Add /v1/messages handler with Anthropic-to-OpenAI translation, upstream
failover, and probe endpoints. Replace shell-based functional test scripts
with pkg/aiproxy/ft and climc test commands; consolidate documentation.
Add auto_register_aiproxy on llm_deployment creation, background sync task,
and register/unregister CLI actions. Link aiproxy providers, routings, and
models to llm replicas via llm_id and llm_deployment_id fields.
Add a generic `feature-config <feature> <on|off>` command alongside the
existing per-feature `feature-config-<name>` commands. A `--force` flag
skips the known-features validation so new features can be toggled
without a code change. Also adds "ai" to the known feature list.
* feat(region, climc): Add kickstart parameters and metadata
- Add kickstart parameters and metadata support
- Add kickstart CLI options and management commands
- Set kickstart metadata in PostCreate
- Add related kickstart metadata fields
* feat(host): Implement automated VM installation
- Mounts the installation ISO to a unique temp path, reusing existing mounts.
- Extracts kernel (`vmlinuz`, `initrd`) to configure QEMU boot.
- Injects kernel parameters to trigger the autoinstall.
- Add kickstart configuration validation.
Modified `guestLauncher` script to handle kernel parameters with spaces
that were previously truncated by the Python script.
* feat(host): Implement Kickstart monitoring
- Automatically updates the VM status upon
receiving `KICKSTART_SUCCESS` or `KICKSTART_FAILED` signals.
- On success, cleans up temporary mount points
and restarts the server via the Region API.
* feat(region): add kickstart type metadata
Adds the `VM_METADATA_KICKSTART_TYPE` to distinguish between
kickstart configurations passed by URL or by direct content.
* feat(host): Handle kickstart installing and timeout
- Set status to 'installing' upon successful kickstart boot
- Handle KICKSTART_INSTALLING serial message for status updates
- Add timeout handling for kickstart status monitoring
* feat(host): provide Kickstart config via content
Embed the provided config content as a file in a
labeled ISO (mkisofs), then mount the ISO during
the Kickstart installation to supply the config.
* refactor(host): Enhance the kickstart installation
- Integrates kickstart status directly into the VM status, replacing metadata.
- Unifies locations for kickstart logs and temporary files.
- Uses regex for more robust status matching.
- Moves helper functions to `kickstart_helper.go`.
- Fixes an incorrect kickstart configuration filename.
- Adjusts kickstart logging.
* feat(region, host): add kickstart support for openEuler
* refactor(host): extract kickstart cleanup function
* feat(region, climc): add kickstart-complete cli
Add this command to manually update the kickstart status and
reboot VM if needed. This is mainly used by users to manually
update the kickstart status when kickstart status detection is abnormal.
* refactor(region): reduce code duplication in Kickstart validation
- unify two validateKickstartConfig functions
- extract determineKickstartType function for type determination
* fix(region, host): fix kickstart auto-reboot
- Adjusted the timing of setting the status to
`kickstart_pending` to avoid incorrect reboot checks
- Skip setting kickstart if the status is already
`kickstart_completed` to prevent unnecessary reboots
- Added cleanup of kickstart files upon VM deletion
* feat(host): download kickstart config from URL
- download content when passing config via URL
- unified boot parameters
* feat(host): copy kickstart kernel files
* feat(host): ensure kickstart completion signal in config
---------
Co-authored-by: happygame <happygame1024@gmail.com>
* feat(llm): init vllm as llm_container
* feat(vllm): fix save-instant-model for vllm
* feat(llm): support quick-model for vllm
* feat(vllm): support preferred-model in llm
* fix(llm): add ValidateSpec in llm_container_driver
* feat(vllm): auto-set vllm params
- climc: add VlanId to host-add-netif, host-remove-netif, host-enable-netif,
host-disable-netif
- region: set bn.VlanId from netif.VlanId when attaching to network
- mcclient: include Vlan_ID in baremetalnetworks list columns
* fix(llm): mcp-agent apiKey encrypt
* fix(llm): error-handle when instant-model import
* feature(llm): add get-available-network for llm
* feature(llm): support history for mcp-agent-chat
* fix(llm): llm create with net instead of networktype & networkid
* fix(llm): delete modelName in llmSku & fix some bugs
Zexi Li