[JobCondition.FREE_SPACE, JobCondition.INTERNET_SYSTEM],
         on_condition=StoreJobError,
     )
@@ -112,6 +113,7 @@ class GitRepo(CoreSysAttributes):
                 raise StoreGitCloneError() from err
 
     @Job(
+        name="git_repo_pull",
         conditions=[JobCondition.FREE_SPACE, JobCondition.INTERNET_SYSTEM],
         on_condition=StoreJobError,
     )

supervisor/supervisor.py

@@ -211,7 +211,11 @@ class Supervisor(CoreSysAttributes):
 
         self.sys_create_task(self.sys_core.stop())
 
-    @Job(conditions=[JobCondition.RUNNING], on_condition=SupervisorJobError)
+    @Job(
+        name="supervisor_restart",
+        conditions=[JobCondition.RUNNING],
+        on_condition=SupervisorJobError,
+    )
     async def restart(self) -> None:
         """Restart Supervisor soft."""
         self.sys_core.exit_code = 100
@@ -255,6 +259,7 @@ class Supervisor(CoreSysAttributes):
             _LOGGER.error("Repair of Supervisor failed")
 
     @Job(
+        name="supervisor_check_connectivity",
         limit=JobExecutionLimit.THROTTLE,
         throttle_period=_check_connectivity_throttle_period,
     )

supervisor/updater.py

@@ -181,6 +181,7 @@ class Updater(FileConfiguration, CoreSysAttributes):
         self._data[ATTR_AUTO_UPDATE] = value
 
     @Job(
+        name="updater_fetch_data",
         conditions=[JobCondition.INTERNET_SYSTEM],
         on_condition=UpdaterJobError,
         limit=JobExecutionLimit.THROTTLE_WAIT,

supervisor/validate.py

@@ -30,6 +30,8 @@ from .const import (
     ATTR_PWNED,
     ATTR_REGISTRIES,
     ATTR_SESSION,
+    ATTR_SESSION_DATA,
+    ATTR_SESSION_DATA_USER,
     ATTR_SUPERVISOR,
     ATTR_TIMEZONE,
     ATTR_USERNAME,
@@ -49,7 +51,9 @@ RE_REGISTRY = re.compile(r"^([a-z0-9]+(-[a-z0-9]+)*\.)+[a-z]{2,}$")
 # pylint: disable=invalid-name
 network_port = vol.All(vol.Coerce(int), vol.Range(min=1, max=65535))
 wait_boot = vol.All(vol.Coerce(int), vol.Range(min=1, max=60))
-docker_image = vol.Match(r"^([a-zA-Z\-\.:\d{}]+/)*?([\-\w{}]+)/([\-\w{}]+)$")
+docker_image = vol.Match(
+    r"^([a-z0-9][a-z0-9.\-]*(:[0-9]+)?/)*?([a-z0-9{][a-z0-9.\-_{}]*/)*?([a-z0-9{][a-z0-9.\-_{}]*)$"
+)
 uuid_match = vol.Match(r"^[0-9a-f]{32}$")
 sha256 = vol.Match(r"^[0-9a-f]{64}$")
 token = vol.Match(r"^[0-9a-f]{32,256}$")
@@ -176,18 +180,33 @@ SCHEMA_DOCKER_CONFIG = vol.Schema(
 
 SCHEMA_AUTH_CONFIG = vol.Schema({sha256: sha256})
 
+SCHEMA_SESSION_DATA = vol.Schema(
+    {
+        token: vol.Schema(
+            {
+                vol.Required(ATTR_SESSION_DATA_USER): vol.Schema(
+                    {
+                        vol.Required("id"): str,
+                        vol.Required("username"): str,
+                        vol.Required("displayname"): str,
+                    }
+                )
+            }
+        )
+    }
+)
 
 SCHEMA_INGRESS_CONFIG = vol.Schema(
     {
         vol.Required(ATTR_SESSION, default=dict): vol.Schema(
             {token: vol.Coerce(float)}
         ),
+        vol.Required(ATTR_SESSION_DATA, default=dict): SCHEMA_SESSION_DATA,
         vol.Required(ATTR_PORTS, default=dict): vol.Schema({str: network_port}),
     },
     extra=vol.REMOVE_EXTRA,
 )
 
-
 # pylint: disable=no-value-for-parameter
 SCHEMA_SECURITY_CONFIG = vol.Schema(
     {

tests/addons/test_addon.py

@@ -18,6 +18,7 @@ from supervisor.docker.addon import DockerAddon
 from supervisor.docker.const import ContainerState
 from supervisor.docker.monitor import DockerContainerStateEvent
 from supervisor.exceptions import AddonsError, AddonsJobError, AudioUpdateError
+from supervisor.ingress import Ingress
 from supervisor.store.repository import Repository
 from supervisor.utils.dt import utcnow
 
@@ -523,18 +524,44 @@ async def test_restore(
     path_extern,
 ) -> None:
     """Test restoring an addon."""
+    coresys.hardware.disk.get_disk_free_space = lambda x: 5000
     install_addon_ssh.path_data.mkdir()
     await install_addon_ssh.load()
 
     tarfile = SecureTarFile(get_fixture_path(f"backup_local_ssh_{status}.tar.gz"), "r")
     with patch.object(DockerAddon, "is_running", return_value=False), patch.object(
         CpuArch, "supported", new=PropertyMock(return_value=["aarch64"])
-    ):
+    ), patch.object(Ingress, "update_hass_panel") as update_hass_panel:
         start_task = await coresys.addons.restore(TEST_ADDON_SLUG, tarfile)
 
+        update_hass_panel.assert_called_once()
+
     assert bool(start_task) is (status == "running")
 
 
+async def test_restore_while_running(
+    coresys: CoreSys,
+    install_addon_ssh: Addon,
+    container: MagicMock,
+    tmp_supervisor_data,
+    path_extern,
+):
+    """Test restore of a running addon."""
+    container.status = "running"
+    coresys.hardware.disk.get_disk_free_space = lambda x: 5000
+    install_addon_ssh.path_data.mkdir()
+    await install_addon_ssh.load()
+
+    tarfile = SecureTarFile(get_fixture_path("backup_local_ssh_stopped.tar.gz"), "r")
+    with patch.object(DockerAddon, "is_running", return_value=True), patch.object(
+        CpuArch, "supported", new=PropertyMock(return_value=["aarch64"])
+    ), patch.object(Ingress, "update_hass_panel"):
+        start_task = await coresys.addons.restore(TEST_ADDON_SLUG, tarfile)
+
+    assert bool(start_task) is False
+    container.stop.assert_called_once()
+
+
 async def test_start_when_running(
     coresys: CoreSys,
     install_addon_ssh: Addon,

tests/addons/test_config.py

@@ -110,7 +110,7 @@ def test_invalid_repository():
     """Validate basic config with invalid repositories."""
     config = load_json_fixture("basic-addon-config.json")
 
-    config["image"] = "something"
+    config["image"] = "-invalid-something"
     with pytest.raises(vol.Invalid):
         vd.SCHEMA_ADDON_CONFIG(config)
 

tests/addons/test_manager.py

@@ -1,6 +1,7 @@
 """Test addon manager."""
 
 import asyncio
+from pathlib import Path
 from unittest.mock import AsyncMock, MagicMock, Mock, PropertyMock, patch
 
 from awesomeversion import AwesomeVersion
@@ -8,6 +9,7 @@ import pytest
 
 from supervisor.addons.addon import Addon
 from supervisor.arch import CpuArch
+from supervisor.config import CoreConfig
 from supervisor.const import AddonBoot, AddonStartup, AddonState, BusEvent
 from supervisor.coresys import CoreSys
 from supervisor.docker.addon import DockerAddon
@@ -22,6 +24,7 @@ from supervisor.exceptions import (
 )
 from supervisor.plugins.dns import PluginDns
 from supervisor.utils import check_exception_chain
+from supervisor.utils.common import write_json_file
 
 from tests.common import load_json_fixture
 from tests.const import TEST_ADDON_SLUG
@@ -57,7 +60,7 @@ async def test_image_added_removed_on_update(
     assert install_addon_ssh.image == "local/amd64-addon-ssh"
     assert coresys.addons.store.get(TEST_ADDON_SLUG).image == "test/amd64-my-ssh-addon"
 
-    with patch.object(DockerInterface, "_install") as install, patch.object(
+    with patch.object(DockerInterface, "install") as install, patch.object(
         DockerAddon, "_build"
     ) as build:
         await install_addon_ssh.update()
@@ -77,7 +80,7 @@ async def test_image_added_removed_on_update(
     assert install_addon_ssh.image == "test/amd64-my-ssh-addon"
     assert coresys.addons.store.get(TEST_ADDON_SLUG).image == "local/amd64-addon-ssh"
 
-    with patch.object(DockerInterface, "_install") as install, patch.object(
+    with patch.object(DockerInterface, "install") as install, patch.object(
         DockerAddon, "_build"
     ) as build:
         await install_addon_ssh.update()
@@ -249,7 +252,7 @@ async def test_update(
 
     assert install_addon_ssh.need_update is True
 
-    with patch.object(DockerInterface, "_install"), patch.object(
+    with patch.object(DockerInterface, "install"), patch.object(
         DockerAddon, "is_running", return_value=False
     ):
         start_task = await coresys.addons.update(TEST_ADDON_SLUG)
@@ -277,3 +280,87 @@ async def test_rebuild(
         start_task = await coresys.addons.rebuild(TEST_ADDON_SLUG)
 
     assert bool(start_task) is (status == "running")
+
+
+async def test_start_wait_cancel_on_uninstall(
+    coresys: CoreSys,
+    install_addon_ssh: Addon,
+    container: MagicMock,
+    caplog: pytest.LogCaptureFixture,
+    tmp_supervisor_data,
+    path_extern,
+) -> None:
+    """Test the addon wait task is cancelled when addon is uninstalled."""
+    install_addon_ssh.path_data.mkdir()
+    container.attrs["Config"] = {"Healthcheck": "exists"}
+    await install_addon_ssh.load()
+    await asyncio.sleep(0)
+    assert install_addon_ssh.state == AddonState.STOPPED
+
+    start_task = asyncio.create_task(await install_addon_ssh.start())
+    assert start_task
+
+    coresys.bus.fire_event(
+        BusEvent.DOCKER_CONTAINER_STATE_CHANGE,
+        DockerContainerStateEvent(
+            name=f"addon_{TEST_ADDON_SLUG}",
+            state=ContainerState.RUNNING,
+            id="abc123",
+            time=1,
+        ),
+    )
+    await asyncio.sleep(0.01)
+
+    assert not start_task.done()
+    assert install_addon_ssh.state == AddonState.STARTUP
+
+    caplog.clear()
+    await coresys.addons.uninstall(TEST_ADDON_SLUG)
+    await asyncio.sleep(0.01)
+    assert start_task.done()
+    assert "Wait for addon startup task cancelled" in caplog.text
+
+
+async def test_repository_file_missing(
+    coresys: CoreSys, tmp_supervisor_data: Path, caplog: pytest.LogCaptureFixture
+):
+    """Test repository file is missing."""
+    with patch.object(
+        CoreConfig,
+        "path_addons_git",
+        new=PropertyMock(return_value=tmp_supervisor_data / "addons" / "git"),
+    ):
+        repo_dir = coresys.config.path_addons_git / "test"
+        repo_dir.mkdir(parents=True)
+
+        await coresys.store.data.update()
+
+    assert f"No repository information exists at {repo_dir.as_posix()}" in caplog.text
+
+
+async def test_repository_file_error(
+    coresys: CoreSys, tmp_supervisor_data: Path, caplog: pytest.LogCaptureFixture
+):
+    """Test repository file is missing."""
+    with patch.object(
+        CoreConfig,
+        "path_addons_git",
+        new=PropertyMock(return_value=tmp_supervisor_data / "addons" / "git"),
+    ):
+        repo_dir = coresys.config.path_addons_git / "test"
+        repo_dir.mkdir(parents=True)
+
+        repo_file = repo_dir / "repository.json"
+
+        with repo_file.open("w") as file:
+            file.write("not json")
+
+        await coresys.store.data.update()
+        assert (
+            f"Can't read repository information from {repo_file.as_posix()}"
+            in caplog.text
+        )
+
+        write_json_file(repo_file, {"invalid": "bad"})
+        await coresys.store.data.update()
+        assert f"Repository parse error {repo_dir.as_posix()}" in caplog.text

tests/api/test_addons.py

@@ -171,6 +171,7 @@ async def test_api_addon_rebuild_healthcheck(
     path_extern,
 ):
     """Test rebuilding an addon waits for healthy."""
+    coresys.hardware.disk.get_disk_free_space = lambda x: 5000
     container.status = "running"
     install_addon_ssh.path_data.mkdir()
     container.attrs["Config"] = {"Healthcheck": "exists"}

tests/api/test_ingress.py

@@ -1,4 +1,5 @@
 """Test ingress API."""
+# pylint: disable=protected-access
 from unittest.mock import patch
 
 import pytest
@@ -37,3 +38,50 @@ async def test_validate_session(api_client, coresys):
         assert await resp.json() == {"result": "ok", "data": {}}
 
         assert coresys.ingress.sessions[session] > valid_time
+
+
+@pytest.mark.asyncio
+async def test_validate_session_with_user_id(api_client, coresys):
+    """Test validating ingress session with user ID passed."""
+    with patch("aiohttp.web_request.BaseRequest.__getitem__", return_value=None):
+        resp = await api_client.post(
+            "/ingress/validate_session",
+            json={"session": "non-existing"},
+        )
+        assert resp.status == 401
+
+    with patch(
+        "aiohttp.web_request.BaseRequest.__getitem__",
+        return_value=coresys.homeassistant,
+    ):
+        client = coresys.homeassistant.websocket._client
+        client.async_send_command.return_value = [
+            {"id": "some-id", "name": "Some Name", "username": "sn"}
+        ]
+
+        resp = await api_client.post("/ingress/session", json={"user_id": "some-id"})
+        result = await resp.json()
+
+        client.async_send_command.assert_called_with({"type": "config/auth/list"})
+
+        assert "session" in result["data"]
+        session = result["data"]["session"]
+        assert session in coresys.ingress.sessions
+
+        valid_time = coresys.ingress.sessions[session]
+
+        resp = await api_client.post(
+            "/ingress/validate_session",
+            json={"session": session},
+        )
+        assert resp.status == 200
+        assert await resp.json() == {"result": "ok", "data": {}}
+
+        assert coresys.ingress.sessions[session] > valid_time
+
+        assert session in coresys.ingress.sessions_data
+        assert coresys.ingress.get_session_data(session).user.id == "some-id"
+        assert coresys.ingress.get_session_data(session).user.username == "sn"
+        assert (
+            coresys.ingress.get_session_data(session).user.display_name == "Some Name"
+        )

tests/api/test_store.py

@@ -125,6 +125,7 @@ async def test_api_store_update_healthcheck(
     path_extern,
 ):
     """Test updating an addon with healthcheck waits for health status."""
+    coresys.hardware.disk.get_disk_free_space = lambda x: 5000
     container.status = "running"
     container.attrs["Config"] = {"Healthcheck": "exists"}
     install_addon_ssh.path_data.mkdir()
@@ -176,7 +177,7 @@ async def test_api_store_update_healthcheck(
         asyncio.create_task(container_events())
 
     with patch.object(DockerAddon, "run", new=container_events_task), patch.object(
-        DockerInterface, "_install"
+        DockerInterface, "install"
     ), patch.object(DockerAddon, "is_running", return_value=False), patch.object(
         CpuArch, "supported", new=PropertyMock(return_value=["amd64"])
     ):

tests/conftest.py

@@ -70,6 +70,13 @@ async def path_extern() -> None:
     yield
 
 
+@pytest.fixture
+async def supervisor_name() -> None:
+    """Set env for supervisor name."""
+    os.environ["SUPERVISOR_NAME"] = "hassio_supervisor"
+    yield
+
+
 @pytest.fixture
 async def docker() -> DockerAPI:
     """Mock DockerAPI."""
@@ -286,7 +293,13 @@ async def fixture_all_dbus_services(
 
 @pytest.fixture
 async def coresys(
-    event_loop, docker, dbus_session_bus, all_dbus_services, aiohttp_client, run_dir
+    event_loop,
+    docker,
+    dbus_session_bus,
+    all_dbus_services,
+    aiohttp_client,
+    run_dir,
+    supervisor_name,
 ) -> CoreSys:
     """Create a CoreSys Mock."""
     with patch("supervisor.bootstrap.initialize_system"), patch(
@@ -409,7 +422,9 @@ def sys_supervisor():
 
 @pytest.fixture
 async def api_client(
-    aiohttp_client, coresys: CoreSys, request: pytest.FixtureRequest
+    aiohttp_client,
+    coresys: CoreSys,
+    request: pytest.FixtureRequest,
 ) -> TestClient:
     """Fixture for RestAPI client."""
 
@@ -428,9 +443,7 @@ async def api_client(
     api = RestAPI(coresys)
     api.webapp = web.Application(middlewares=[_security_middleware])
     api.start = AsyncMock()
-    with patch("supervisor.docker.supervisor.os") as os:
-        os.environ = {"SUPERVISOR_NAME": "hassio_supervisor"}
-        await api.load()
+    await api.load()
     yield await aiohttp_client(api.webapp)
 
 
@@ -593,16 +606,12 @@ async def journald_logs(coresys: CoreSys) -> MagicMock:
 
 
 @pytest.fixture
-async def docker_logs(docker: DockerAPI) -> MagicMock:
+async def docker_logs(docker: DockerAPI, supervisor_name) -> MagicMock:
     """Mock log output for a container from docker."""
     container_mock = MagicMock()
     container_mock.logs.return_value = load_binary_fixture("logs_docker_container.txt")
     docker.containers.get.return_value = container_mock
-
-    with patch("supervisor.docker.supervisor.os") as os:
-        os.environ = {"SUPERVISOR_NAME": "hassio_supervisor"}
-
-        yield container_mock.logs
+    yield container_mock.logs
 
 
 @pytest.fixture
@@ -640,7 +649,6 @@ async def os_available(request: pytest.FixtureRequest) -> None:
 @pytest.fixture
 async def mount_propagation(docker: DockerAPI, coresys: CoreSys) -> None:
     """Mock supervisor connected to container with propagation set."""
-    os.environ["SUPERVISOR_NAME"] = "hassio_supervisor"
     docker.containers.get.return_value = supervisor = MagicMock()
     supervisor.attrs = {
         "Mounts": [

tests/dbus/network/setting/test_generate.py

@@ -1,6 +1,9 @@
 """Test settings generation from interface."""
 
+from unittest.mock import PropertyMock, patch
+
 from supervisor.dbus.network import NetworkManager
+from supervisor.dbus.network.interface import NetworkInterface
 from supervisor.dbus.network.setting.generate import get_connection_from_interface
 from supervisor.host.network import Interface
 
@@ -17,13 +20,24 @@ async def test_get_connection_from_interface(network_manager: NetworkManager):
 
     assert "interface-name" not in connection_payload["connection"]
     assert connection_payload["connection"]["type"].value == "802-3-ethernet"
-    assert (
-        connection_payload["device"]["match-device"].value
-        == "mac:AA:BB:CC:DD:EE:FF,interface-name:eth0"
-    )
+    assert connection_payload["match"]["path"].value == ["platform-ff3f0000.ethernet"]
 
     assert connection_payload["ipv4"]["method"].value == "auto"
     assert "address-data" not in connection_payload["ipv4"]
 
     assert connection_payload["ipv6"]["method"].value == "auto"
     assert "address-data" not in connection_payload["ipv6"]
+
+
+async def test_get_connection_no_path(network_manager: NetworkManager):
+    """Test network interface without a path."""
+    dbus_interface = network_manager.get(TEST_INTERFACE)
+    with patch.object(NetworkInterface, "path", new=PropertyMock(return_value=None)):
+        interface = Interface.from_dbus_interface(dbus_interface)
+
+    connection_payload = get_connection_from_interface(interface)
+
+    assert "connection" in connection_payload
+    assert "match" not in connection_payload
+
+    assert connection_payload["connection"]["interface-name"].value == "eth0"

tests/dbus/network/setting/test_init.py

@@ -58,9 +58,7 @@ async def test_update(
     )
     assert settings["connection"]["autoconnect"] == Variant("b", True)
 
-    assert settings["device"] == {
-        "match-device": Variant("s", "mac:AA:BB:CC:DD:EE:FF,interface-name:eth0")
-    }
+    assert settings["match"] == {"path": Variant("as", ["platform-ff3f0000.ethernet"])}
 
     assert "ipv4" in settings
     assert settings["ipv4"]["method"] == Variant("s", "auto")

tests/dbus_service_mocks/network_connection_settings.py

@@ -77,9 +77,7 @@ SETINGS_FIXTURES: dict[str, dict[str, dict[str, Variant]]] = {
         "proxy": {},
         "802-3-ethernet": SETTINGS_FIXTURE["802-3-ethernet"],
         "802-11-wireless": SETTINGS_FIXTURE["802-11-wireless"],
-        "device": {
-            "match-device": Variant("s", "mac:AA:BB:CC:DD:EE:FF,interface-name:eth0"),
-        },
+        "match": {"path": Variant("as", ["platform-ff3f0000.ethernet"])},
     },
 }
 

tests/docker/test_addon.py

@@ -194,7 +194,7 @@ async def test_addon_run_docker_error(
         coresys, addonsdata_system, "basic-addon-config.json"
     )
 
-    with patch.object(DockerAddon, "_stop"), patch.object(
+    with patch.object(DockerAddon, "stop"), patch.object(
         AddonOptions, "validate", new=PropertyMock(return_value=lambda _: None)
     ), pytest.raises(DockerNotFound):
         await docker_addon.run()
@@ -218,7 +218,7 @@ async def test_addon_run_add_host_error(
         coresys, addonsdata_system, "basic-addon-config.json"
     )
 
-    with patch.object(DockerAddon, "_stop"), patch.object(
+    with patch.object(DockerAddon, "stop"), patch.object(
         AddonOptions, "validate", new=PropertyMock(return_value=lambda _: None)
     ), patch.object(PluginDns, "add_host", side_effect=(err := CoreDNSError())):
         await docker_addon.run()

tests/homeassistant/test_core.py

@@ -92,7 +92,7 @@ async def test_install_docker_error(
 ):
     """Test install fails due to docker error."""
     coresys.security.force = True
-    with patch.object(HomeAssistantCore, "_start"), patch.object(
+    with patch.object(HomeAssistantCore, "start"), patch.object(
         DockerHomeAssistant, "cleanup"
     ), patch.object(
         Updater, "image_homeassistant", new=PropertyMock(return_value="homeassistant")
@@ -119,7 +119,7 @@ async def test_install_other_error(
     """Test install fails due to other error."""
     coresys.docker.images.pull.side_effect = [(err := OSError()), MagicMock()]
 
-    with patch.object(HomeAssistantCore, "_start"), patch.object(
+    with patch.object(HomeAssistantCore, "start"), patch.object(
         DockerHomeAssistant, "cleanup"
     ), patch.object(
         Updater, "image_homeassistant", new=PropertyMock(return_value="homeassistant")

tests/host/test_network.py

@@ -13,7 +13,6 @@ from supervisor.exceptions import HostNotSupportedError
 from supervisor.homeassistant.const import WSEvent, WSType
 from supervisor.host.const import WifiMode
 
-from tests.common import mock_dbus_services
 from tests.dbus_service_mocks.base import DBusServiceMock
 from tests.dbus_service_mocks.network_active_connection import (
     ActiveConnection as ActiveConnectionService,
@@ -253,47 +252,3 @@ async def test_host_connectivity_disabled(
         }
     )
     assert "connectivity_check" not in coresys.resolution.unsupported
-
-
-@pytest.mark.parametrize(
-    "interface_obj_path",
-    [
-        "/org/freedesktop/NetworkManager/Devices/4",
-        "/org/freedesktop/NetworkManager/Devices/5",
-    ],
-)
-async def test_load_with_mac_or_name_change(
-    coresys: CoreSys,
-    network_manager_service: NetworkManagerService,
-    interface_obj_path: str,
-):
-    """Test load fixes match-device settings if mac address or interface name has changed."""
-    await mock_dbus_services(
-        {
-            "network_active_connection": "/org/freedesktop/NetworkManager/ActiveConnection/2",
-            "network_connection_settings": "/org/freedesktop/NetworkManager/Settings/2",
-            "network_device": interface_obj_path,
-        },
-        coresys.dbus.bus,
-    )
-    await coresys.dbus.network.update({"Devices": [interface_obj_path]})
-
-    network_manager_service.ActivateConnection.calls.clear()
-    assert len(coresys.dbus.network.interfaces) == 1
-    interface = next(iter(coresys.dbus.network.interfaces))
-    assert interface.object_path == interface_obj_path
-    expected_match_device = (
-        f"mac:{interface.hw_address},interface-name:{interface.name}"
-    )
-    assert interface.settings.device.match_device != expected_match_device
-
-    await coresys.host.network.load()
-
-    assert network_manager_service.ActivateConnection.calls == [
-        (
-            "/org/freedesktop/NetworkManager/Settings/2",
-            interface_obj_path,
-            "/",
-        )
-    ]
-    assert interface.settings.device.match_device == expected_match_device

tests/jobs/test_job_decorator.py

@@ -3,6 +3,7 @@
 import asyncio
 from datetime import timedelta
 from unittest.mock import AsyncMock, Mock, PropertyMock, patch
+from uuid import uuid4
 
 from aiohttp.client_exceptions import ClientError
 import pytest
@@ -18,8 +19,10 @@ from supervisor.exceptions import (
 )
 from supervisor.host.const import HostFeature
 from supervisor.host.manager import HostManager
+from supervisor.jobs import SupervisorJob
 from supervisor.jobs.const import JobExecutionLimit
 from supervisor.jobs.decorator import Job, JobCondition
+from supervisor.jobs.job_group import JobGroup
 from supervisor.plugins.audio import PluginAudio
 from supervisor.resolution.const import UnhealthyReason
 from supervisor.utils.dt import utcnow
@@ -35,7 +38,7 @@ async def test_healthy(coresys: CoreSys, caplog: pytest.LogCaptureFixture):
             """Initialize the test class."""
             self.coresys = coresys
 
-        @Job(conditions=[JobCondition.HEALTHY])
+        @Job(name="test_healthy_execute", conditions=[JobCondition.HEALTHY])
         async def execute(self):
             """Execute the class method."""
             return True
@@ -77,12 +80,18 @@ async def test_internet(
             """Initialize the test class."""
             self.coresys = coresys
 
-        @Job(conditions=[JobCondition.INTERNET_HOST])
+        @Job(
+            name=f"test_internet_execute_host_{uuid4().hex}",
+            conditions=[JobCondition.INTERNET_HOST],
+        )
         async def execute_host(self):
             """Execute the class method."""
             return True
 
-        @Job(conditions=[JobCondition.INTERNET_SYSTEM])
+        @Job(
+            name=f"test_internet_execute_system_{uuid4().hex}",
+            conditions=[JobCondition.INTERNET_SYSTEM],
+        )
         async def execute_system(self):
             """Execute the class method."""
             return True
@@ -118,7 +127,7 @@ async def test_free_space(coresys: CoreSys):
             """Initialize the test class."""
             self.coresys = coresys
 
-        @Job(conditions=[JobCondition.FREE_SPACE])
+        @Job(name="test_free_space_execute", conditions=[JobCondition.FREE_SPACE])
         async def execute(self):
             """Execute the class method."""
             return True
@@ -144,7 +153,7 @@ async def test_haos(coresys: CoreSys):
             """Initialize the test class."""
             self.coresys = coresys
 
-        @Job(conditions=[JobCondition.HAOS])
+        @Job(name="test_haos_execute", conditions=[JobCondition.HAOS])
         async def execute(self):
             """Execute the class method."""
             return True
@@ -170,7 +179,7 @@ async def test_exception(coresys: CoreSys, capture_exception: Mock):
             """Initialize the test class."""
             self.coresys = coresys
 
-        @Job(conditions=[JobCondition.HEALTHY])
+        @Job(name="test_exception_execute", conditions=[JobCondition.HEALTHY])
         async def execute(self):
             """Execute the class method."""
             raise HassioError()
@@ -194,7 +203,9 @@ async def test_exception_not_handle(coresys: CoreSys, capture_exception: Mock):
             """Initialize the test class."""
             self.coresys = coresys
 
-        @Job(conditions=[JobCondition.HEALTHY])
+        @Job(
+            name="test_exception_not_handle_execute", conditions=[JobCondition.HEALTHY]
+        )
         async def execute(self):
             """Execute the class method."""
             raise err
@@ -217,7 +228,7 @@ async def test_running(coresys: CoreSys):
             """Initialize the test class."""
             self.coresys = coresys
 
-        @Job(conditions=[JobCondition.RUNNING])
+        @Job(name="test_running_execute", conditions=[JobCondition.RUNNING])
         async def execute(self):
             """Execute the class method."""
             return True
@@ -244,7 +255,11 @@ async def test_exception_conditions(coresys: CoreSys):
             """Initialize the test class."""
             self.coresys = coresys
 
-        @Job(conditions=[JobCondition.RUNNING], on_condition=HassioError)
+        @Job(
+            name="test_exception_conditions_execute",
+            conditions=[JobCondition.RUNNING],
+            on_condition=HassioError,
+        )
         async def execute(self):
             """Execute the class method."""
             return True
@@ -272,7 +287,10 @@ async def test_execution_limit_single_wait(
             self.coresys = coresys
             self.run = asyncio.Lock()
 
-        @Job(limit=JobExecutionLimit.SINGLE_WAIT)
+        @Job(
+            name="test_execution_limit_single_wait_execute",
+            limit=JobExecutionLimit.SINGLE_WAIT,
+        )
         async def execute(self, sleep: float):
             """Execute the class method."""
             assert not self.run.locked()
@@ -298,7 +316,11 @@ async def test_execution_limit_throttle_wait(
             self.run = asyncio.Lock()
             self.call = 0
 
-        @Job(limit=JobExecutionLimit.THROTTLE_WAIT, throttle_period=timedelta(hours=1))
+        @Job(
+            name="test_execution_limit_throttle_wait_execute",
+            limit=JobExecutionLimit.THROTTLE_WAIT,
+            throttle_period=timedelta(hours=1),
+        )
         async def execute(self, sleep: float):
             """Execute the class method."""
             assert not self.run.locked()
@@ -331,6 +353,7 @@ async def test_execution_limit_throttle_rate_limit(
             self.call = 0
 
         @Job(
+            name=f"test_execution_limit_throttle_rate_limit_execute_{uuid4().hex}",
             limit=JobExecutionLimit.THROTTLE_RATE_LIMIT,
             throttle_period=timedelta(hours=1),
             throttle_max_calls=2,
@@ -368,7 +391,11 @@ async def test_execution_limit_throttle(coresys: CoreSys, loop: asyncio.BaseEven
             self.run = asyncio.Lock()
             self.call = 0
 
-        @Job(limit=JobExecutionLimit.THROTTLE, throttle_period=timedelta(hours=1))
+        @Job(
+            name="test_execution_limit_throttle_execute",
+            limit=JobExecutionLimit.THROTTLE,
+            throttle_period=timedelta(hours=1),
+        )
         async def execute(self, sleep: float):
             """Execute the class method."""
             assert not self.run.locked()
@@ -396,7 +423,11 @@ async def test_execution_limit_once(coresys: CoreSys, loop: asyncio.BaseEventLoo
             self.coresys = coresys
             self.run = asyncio.Lock()
 
-        @Job(limit=JobExecutionLimit.ONCE, on_condition=JobException)
+        @Job(
+            name="test_execution_limit_once_execute",
+            limit=JobExecutionLimit.ONCE,
+            on_condition=JobException,
+        )
         async def execute(self, sleep: float):
             """Execute the class method."""
             assert not self.run.locked()
@@ -423,7 +454,10 @@ async def test_supervisor_updated(coresys: CoreSys):
             """Initialize the test class."""
             self.coresys = coresys
 
-        @Job(conditions=[JobCondition.SUPERVISOR_UPDATED])
+        @Job(
+            name="test_supervisor_updated_execute",
+            conditions=[JobCondition.SUPERVISOR_UPDATED],
+        )
         async def execute(self) -> bool:
             """Execute the class method."""
             return True
@@ -451,7 +485,10 @@ async def test_plugins_updated(coresys: CoreSys):
             """Initialize the test class."""
             self.coresys = coresys
 
-        @Job(conditions=[JobCondition.PLUGINS_UPDATED])
+        @Job(
+            name="test_plugins_updated_execute",
+            conditions=[JobCondition.PLUGINS_UPDATED],
+        )
         async def execute(self) -> bool:
             """Execute the class method."""
             return True
@@ -484,7 +521,7 @@ async def test_auto_update(coresys: CoreSys):
             """Initialize the test class."""
             self.coresys = coresys
 
-        @Job(conditions=[JobCondition.AUTO_UPDATE])
+        @Job(name="test_auto_update_execute", conditions=[JobCondition.AUTO_UPDATE])
         async def execute(self) -> bool:
             """Execute the class method."""
             return True
@@ -510,7 +547,7 @@ async def test_os_agent(coresys: CoreSys):
             """Initialize the test class."""
             self.coresys = coresys
 
-        @Job(conditions=[JobCondition.OS_AGENT])
+        @Job(name="test_os_agent_execute", conditions=[JobCondition.OS_AGENT])
         async def execute(self) -> bool:
             """Execute the class method."""
             return True
@@ -539,7 +576,7 @@ async def test_host_network(coresys: CoreSys):
             """Initialize the test class."""
             self.coresys = coresys
 
-        @Job(conditions=[JobCondition.HOST_NETWORK])
+        @Job(name="test_host_network_execute", conditions=[JobCondition.HOST_NETWORK])
         async def execute(self) -> bool:
             """Execute the class method."""
             return True
@@ -552,3 +589,354 @@ async def test_host_network(coresys: CoreSys):
 
     coresys.jobs.ignore_conditions = [JobCondition.HOST_NETWORK]
     assert await test.execute()
+
+
+async def test_job_group_once(coresys: CoreSys, loop: asyncio.BaseEventLoop):
+    """Test job group once execution limitation."""
+
+    class TestClass(JobGroup):
+        """Test class."""
+
+        def __init__(self, coresys: CoreSys):
+            """Initialize the test class."""
+            super().__init__(coresys, "TestClass")
+            self.event = asyncio.Event()
+
+        @Job(
+            name="test_job_group_once_inner_execute",
+            limit=JobExecutionLimit.GROUP_ONCE,
+            on_condition=JobException,
+        )
+        async def inner_execute(self) -> bool:
+            """Inner class method called by execute, group level lock allows this."""
+            await self.event.wait()
+            return True
+
+        @Job(
+            name="test_job_group_once_execute",
+            limit=JobExecutionLimit.GROUP_ONCE,
+            on_condition=JobException,
+        )
+        async def execute(self) -> bool:
+            """Execute the class method."""
+            return await self.inner_execute()
+
+        @Job(
+            name="test_job_group_once_separate_execute",
+            limit=JobExecutionLimit.GROUP_ONCE,
+            on_condition=JobException,
+        )
+        async def separate_execute(self) -> bool:
+            """Alternate execute method that shares group lock."""
+            return True
+
+        @Job(
+            name="test_job_group_once_unrelated",
+            limit=JobExecutionLimit.ONCE,
+            on_condition=JobException,
+        )
+        async def unrelated_method(self) -> bool:
+            """Unrelated method, sparate job with separate lock."""
+            return True
+
+    test = TestClass(coresys)
+    run_task = loop.create_task(test.execute())
+    await asyncio.sleep(0)
+
+    # All methods with group limits should be locked
+    with pytest.raises(JobException):
+        await test.execute()
+
+    with pytest.raises(JobException):
+        await test.inner_execute()
+
+    with pytest.raises(JobException):
+        await test.separate_execute()
+
+    # The once method is still callable
+    assert await test.unrelated_method()
+
+    test.event.set()
+    assert await run_task
+
+
+async def test_job_group_wait(coresys: CoreSys, loop: asyncio.BaseEventLoop):
+    """Test job group wait execution limitation."""
+
+    class TestClass(JobGroup):
+        """Test class."""
+
+        def __init__(self, coresys: CoreSys):
+            """Initialize the test class."""
+            super().__init__(coresys, "TestClass")
+            self.execute_count = 0
+            self.other_count = 0
+            self.event = asyncio.Event()
+
+        @Job(
+            name="test_job_group_wait_inner_execute",
+            limit=JobExecutionLimit.GROUP_WAIT,
+            on_condition=JobException,
+        )
+        async def inner_execute(self) -> None:
+            """Inner class method called by execute, group level lock allows this."""
+            self.execute_count += 1
+            await self.event.wait()
+
+        @Job(
+            name="test_job_group_wait_execute",
+            limit=JobExecutionLimit.GROUP_WAIT,
+            on_condition=JobException,
+        )
+        async def execute(self) -> None:
+            """Execute the class method."""
+            await self.inner_execute()
+
+        @Job(
+            name="test_job_group_wait_separate_execute",
+            limit=JobExecutionLimit.GROUP_WAIT,
+            on_condition=JobException,
+        )
+        async def separate_execute(self) -> None:
+            """Alternate execute method that shares group lock."""
+            self.other_count += 1
+
+    test = TestClass(coresys)
+    run_task = loop.create_task(test.execute())
+    await asyncio.sleep(0)
+
+    repeat_task = loop.create_task(test.execute())
+    other_task = loop.create_task(test.separate_execute())
+    await asyncio.sleep(0)
+
+    assert test.execute_count == 1
+    assert test.other_count == 0
+
+    test.event.set()
+    await run_task
+    await repeat_task
+    await other_task
+
+    assert test.execute_count == 2
+    assert test.other_count == 1
+
+
+async def test_job_cleanup(coresys: CoreSys, loop: asyncio.BaseEventLoop):
+    """Test job is cleaned up."""
+
+    class TestClass:
+        """Test class."""
+
+        def __init__(self, coresys: CoreSys):
+            """Initialize the test class."""
+            self.coresys = coresys
+            self.event = asyncio.Event()
+            self.job: SupervisorJob | None = None
+
+        @Job(name="test_job_cleanup_execute", limit=JobExecutionLimit.ONCE)
+        async def execute(self):
+            """Execute the class method."""
+            self.job = coresys.jobs.get_job()
+            await self.event.wait()
+            return True
+
+    test = TestClass(coresys)
+    run_task = loop.create_task(test.execute())
+    await asyncio.sleep(0)
+
+    assert coresys.jobs.jobs == [test.job]
+    assert not test.job.done
+
+    test.event.set()
+    assert await run_task
+
+    assert coresys.jobs.jobs == []
+    assert test.job.done
+
+
+async def test_job_skip_cleanup(coresys: CoreSys, loop: asyncio.BaseEventLoop):
+    """Test job is left in job manager when cleanup is false."""
+
+    class TestClass:
+        """Test class."""
+
+        def __init__(self, coresys: CoreSys):
+            """Initialize the test class."""
+            self.coresys = coresys
+            self.event = asyncio.Event()
+            self.job: SupervisorJob | None = None
+
+        @Job(
+            name="test_job_skip_cleanup_execute",
+            limit=JobExecutionLimit.ONCE,
+            cleanup=False,
+        )
+        async def execute(self):
+            """Execute the class method."""
+            self.job = coresys.jobs.get_job()
+            await self.event.wait()
+            return True
+
+    test = TestClass(coresys)
+    run_task = loop.create_task(test.execute())
+    await asyncio.sleep(0)
+
+    assert coresys.jobs.jobs == [test.job]
+    assert not test.job.done
+
+    test.event.set()
+    assert await run_task
+
+    assert coresys.jobs.jobs == [test.job]
+    assert test.job.done
+
+
+async def test_execution_limit_group_throttle(
+    coresys: CoreSys, loop: asyncio.BaseEventLoop
+):
+    """Test the group throttle execution limit."""
+
+    class TestClass(JobGroup):
+        """Test class."""
+
+        def __init__(self, coresys: CoreSys, reference: str):
+            """Initialize the test class."""
+            super().__init__(coresys, f"test_class_{reference}", reference)
+            self.run = asyncio.Lock()
+            self.call = 0
+
+        @Job(
+            name="test_execution_limit_group_throttle_execute",
+            limit=JobExecutionLimit.GROUP_THROTTLE,
+            throttle_period=timedelta(milliseconds=95),
+        )
+        async def execute(self, sleep: float):
+            """Execute the class method."""
+            assert not self.run.locked()
+            async with self.run:
+                await asyncio.sleep(sleep)
+            self.call += 1
+
+    test1 = TestClass(coresys, "test1")
+    test2 = TestClass(coresys, "test2")
+
+    # One call of each should work. The subsequent calls will be silently throttled due to period
+    await asyncio.gather(
+        test1.execute(0), test1.execute(0), test2.execute(0), test2.execute(0)
+    )
+    assert test1.call == 1
+    assert test2.call == 1
+
+    # First call to each will work again since period cleared. Second throttled once more as they don't wait
+    with time_machine.travel(utcnow() + timedelta(milliseconds=100)):
+        await asyncio.gather(
+            test1.execute(0.1),
+            test1.execute(0.1),
+            test2.execute(0.1),
+            test2.execute(0.1),
+        )
+
+    assert test1.call == 2
+    assert test2.call == 2
+
+
+async def test_execution_limit_group_throttle_wait(
+    coresys: CoreSys, loop: asyncio.BaseEventLoop
+):
+    """Test the group throttle wait job execution limit."""
+
+    class TestClass(JobGroup):
+        """Test class."""
+
+        def __init__(self, coresys: CoreSys, reference: str):
+            """Initialize the test class."""
+            super().__init__(coresys, f"test_class_{reference}", reference)
+            self.run = asyncio.Lock()
+            self.call = 0
+
+        @Job(
+            name="test_execution_limit_group_throttle_wait_execute",
+            limit=JobExecutionLimit.GROUP_THROTTLE_WAIT,
+            throttle_period=timedelta(milliseconds=95),
+        )
+        async def execute(self, sleep: float):
+            """Execute the class method."""
+            assert not self.run.locked()
+            async with self.run:
+                await asyncio.sleep(sleep)
+            self.call += 1
+
+    test1 = TestClass(coresys, "test1")
+    test2 = TestClass(coresys, "test2")
+
+    # One call of each should work. The subsequent calls will be silently throttled after waiting due to period
+    await asyncio.gather(
+        *[test1.execute(0), test1.execute(0), test2.execute(0), test2.execute(0)]
+    )
+    assert test1.call == 1
+    assert test2.call == 1
+
+    # All calls should work as we cleared the period. And tasks take longer then period and are queued
+    with time_machine.travel(utcnow() + timedelta(milliseconds=100)):
+        await asyncio.gather(
+            *[
+                test1.execute(0.1),
+                test1.execute(0.1),
+                test2.execute(0.1),
+                test2.execute(0.1),
+            ]
+        )
+
+    assert test1.call == 3
+    assert test2.call == 3
+
+
+@pytest.mark.parametrize("error", [None, PluginJobError])
+async def test_execution_limit_group_throttle_rate_limit(
+    coresys: CoreSys, loop: asyncio.BaseEventLoop, error: JobException | None
+):
+    """Test the group throttle rate limit job execution limit."""
+
+    class TestClass(JobGroup):
+        """Test class."""
+
+        def __init__(self, coresys: CoreSys, reference: str):
+            """Initialize the test class."""
+            super().__init__(coresys, f"test_class_{reference}", reference)
+            self.run = asyncio.Lock()
+            self.call = 0
+
+        @Job(
+            name=f"test_execution_limit_group_throttle_rate_limit_execute_{uuid4().hex}",
+            limit=JobExecutionLimit.GROUP_THROTTLE_RATE_LIMIT,
+            throttle_period=timedelta(hours=1),
+            throttle_max_calls=2,
+            on_condition=error,
+        )
+        async def execute(self):
+            """Execute the class method."""
+            self.call += 1
+
+    test1 = TestClass(coresys, "test1")
+    test2 = TestClass(coresys, "test2")
+
+    await asyncio.gather(
+        *[test1.execute(), test1.execute(), test2.execute(), test2.execute()]
+    )
+    assert test1.call == 2
+    assert test2.call == 2
+
+    with pytest.raises(JobException if error is None else error):
+        await test1.execute()
+    with pytest.raises(JobException if error is None else error):
+        await test2.execute()
+
+    assert test1.call == 2
+    assert test2.call == 2
+
+    with time_machine.travel(utcnow() + timedelta(hours=1)):
+        await test1.execute()
+        await test2.execute()
+
+    assert test1.call == 3
+    assert test2.call == 3

tests/jobs/test_job_manager.py

@@ -1,39 +1,76 @@
 """Test the condition decorators."""
+
+import pytest
+
 # pylint: disable=protected-access,import-error
 from supervisor.coresys import CoreSys
+from supervisor.exceptions import JobStartException
 
 TEST_JOB = "test"
 
 
 async def test_add_job(coresys: CoreSys):
     """Test adding jobs."""
-    job = coresys.jobs.get_job(TEST_JOB)
+    job = coresys.jobs.new_job(TEST_JOB)
 
-    assert job.name in coresys.jobs.jobs
+    assert job in coresys.jobs.jobs
 
 
-async def test_remove_job_directly(coresys: CoreSys):
+async def test_remove_job_directly(coresys: CoreSys, caplog: pytest.LogCaptureFixture):
     """Test removing jobs  from manager."""
-    job = coresys.jobs.get_job(TEST_JOB)
+    job = coresys.jobs.new_job(TEST_JOB)
+
+    assert job in coresys.jobs.jobs
 
-    assert job.name in coresys.jobs.jobs
     coresys.jobs.remove_job(job)
-    assert job.name not in coresys.jobs.jobs
+    assert job not in coresys.jobs.jobs
+    assert f"Removing incomplete job {job.name}" in caplog.text
 
 
-async def test_remove_job_with_progress(coresys: CoreSys):
-    """Test removing jobs by setting progress to 100."""
-    job = coresys.jobs.get_job(TEST_JOB)
+async def test_job_done(coresys: CoreSys):
+    """Test done set correctly with jobs."""
+    job = coresys.jobs.new_job(TEST_JOB)
+    assert not job.done
+    assert coresys.jobs.get_job() != job
 
-    assert job.name in coresys.jobs.jobs
-    job.update(progress=100)
-    assert job.name not in coresys.jobs.jobs
+    with job.start():
+        assert coresys.jobs.get_job() == job
+        assert not job.done
+
+    assert coresys.jobs.get_job() != job
+    assert job.done
+
+    with pytest.raises(JobStartException):
+        with job.start():
+            pass
+
+
+async def test_job_start_bad_parent(coresys: CoreSys):
+    """Test job cannot be started outside of parent."""
+    job = coresys.jobs.new_job(TEST_JOB)
+    job2 = coresys.jobs.new_job(f"{TEST_JOB}_2")
+
+    with job.start():
+        with pytest.raises(JobStartException):
+            with job2.start():
+                pass
+
+    with job2.start():
+        assert coresys.jobs.get_job() == job2
 
 
 async def test_update_job(coresys: CoreSys):
     """Test updating jobs."""
-    job = coresys.jobs.get_job(TEST_JOB)
+    job = coresys.jobs.new_job(TEST_JOB)
 
-    job.update(progress=50, stage="stage")
+    job.progress = 50
     assert job.progress == 50
+
+    job.stage = "stage"
     assert job.stage == "stage"
+
+    with pytest.raises(ValueError):
+        job.progress = 110
+
+    with pytest.raises(ValueError):
+        job.progress = -10

tests/plugins/test_audio.py

@@ -6,8 +6,16 @@ import pytest
 
 from supervisor.const import LogLevel
 from supervisor.coresys import CoreSys
+from supervisor.docker.audio import DockerAudio
 
-from tests.plugins.test_dns import fixture_docker_interface  # noqa: F401
+
+@pytest.fixture(name="docker_interface")
+async def fixture_docker_interface() -> tuple[AsyncMock, AsyncMock]:
+    """Mock docker interface methods."""
+    with patch.object(DockerAudio, "run") as run, patch.object(
+        DockerAudio, "restart"
+    ) as restart:
+        yield (run, restart)
 
 
 @pytest.fixture(name="write_json")

tests/plugins/test_dns.py

@@ -9,7 +9,7 @@ import pytest
 from supervisor.const import BusEvent, LogLevel
 from supervisor.coresys import CoreSys
 from supervisor.docker.const import ContainerState
-from supervisor.docker.interface import DockerInterface
+from supervisor.docker.dns import DockerDNS
 from supervisor.docker.monitor import DockerContainerStateEvent
 from supervisor.plugins.dns import HostEntry
 from supervisor.resolution.const import ContextType, IssueType, SuggestionType
@@ -19,8 +19,8 @@ from supervisor.resolution.data import Issue, Suggestion
 @pytest.fixture(name="docker_interface")
 async def fixture_docker_interface() -> tuple[AsyncMock, AsyncMock]:
     """Mock docker interface methods."""
-    with patch.object(DockerInterface, "run") as run, patch.object(
-        DockerInterface, "restart"
+    with patch.object(DockerDNS, "run") as run, patch.object(
+        DockerDNS, "restart"
     ) as restart:
         yield (run, restart)
 

tests/store/test_reading_addons.py

@@ -5,7 +5,7 @@ from unittest.mock import patch
 from supervisor.coresys import CoreSys
 
 
-def test_read_addon_files(coresys: CoreSys):
+async def test_read_addon_files(coresys: CoreSys):
     """Test that we are reading add-on files correctly."""
     with patch(
         "pathlib.Path.glob",
@@ -19,7 +19,7 @@ def test_read_addon_files(coresys: CoreSys):
             Path(".circleci/config.yml"),
         ],
     ):
-        addon_list = coresys.store.data._find_addons(Path("test"), {})
+        addon_list = await coresys.store.data._find_addons(Path("test"), {})
 
         assert len(addon_list) == 1
         assert str(addon_list[0]) == "addon/config.yml"

tests/store/test_translation_load.py

@@ -1,16 +1,20 @@
 """Test loading add-translation."""
 # pylint: disable=import-error,protected-access
 import os
+from pathlib import Path
+
+import pytest
 
 from supervisor.coresys import CoreSys
+from supervisor.store.data import _read_addon_translations
 from supervisor.utils.common import write_json_or_yaml_file
 
 
-def test_loading_traslations(coresys: CoreSys, tmp_path):
+def test_loading_traslations(coresys: CoreSys, tmp_path: Path):
     """Test loading add-translation."""
     os.makedirs(tmp_path / "translations")
     # no transaltions
-    assert coresys.store.data._read_addon_translations(tmp_path) == {}
+    assert _read_addon_translations(tmp_path) == {}
 
     for file in ("en.json", "es.json"):
         write_json_or_yaml_file(
@@ -27,7 +31,7 @@ def test_loading_traslations(coresys: CoreSys, tmp_path):
             },
         )
 
-    translations = coresys.store.data._read_addon_translations(tmp_path)
+    translations = _read_addon_translations(tmp_path)
 
     assert translations["en"]["configuration"]["test"]["name"] == "test"
     assert translations["es"]["configuration"]["test"]["name"] == "test"
@@ -37,3 +41,22 @@ def test_loading_traslations(coresys: CoreSys, tmp_path):
     assert "test" not in translations["en"]["configuration"]["test"]
 
     assert translations["no"]["network"]["80/tcp"] == "Webserver port"
+
+
+def test_translation_file_failure(
+    coresys: CoreSys, tmp_path: Path, caplog: pytest.LogCaptureFixture
+):
+    """Test translations load if one fails."""
+    os.makedirs(tmp_path / "translations")
+    write_json_or_yaml_file(
+        tmp_path / "translations" / "en.json",
+        {"configuration": {"test": {"name": "test", "test": "test"}}},
+    )
+    fail_path = tmp_path / "translations" / "de.json"
+    with fail_path.open("w") as de_file:
+        de_file.write("not json")
+
+    translations = _read_addon_translations(tmp_path)
+
+    assert translations["en"]["configuration"]["test"]["name"] == "test"
+    assert f"Can't read translations from {fail_path.as_posix()}" in caplog.text

tests/test_arch.py

@@ -103,6 +103,26 @@ async def test_raspberrypi4_64_arch(coresys, sys_machine, sys_supervisor):
     assert coresys.arch.supported == ["aarch64", "armv7", "armhf"]
 
 
+async def test_yellow_arch(coresys, sys_machine, sys_supervisor):
+    """Test arch for yellow."""
+    sys_machine.return_value = "yellow"
+    sys_supervisor.arch = "aarch64"
+    await coresys.arch.load()
+
+    assert coresys.arch.default == "aarch64"
+    assert coresys.arch.supported == ["aarch64", "armv7", "armhf"]
+
+
+async def test_green_arch(coresys, sys_machine, sys_supervisor):
+    """Test arch for green."""
+    sys_machine.return_value = "green"
+    sys_supervisor.arch = "aarch64"
+    await coresys.arch.load()
+
+    assert coresys.arch.default == "aarch64"
+    assert coresys.arch.supported == ["aarch64", "armv7", "armhf"]
+
+
 async def test_tinker_arch(coresys, sys_machine, sys_supervisor):
     """Test arch for tinker."""
     sys_machine.return_value = "tinker"

tests/test_ingress.py

@@ -1,6 +1,7 @@
 """Test ingress."""
 from datetime import timedelta
 
+from supervisor.const import ATTR_SESSION_DATA_USER_ID
 from supervisor.utils.dt import utc_from_timestamp
 
 
@@ -20,6 +21,21 @@ def test_session_handling(coresys):
     assert not coresys.ingress.validate_session(session)
     assert not coresys.ingress.validate_session("invalid session")
 
+    session_data = coresys.ingress.get_session_data(session)
+    assert session_data is None
+
+
+def test_session_handling_with_session_data(coresys):
+    """Create and test session."""
+    session = coresys.ingress.create_session(
+        dict([(ATTR_SESSION_DATA_USER_ID, "some-id")])
+    )
+
+    assert session
+
+    session_data = coresys.ingress.get_session_data(session)
+    assert session_data[ATTR_SESSION_DATA_USER_ID] == "some-id"
+
 
 async def test_save_on_unload(coresys):
     """Test called save on unload."""

tests/test_validate.py

@@ -16,6 +16,24 @@ DNS_GOOD_V6 = [
     "DNS://2606:4700:4700::1001",  # cloudflare
 ]
 DNS_BAD = ["hello world", "https://foo.bar", "", "dns://example.com"]
+IMAGE_NAME_GOOD = [
+    "ghcr.io/home-assistant/{machine}-homeassistant",
+    "ghcr.io/home-assistant/{arch}-homeassistant",
+    "homeassistant/{arch}-homeassistant",
+    "doocker.io/homeassistant/{arch}-homeassistant",
+    "ghcr.io/home-assistant/amd64-homeassistant",
+    "homeassistant/amd64-homeassistant",
+    "ttl.sh/homeassistant",
+    "myreg.local:8080/homeassistant",
+]
+IMAGE_NAME_BAD = [
+    "ghcr.io/home-assistant/homeassistant:123",
+    ".ghcr.io/home-assistant/homeassistant",
+    "HOMEASSISTANT/homeassistant",
+    "homeassistant/HOMEASSISTANT",
+    "homeassistant/_homeassistant",
+    "homeassistant/-homeassistant",
+]
 
 
 async def test_dns_url_v4_good():
@@ -72,6 +90,19 @@ def test_dns_server_list_bad_combined():
         assert validate.dns_server_list(combined)
 
 
+def test_image_name_good():
+    """Test container image names validator with known-good image names."""
+    for image_name in IMAGE_NAME_GOOD:
+        assert validate.docker_image(image_name)
+
+
+def test_image_name_bad():
+    """Test container image names validator with known-bad image names."""
+    for image_name in IMAGE_NAME_BAD:
+        with pytest.raises(vol.error.Invalid):
+            assert validate.docker_image(image_name)
+
+
 def test_version_complex():
     """Test version simple with good version."""
     for version in (