From 55a8f61e304f2c37bbf4e70135af03713df4527c Mon Sep 17 00:00:00 2001 From: Isteb4k Date: Wed, 10 Jun 2026 13:53:56 +0000 Subject: [PATCH] Re-generate changelog v1.9.0 Signed-off-by: deckhouse-BOaTswain <89150800+deckhouse-boatswain@users.noreply.github.com> --- CHANGELOG/CHANGELOG-v1.9.0.yml | 229 +++++++++++++++++++++++++++++++++ CHANGELOG/CHANGELOG-v1.9.md | 35 +++++ 2 files changed, 264 insertions(+) create mode 100644 CHANGELOG/CHANGELOG-v1.9.0.yml create mode 100644 CHANGELOG/CHANGELOG-v1.9.md diff --git a/CHANGELOG/CHANGELOG-v1.9.0.yml b/CHANGELOG/CHANGELOG-v1.9.0.yml new file mode 100644 index 0000000000..37751ffa51 --- /dev/null +++ b/CHANGELOG/CHANGELOG-v1.9.0.yml @@ -0,0 +1,229 @@ +ci: + features: + - summary: Add Python-generated E2E duration charts and slowest-specs workflow artifacts. + pull_request: https://github.com/deckhouse/virtualization/pull/2361 + - summary: Add step to switch modules to maintenance mode before e2e tests. + pull_request: https://github.com/deckhouse/virtualization/pull/2321 + - summary: >- + Deploy Prod workflow can prepare GitHub releases from changelog PRs and report release + status to Loop. + pull_request: https://github.com/deckhouse/virtualization/pull/2291 + - summary: test rolling over release to release + pull_request: https://github.com/deckhouse/virtualization/pull/2195 + fixes: + - summary: The SDN readiness timeout during nested e2e bootstrap has been increased. + pull_request: https://github.com/deckhouse/virtualization/pull/2413 + - summary: >- + Increase release site validation wait time to reduce CI flakes while module versions + propagate. + pull_request: https://github.com/deckhouse/virtualization/pull/2386 + - summary: Release-channel validation now checks module versions on the module-specific release page. + pull_request: https://github.com/deckhouse/virtualization/pull/2381 + - summary: Fix E2E workflow artifact upload and download handling for encrypted generated files. + pull_request: https://github.com/deckhouse/virtualization/pull/2364 + - summary: >- + Improve E2E messenger report failure reasons, suite setup failure reporting, and + success-rate calculation. + pull_request: https://github.com/deckhouse/virtualization/pull/2355 + - summary: >- + Harden nested e2e module setup against Deckhouse module propagation and admission readiness + races. + pull_request: https://github.com/deckhouse/virtualization/pull/2349 + - summary: use discovered SSH credentials for Alpine VMs in static cluster e2e infrastructure. + pull_request: https://github.com/deckhouse/virtualization/pull/2337 + - summary: Wait longer for nested e2e kubeconfig authorization to become ready. + pull_request: https://github.com/deckhouse/virtualization/pull/2336 + - summary: docker login step + pull_request: https://github.com/deckhouse/virtualization/pull/2325 + - summary: Retry applying Virtualization configuration in nested E2E workflow. + pull_request: https://github.com/deckhouse/virtualization/pull/2319 + - summary: Make nested E2E bootstrap diagnostics tolerate temporary Kubernetes API failures. + pull_request: https://github.com/deckhouse/virtualization/pull/2317 + - summary: >- + Enable source image tag-based werf imports in prod build workflows to avoid unnecessary + rebuilds caused by cache metadata mismatch between werf versions. + pull_request: https://github.com/deckhouse/virtualization/pull/2310 + - summary: >- + Skip USB nested e2e on Kubernetes versions below 1.34 and fall back to the default NFS + storage class. + pull_request: https://github.com/deckhouse/virtualization/pull/2307 + - summary: >- + Prevent E2E report parsing failures by separating the raw Ginkgo JSON report from the + summary report in CI artifacts. + pull_request: https://github.com/deckhouse/virtualization/pull/2305 + - summary: add SSH keepalive options to d8-based nested cluster access in CI and e2e helpers + pull_request: https://github.com/deckhouse/virtualization/pull/2301 + - summary: >- + Refactor nested E2E report generation into tested Node.js scripts with JSON reports and + improve Loop reporting for missing artifacts, skipped tests, and stage failures. + pull_request: https://github.com/deckhouse/virtualization/pull/2245 +core: + features: + - summary: Add live migration transfer details to VMOP condition messages. + pull_request: https://github.com/deckhouse/virtualization/pull/2436 + - summary: >- + CLI lifecycle commands (start, stop, restart, evict, migrate) now support multiple VM + targets, --all flag, --label-selector flag for label-based targeting, and --yes flag for + non-interactive confirmation. + pull_request: https://github.com/deckhouse/virtualization/pull/2412 + - summary: >- + Compatible VirtualMachineOperation resources can now supersede another active operation on + the same VM. + pull_request: https://github.com/deckhouse/virtualization/pull/2338 + - summary: >- + Added the `Uptime` column to `VirtualMachine` resources, showing the time since the VM + started. + pull_request: https://github.com/deckhouse/virtualization/pull/2279 + fixes: + - summary: >- + Fix USB startupPolicy='optional' in containers where sysfs exposes host devices but device + nodes are absent from the mount namespace. + pull_request: https://github.com/deckhouse/virtualization/pull/2432 + - summary: Create network interfaces with deckhouse user ownership (UID/GID 64535). + pull_request: https://github.com/deckhouse/virtualization/pull/2430 + - summary: Fix panic in kubevirt CRD event handler on tombstone object delivery. + pull_request: https://github.com/deckhouse/virtualization/pull/2420 + - summary: Fix cleanup of stale hotplug attachment pods after removing VM disks. + pull_request: https://github.com/deckhouse/virtualization/pull/2409 + - summary: Prevent hotplug volume mount leaks on failed migration cleanup by bumping KubeVirt. + pull_request: https://github.com/deckhouse/virtualization/pull/2385 + - summary: >- + Better handling Windows guests: start and migration should work in clusters with frequent + CPU frequencies drifts + pull_request: https://github.com/deckhouse/virtualization/pull/2345 + - summary: Preserve VirtualMachine uptime during migration by tracking the last start time in status. + pull_request: https://github.com/deckhouse/virtualization/pull/2304 + - summary: >- + Restore MigrationPending with 0% progress and treat target scheduling as in-progress for VM + migration operations. + pull_request: https://github.com/deckhouse/virtualization/pull/2272 + - summary: Propagate target migration pod unschedulable details to VMOP conditions. + pull_request: https://github.com/deckhouse/virtualization/pull/2261 + - summary: >- + Reconnect logic in d8 v vnc and d8 v console now refreshes client config before reconnect, + avoiding stale OIDC refresh token failures. + pull_request: https://github.com/deckhouse/virtualization/pull/2253 +dvcr: + fixes: + - summary: >- + DVCR garbage collection no longer fails on a fresh PVC-backed registry before the first + image upload. + pull_request: https://github.com/deckhouse/virtualization/pull/2447 +module: + fixes: + - summary: >- + Prevent the USB/IP module installation step from retrying indefinitely when kernel module + packages are unavailable. + pull_request: https://github.com/deckhouse/virtualization/pull/2371 + - summary: >- + Skip USBGateway detach cleanup for untracked USB devices and detach tracked devices only on + the last active consumer. + pull_request: https://github.com/deckhouse/virtualization/pull/2298 + - summary: >- + Fixed an issue where invalid `virtualization` module ModuleConfig settings could block the + Deckhouse queue. + pull_request: https://github.com/deckhouse/virtualization/pull/2246 + - summary: >- + Add idempotency for USB device cleanup in virtualization-dra to prevent errors on repeated + unprepare calls. + pull_request: https://github.com/deckhouse/virtualization/pull/2216 + - summary: Fixed duplicate series on the `Virtualization / Overview` dashboard. + pull_request: https://github.com/deckhouse/virtualization/pull/2189 +test: + features: + - summary: >- + Add precheck system for e2e tests that validates cluster configuration before test + execution. + pull_request: https://github.com/deckhouse/virtualization/pull/2269 + fixes: + - summary: Make e2e precheck label parsing compatible with Ginkgo negated label filters. + pull_request: https://github.com/deckhouse/virtualization/pull/2439 + - summary: VirtualDiskResizing e2e test now waits for SSH before checking disk size after resize. + pull_request: https://github.com/deckhouse/virtualization/pull/2424 + - summary: Simplify VirtualDiskResizing disk size checks to reduce SSH command flakiness. + pull_request: https://github.com/deckhouse/virtualization/pull/2405 + - summary: Increase CPU core fraction for hotplug e2e VMs to avoid SSH disk probe timeouts. + pull_request: https://github.com/deckhouse/virtualization/pull/2403 + - summary: >- + Failed e2e dump artifacts are now stored in e2e_failed//... for both new and + legacy tests. + pull_request: https://github.com/deckhouse/virtualization/pull/2383 + - summary: >- + Improve reliability of flaky USB reconnection e2e test by splitting mount logic into find, + format, and mount steps. + pull_request: https://github.com/deckhouse/virtualization/pull/2332 + - summary: Fix e2e precheck prepare for focus/label filters containing spaces. + pull_request: https://github.com/deckhouse/virtualization/pull/2318 + - summary: >- + Make the e2e external connectivity check rely on successful outbound access to fallback + hosts instead of HTTP 200 from a single site. + pull_request: https://github.com/deckhouse/virtualization/pull/2313 + - summary: >- + Stabilized the VirtualMachineUSB e2e test by improving guest USB disk detection, remount + flow, and timeouts around migration. + pull_request: https://github.com/deckhouse/virtualization/pull/2297 + - summary: Improve VirtualMachineAdditionalNetworkInterfaces e2e test reliability and diagnostics. + pull_request: https://github.com/deckhouse/virtualization/pull/2290 + - summary: Increase timeout for repeated affinity migration in the legacy affinity/toleration e2e test. + pull_request: https://github.com/deckhouse/virtualization/pull/2267 + - summary: >- + Increase VirtualMachineSnapshot readiness timeout in VirtualMachineOperationRestore e2e test + to fix flakiness on NFS storage. + pull_request: https://github.com/deckhouse/virtualization/pull/2254 +vd: + fixes: + - summary: >- + Time spent in the `WaitForFirstConsumer` phase is no longer included in + `.status.stats.creationDuration.totalProvisioning` of virtual disks. + pull_request: https://github.com/deckhouse/virtualization/pull/2379 + - summary: Allow ingress from virtualization namespace to importer pods + pull_request: https://github.com/deckhouse/virtualization/pull/2356 + - summary: Restore requested VirtualDisk size from snapshot metadata. + pull_request: https://github.com/deckhouse/virtualization/pull/2278 +vm: + features: + - summary: >- + The VM status now includes a `No bootable device` message when the VM cannot find a bootable + disk to start. + pull_request: https://github.com/deckhouse/virtualization/pull/2404 + - summary: Add domain jobs and block-jobs info subcommands to vlctl. + pull_request: https://github.com/deckhouse/virtualization/pull/2280 + - summary: >- + Added the ability to change `coreFraction` on a running VM without a restart. The new value + is applied via live migration. + pull_request: https://github.com/deckhouse/virtualization/pull/2210 + - summary: >- + Added the ability to attach additional network interfaces without a restart via the virtual + machine's `.spec.networks`. + pull_request: https://github.com/deckhouse/virtualization/pull/2187 + - summary: Provide "no bootable device" event to virt-launcher. + pull_request: https://github.com/deckhouse/virtualization/pull/2183 + - summary: >- + System virtual machine resources (pods with `d8v-hp-` and `d8v-vm-` prefixes) now run as the + `deckhouse` user, without root privileges. + pull_request: https://github.com/deckhouse/virtualization/pull/2097 + - summary: >- + A restart is no longer required to attach and detach virtual disks and images via the + virtual machine's `.spec.blockDeviceRefs`: + + - Works for new virtual machines starting from v1.9.0. + + - For previously created virtual machines, a restart is required to enable this behavior. + pull_request: https://github.com/deckhouse/virtualization/pull/2033 + fixes: + - summary: resolve xml dir for non root setup + pull_request: https://github.com/deckhouse/virtualization/pull/2434 + - summary: >- + Fixed disk update for stopped VMs with WaitForFirstConsumer storage class. Previously, when + a VM was stopped and disks were changed, the KVVM was not updated, causing the VM to get + stuck in starting phase. + pull_request: https://github.com/deckhouse/virtualization/pull/2407 + - summary: >- + Fix VM migration failure caused by incorrect target disk size for filesystem-backed hotplug + volumes. + pull_request: https://github.com/deckhouse/virtualization/pull/2402 + - summary: Fix possible scheduling problems after changing vmclass from Discovery type to Model type. + pull_request: https://github.com/deckhouse/virtualization/pull/2352 + - summary: Fixed an issue with VM migration cancellation that prevented new migrations from starting. + pull_request: https://github.com/deckhouse/virtualization/pull/2282 + diff --git a/CHANGELOG/CHANGELOG-v1.9.md b/CHANGELOG/CHANGELOG-v1.9.md new file mode 100644 index 0000000000..5d9c88f681 --- /dev/null +++ b/CHANGELOG/CHANGELOG-v1.9.md @@ -0,0 +1,35 @@ +# Changelog v1.9 + +## Features + + + - **[core]** CLI lifecycle commands (start, stop, restart, evict, migrate) now support multiple VM targets, --all flag, --label-selector flag for label-based targeting, and --yes flag for non-interactive confirmation. [#2412](https://github.com/deckhouse/virtualization/pull/2412) + - **[core]** Compatible VirtualMachineOperation resources can now supersede another active operation on the same VM. [#2338](https://github.com/deckhouse/virtualization/pull/2338) + - **[core]** Added the `Uptime` column to `VirtualMachine` resources, showing the time since the VM started. [#2279](https://github.com/deckhouse/virtualization/pull/2279) + - **[vm]** The VM status now includes a `No bootable device` message when the VM cannot find a bootable disk to start. [#2404](https://github.com/deckhouse/virtualization/pull/2404) + - **[vm]** Add domain jobs and block-jobs info subcommands to vlctl. [#2280](https://github.com/deckhouse/virtualization/pull/2280) + - **[vm]** Added the ability to change `coreFraction` on a running VM without a restart. The new value is applied via live migration. [#2210](https://github.com/deckhouse/virtualization/pull/2210) + - **[vm]** Added the ability to attach additional network interfaces without a restart via the virtual machine's `.spec.networks`. [#2187](https://github.com/deckhouse/virtualization/pull/2187) + - **[vm]** System virtual machine resources (pods with `d8v-hp-` and `d8v-vm-` prefixes) now run as the `deckhouse` user, without root privileges. [#2097](https://github.com/deckhouse/virtualization/pull/2097) + - **[vm]** A restart is no longer required to attach and detach virtual disks and images via the virtual machine's `.spec.blockDeviceRefs`: + - Works for new virtual machines starting from v1.9.0. + - For previously created virtual machines, a restart is required to enable this behavior. [#2033](https://github.com/deckhouse/virtualization/pull/2033) + +## Fixes + + + - **[core]** Better handling Windows guests: start and migration should work in clusters with frequent CPU frequencies drifts [#2345](https://github.com/deckhouse/virtualization/pull/2345) + - **[module]** Fixed an issue where invalid `virtualization` module ModuleConfig settings could block the Deckhouse queue. [#2246](https://github.com/deckhouse/virtualization/pull/2246) + - **[module]** Fixed duplicate series on the `Virtualization / Overview` dashboard. [#2189](https://github.com/deckhouse/virtualization/pull/2189) + - **[vd]** Time spent in the `WaitForFirstConsumer` phase is no longer included in `.status.stats.creationDuration.totalProvisioning` of virtual disks. [#2379](https://github.com/deckhouse/virtualization/pull/2379) + - **[vd]** Allow ingress from virtualization namespace to importer pods [#2356](https://github.com/deckhouse/virtualization/pull/2356) + - **[vm]** Fixed disk update for stopped VMs with WaitForFirstConsumer storage class. Previously, when a VM was stopped and disks were changed, the KVVM was not updated, causing the VM to get stuck in starting phase. [#2407](https://github.com/deckhouse/virtualization/pull/2407) + - **[vm]** Fix VM migration failure caused by incorrect target disk size for filesystem-backed hotplug volumes. [#2402](https://github.com/deckhouse/virtualization/pull/2402) + - **[vm]** Fix possible scheduling problems after changing vmclass from Discovery type to Model type. [#2352](https://github.com/deckhouse/virtualization/pull/2352) + - **[vm]** Fixed an issue with VM migration cancellation that prevented new migrations from starting. [#2282](https://github.com/deckhouse/virtualization/pull/2282) + +## Chore + + + - **[api]** Removed the deprecated `VirtualMachineRestore` resource. Use `VirtualMachineOperation` with the `Clone` or `Restore` type, or `VirtualMachineSnapshotOperation` instead. [#2368](https://github.com/deckhouse/virtualization/pull/2368) +