Add exponential backoff strategy to watchdog restart (#6262)

Bumps [actions/upload-artifact](https://github.com/actions/upload-artifact) from 4.6.2 to 5.0.0.
- [Release notes](https://github.com/actions/upload-artifact/releases)
- [Commits](ea165f8d65...330a01c490)

---
updated-dependencies:
- dependency-name: actions/upload-artifact
  dependency-version: 5.0.0
  dependency-type: direct:production
  update-type: version-update:semver-major
...

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

.github/workflows/ci.yaml

@@ -386,7 +386,7 @@ jobs:
             -o console_output_style=count \
             tests
       - name: Upload coverage artifact
-        uses: actions/upload-artifact@ea165f8d65b6e75b540449e92b4886f43607fa02 # v4.6.2
+        uses: actions/upload-artifact@330a01c490aca151604b8cf639adc76d48f6c5d4 # v5.0.0
         with:
           name: coverage
           path: .coverage
@@ -417,7 +417,7 @@ jobs:
           echo "Failed to restore Python virtual environment from cache"
           exit 1
       - name: Download all coverage artifacts
-        uses: actions/download-artifact@634f93cb2916e3fdff6788551b99b062d0335ce0 # v5.0.0
+        uses: actions/download-artifact@018cc2cf5baa6db3ef3c5f8a56943fffe632ef53 # v6.0.0
         with:
           name: coverage
           path: coverage/

requirements.txt

@@ -17,7 +17,7 @@ faust-cchardet==2.1.19
 gitpython==3.1.45
 jinja2==3.1.6
 log-rate-limit==1.4.2
-orjson==3.11.3
+orjson==3.11.4
 pulsectl==24.12.0
 pyudev==0.24.4
 PyYAML==6.0.3

requirements_tests.txt

@@ -8,7 +8,7 @@ pytest-asyncio==0.25.2
 pytest-cov==7.0.0
 pytest-timeout==2.4.0
 pytest==8.4.2
-ruff==0.14.1
+ruff==0.14.2
 time-machine==2.19.0
 types-docker==7.1.0.20251009
 types-pyyaml==6.0.12.20250915

supervisor/addons/addon.py

@@ -1562,7 +1562,15 @@ class Addon(AddonModel):
                 )
                 break
 
-            await asyncio.sleep(WATCHDOG_RETRY_SECONDS)
+            # Exponential backoff to spread retries over the throttle window
+            delay = WATCHDOG_RETRY_SECONDS * (1 << max(attempts - 1, 0))
+            _LOGGER.debug(
+                "Watchdog will retry addon %s in %s seconds (attempt %s)",
+                self.name,
+                delay,
+                attempts + 1,
+            )
+            await asyncio.sleep(delay)
 
     async def container_state_changed(self, event: DockerContainerStateEvent) -> None:
         """Set addon state from container state."""

supervisor/docker/interface.py

@@ -306,6 +306,8 @@ class DockerInterface(JobGroup, ABC):
         # Our filters have all passed. Time to update the job
         # Only downloading and extracting have progress details. Use that to set extra
         # We'll leave it around on later stages as the total bytes may be useful after that stage
+        # Enforce range to prevent float drift error
+        progress = max(0, min(progress, 100))
         if (
             stage in {PullImageLayerStage.DOWNLOADING, PullImageLayerStage.EXTRACTING}
             and reference.progress_detail
@@ -371,7 +373,7 @@ class DockerInterface(JobGroup, ABC):
 
         # To reduce noise, limit updates to when result has changed by an entire percent or when stage changed
         if stage != install_job.stage or progress >= install_job.progress + 1:
-            install_job.update(stage=stage.status, progress=progress)
+            install_job.update(stage=stage.status, progress=max(0, min(progress, 100)))
 
     @Job(
         name="docker_interface_install",

supervisor/hardware/disk.py

@@ -9,7 +9,12 @@ from typing import Any
 from supervisor.resolution.const import UnhealthyReason
 
 from ..coresys import CoreSys, CoreSysAttributes
-from ..exceptions import DBusError, DBusObjectError, HardwareNotFound
+from ..exceptions import (
+    DBusError,
+    DBusNotConnectedError,
+    DBusObjectError,
+    HardwareNotFound,
+)
 from .const import UdevSubsystem
 from .data import Device
 
@@ -207,6 +212,8 @@ class HwDisk(CoreSysAttributes):
         try:
             block_device = self.sys_dbus.udisks2.get_block_device_by_path(device_path)
             drive = self.sys_dbus.udisks2.get_drive(block_device.drive)
+        except DBusNotConnectedError:
+            return None
         except DBusObjectError:
             _LOGGER.warning(
                 "Unable to find UDisks2 drive for device at %s", device_path.as_posix()

tests/hardware/test_disk.py

@@ -376,3 +376,14 @@ async def test_try_get_nvme_life_time_missing_percent_used(
         coresys.config.path_supervisor
     )
     assert lifetime is None
+
+
+async def test_try_get_nvme_life_time_dbus_not_connected(coresys: CoreSys):
+    """Test getting lifetime info from an NVMe when DBUS is not connected."""
+    # Set the dbus for udisks2 bus to be None, to make it forcibly disconnected.
+    coresys.dbus.udisks2.dbus = None
+
+    lifetime = await coresys.hardware.disk.get_disk_life_time(
+        coresys.config.path_supervisor
+    )
+    assert lifetime is None