Backend.ai

--------------------

* FIX: In some kernels, git command has failed due to "undefined symbol: dlsym" error.
It's fixed by adding `-ldl` option to the linker flag of libbaihook.
* FIX: Reconnection and cancellation of etcd watchers used for manager launch detection

-----------------------

This is the last preview, feature-freeze release for v19.09 series.
Stability updates will follow in the v19.09.0 and possibly a few more v19.09.x releases.

* NEW: Image import ([171](https://github.com/lablup/backend.ai-manager/issues/171)) - currently this is limited to import Python-based kernels only.
This is implemented on top of batch tasks, with some specialization to prevent security issues
due to direct access to agent host's Docker daemon. Importing as service-port only image support will
be added in future releases.
* NEW: Batch tasks ([199](https://github.com/lablup/backend.ai-manager/issues/199)) - the kernel creation API now has an extra "type" parameter (default:
"interactive") which can be set "batch" with another new parameter "startupCommand". The batch-type
sessions (batch tasks) run the startup command immdediately after starting and automatically terminates
after finishing it, with a success/failure result recorded in the database retrievable as the "result"
column.

The execution logs are stored in the ".logs" user vfolder if present, which will be mounted all
user-owned compute sessions automatically. This log can be retrieved at any time using the new
`kernel/_/logs` API (task-logs API).
* IMPROVE: Allow admins to create sessions on behalf of other users.
* SECURITY-FIX: Privilege escalation because domain-admins could run sessions on behalf of super-admins
in the same domain.
* Various bug fixes and improvements.

-----------------------

This is the last preview, feature-freeze release for v19.09 series.
Stability updates will follow in the v19.09.0 and possibly a few more v19.09.x releases.
* NEW: Support batch tasks (148, lablup/backend.ai199)
* NEW: Support image import tasks, with internal-purpose security flag implementations (149,
lablup/backend.ai171)
* NEW: Intrinsic SSH support to any session, as "sshd" service port.
The host key and user keypair is randomly generated. To pin your own SSH keypair, create a
".ssh" user vfolder which will be automatically mounted to all your compute sessions.
* NEW: Add support for a new service port: "sftp" for large-file transfers with vfolders using
a special dedicated kernel.
* NEW: Add support for a new service port: "vscode" to access Visual Studio Code running as an
web application in the interactive sessions. Note that the sessions running VSCode are recommended to
have more than 2 GiB of free main memory. (147)
* IMPROVE: Enable the debugger port in TensorBoard. Note that this port is for private-use only
so that a TensorFlow process can send debug-logging data to it in the same container.
* IMPROVE: Add support for multiple TCP ports to be mapped for a single service.

-----------------------

* Minor bug fixes
* CHANGE: The default of "debug.coredump" config becomes false in the halfstack configuration.

-----------------------

* NEW: Add a new intrinsic service port "ttyd" for all kernels, which provides a clean and slick
web-based shell access.
* NEW: Add support for sftp service if the kernel supports it (146).
* FIX: Now "kernel_terminated" events carry the correct "reason" field, which is stored in the
"status_info" in the manager's kernels table.
* FIX: Avoid binary-level conflicts of Python library (libpythonmX.Y.so) in containers due to
"/opt/backend.ai/lib" mounts. This had crashed some vendor-specific images which relies on
Python 3.6.4 while our krunner daemon uses Python 3.6.8.
* CHANGE: The agent-to-manager notifications use Redis instead of ZeroMQ (144,
lablup/backend.ai-manager192, lablup/backend.ai-manager125), and make the agent to survive
intermittent Redis connection disruptions.

-----------------------

* NEW: A superadmin API to list all vfolder hosts and docker registires.
* UPDATE: resource/check-presets API is updated to return per-scaling-group remainings and
group/domain resource limits. ([184](https://github.com/lablup/backend.ai-manager/issues/184))
* UPDATE: Compute session GQL queries now include the `resource_opts` field.
* Minor bug fixes.