12 Releases published by 1 person

• 25.6.5

  published May 7, 2025

• 25.6.6rc1

  published May 21, 2025

• 25.6.6

  published May 23, 2025

• 25.8.0

  published May 23, 2025

• 25.8.1rc1

  published May 23, 2025

• 25.6.7

  published May 23, 2025

• 24.09.9

  published May 23, 2025

• 25.8.1

  published May 23, 2025

• 25.9.0rc1

  published May 29, 2025

• 25.6.8rc1

  published May 29, 2025

• 25.9.0rc2

  published May 29, 2025

• 25.9.0rc3

  published May 30, 2025

125 Pull requests merged by 10 people

• fix(BA-1447): Fix wrong request format in agent watcher sdk

  #4569 merged May 30, 2025

• fix(BA-1475): Add missing message to BgTaskFailedEvent

  #4563 merged May 30, 2025

• ci: Add error handling to build scripts

  #4564 merged May 30, 2025

• fix(BA-1438): Skip processing messages with None data in RedisQueue

  #4561 merged May 30, 2025

• fix(BA-1465): Fix broken Network SDK implementations

  #4558 merged May 30, 2025

• fix(BA-1438): Skip processing messages with None data in RedisQueue

  #4559 merged May 30, 2025

• fix: idle checker init arguments

  #4557 merged May 29, 2025

• feat: Add missing dependency for manager/auth in python distribution

  #4556 merged May 29, 2025

• feat(BA-1453): Add missing session status_history API

  #4543 merged May 29, 2025

• fix(BA-1454): Broken stream_pty method in Session SDK

  #4548 merged May 29, 2025

• refactor(BA-1421): Apply service layer in auth apis

  #4535 merged May 29, 2025

• fix(BA-1460): Broken Resource.usage_per_month SDK method

  #4546 merged May 29, 2025

• feat: Fix missing entity id in processor

  #4555 merged May 29, 2025

• fix(BA-1459): Broken list_presets API, and SDK

  #4541 merged May 29, 2025

• fix(BA-1457): Broken Keypair SDK methods (#4547)

  #4553 merged May 29, 2025

• fix(BA-1457): Broken Keypair SDK methods (#4547)

  #4552 merged May 29, 2025

• fix(BA-1457): Broken Keypair SDK methods

  #4547 merged May 29, 2025

• fix(BA-1473): Fix missing entity id in processor

  #4550 merged May 29, 2025

• refactor: Improve logging for error handling

  #4544 merged May 29, 2025

• feat(BA-1416): make resource fragmentation configurable

  #4533 merged May 29, 2025

• fix(BA-1442): Remove outdated Image SDK methods

  #4537 merged May 29, 2025

• fix: Correct typo in exception label for event failure metrics

  #4536 merged May 29, 2025

• refactor(BA-1471): Improve logging for error handling in various modules

  #4540 merged May 29, 2025

• fix(BA-1456): Remove useless print in ScalingGroup.list_available

  #4538 merged May 29, 2025

• feat(BA-1289): Allow anonymous users register TOTP key

  #4354 merged May 28, 2025

• feat(BA-984): Add Action Tests for Image

  #4048 merged May 28, 2025

• fix: mock accelerator resource not allocated after agent restart

  #4532 merged May 28, 2025

• feat(BA-1443): backend.ai mgr scheduler last-execution-time command

  #4507 merged May 28, 2025

• refactor: fix wrong fragment filename

  #4531 merged May 28, 2025

• feat(BA-1444): Add stage package to support deterministic step-by-step execution

  #4509 merged May 27, 2025

• refactor(BA-1325): Decouple keypair prepare logic from GraphQL

  #4510 merged May 27, 2025

• fix: GQL log is not printing

  #4505 merged May 27, 2025

• feat: Update api schema github actions

  #4511 merged May 27, 2025

• fix: Enhance error logging in BackgroundTaskManager to include exception details

  #4504 merged May 26, 2025

• feat(BA-1428): Refactor event dispatcher and handlers directory structure

  #4497 merged May 26, 2025

• fix(BA-1430): heartbeat register service when service is dead

  #4492 merged May 26, 2025

• feat(BA-1437): Add EventDomain.WORKFLOW enum value to support workflow-related events

  #4499 merged May 25, 2025

• fix(BA-461): Filter vfolder deletion info by vfolder status (#3446)

  #3488 merged May 23, 2025

• fix: Silent failure of DockerAgent's push_image(), pull_image() (#2572)

  #3389 merged May 23, 2025

• fix(BA-1380): Update Service.create() to comply with validation schema (#4449)

  #4484 merged May 23, 2025

• fix: Add missing validator to Redis Sentinel Config

  #4483 merged May 23, 2025

• fix(BA-1380): Update Service.create() to comply with validation schema

  #4449 merged May 23, 2025

• fix(BA-1379): Truncate generated session name in service creation to max allowed length (#4450)

  #4472 merged May 23, 2025

• feat(BA-1414): Add OpenTelemetry dependencies for enhanced observability

  #4479 merged May 23, 2025

• chore: Changed the number of CPUs in the dev environment to one (#4473)

  #4480 merged May 23, 2025

• chore: Changed the number of CPUs in the dev environment to one (#4473)

  #4481 merged May 23, 2025

• chore: Changed the number of CPUs in the dev environment to one

  #4473 merged May 23, 2025

• fix(BA-1413): Fix wrong method name in rpc call metric

  #4475 merged May 22, 2025

• fix(BA-1404): Calculate correct permissions for admins

  #4459 merged May 22, 2025

• fix(BA-1379): Truncate generated session name in service creation to max allowed length

  #4450 merged May 22, 2025

• fix(BA-1002): Incorrect handling of disallowed permission in GQL middleware

  #4463 merged May 22, 2025

• fix(BA-1402): Handle NoItems exception correctly in CLI framework

  #4465 merged May 22, 2025

• fix(BA-1403): Resolve file-upload to session issue (#4457)

  #4461 merged May 21, 2025

• fix(BA-1403): Resolve file-upload to session issue (#4457)

  #4460 merged May 21, 2025

• fix(BA-1403): Resolve file-upload to session issue

  #4457 merged May 21, 2025

• fix(BA-1401): Add missing defaults to BootstrapConfig

  #4453 merged May 21, 2025

• fix(BA-1399): Check unregistered email and update error code when vfolder invitation conflicts

  #4448 merged May 21, 2025

• feat(BA-1209): Add event log

  #4387 merged May 21, 2025

• fix: Fixed a loophole where consume could be missing at event dispatcher startup time

  #4444 merged May 21, 2025

• fix(BA-1369): Resolve admins vfolder leave issue (#4446)

  #4452 merged May 21, 2025

• fix(BA-1369): Resolve admins vfolder leave issue

  #4446 merged May 21, 2025

• fix(BA-1382): Initialize route session environment variables with fallback for None (#4447)

  #4451 merged May 21, 2025

• fix(BA-1382): Initialize route session environment variables with fallback for None

  #4447 merged May 21, 2025

• feat(BA-1305): Register service discovery and add http service discovery for prometheus

  #4438 merged May 20, 2025

• fix(BA-1394): Increase blocking timeout for message retrieval in redis message queue

  #4441 merged May 20, 2025

• fix(BA-1397): Add UUID serialization support in ExtendedJSONEncoder

  #4442 merged May 20, 2025

• fix(BA-1398): Improve error handling for token generation in ModelServingService

  #4443 merged May 20, 2025

• fix(BA-1276): Prevent invalid resource slot creation, and mutation

  #4314 merged May 20, 2025

• fix(BA-1393): resource usage-per-period CLI command not working (#4429)

  #4437 merged May 20, 2025

• fix(BA-1393): resource usage-per-period CLI command not working

  #4429 merged May 20, 2025

• fix: Improve handling of process termination races in the krunner's OOM logger (#4008)

  #4021 merged May 20, 2025

• fix(BA-1353): wrong location of event handling observer (#4392)

  #4428 merged May 20, 2025

• fix(BA-1395): Add an import statement from .kernel and updates the all variable accordingly

  #4436 merged May 20, 2025

• fix: ActionReporter not found crash (#4406)

  #4433 merged May 20, 2025

• feat(BA-1363): Remove subscribed_actions config, and change AuditLogMonitor (#4400)

  #4432 merged May 20, 2025

• fix(BA-1231): BgTaskFailedError is not propagated to the client

  #4272 merged May 19, 2025

• feat(BA-1265): Add ServiceConfig GQL API

  #4376 merged May 19, 2025

• fix(BA-1353): wrong location of event handling observer

  #4392 merged May 19, 2025

• fix(BA-1389): vfolder ls cmd not working (#4426)

  #4427 merged May 19, 2025

• fix(BA-1389): vfolder ls cmd not working

  #4426 merged May 19, 2025

• feat(BA-1321): Add force-delete support for VFolder in Python client SDK (#4353)

  #4420 merged May 19, 2025

• feat(BA-1321): Add force-delete support for VFolder in Python client SDK

  #4353 merged May 19, 2025

• fix(BA-1366): Add missing UserBgtaskEvent implementation

  #4404 merged May 17, 2025

• feat(BA-1374): Support relative path for AutoDirectoryPath

  #4413 merged May 17, 2025

• feat(BA-1211): Add kernel last seen event and handler

  #4386 merged May 16, 2025

• feat(BA-1292): Introduce LabelName enum

  #4328 merged May 16, 2025

• fix(BA-1367): Change pyzmq version on the python-kernel, compatible with python 3.13 (#4405)

  #4414 merged May 16, 2025

• fix(BA-1367): Change pyzmq version on the python-kernel, compatible with python 3.13 (#4405)

  #4415 merged May 16, 2025

• fix(BA-1367): Change pyzmq version on the python-kernel, compatible with python 3.13

  #4405 merged May 16, 2025

• fix(BA-1370): Use orjson in BackendAIError

  #4409 merged May 15, 2025

• fix: ActionReporter not found crash

  #4406 merged May 15, 2025

• feat(BA-1363): Remove subscribed_actions config, and change AuditLogReporter to AuditLogMonitor

  #4400 merged May 15, 2025

• fix: Remove wrong unified config injection

  #4401 merged May 15, 2025

• feat(BA-1364): Introduce ActionSpec

  #4393 merged May 15, 2025

• fix(BA-1362): Revert sane default config update

  #4395 merged May 15, 2025

• fix(BA-1331): Add validation_alias and serialization_alias to ManagerSharedConfig configs

  #4365 merged May 15, 2025

• feat(BA-1333): Make all manager configs to use the same ConfigLoader

  #4370 merged May 15, 2025

• feat(BA-1158): Introduce ProcessorPackage

  #4379 merged May 14, 2025

• fix(BA-1141): Wrong project-id parsing when creating project vfolder (#4144)

  #4382 merged May 13, 2025

• fix(BA-1345): Revert addition of SessionStatus.ERROR and KernelStatus.ERROR to dead status sets (#4384)

  #4385 merged May 13, 2025

• fix(BA-1345): Revert addition of SessionStatus.ERROR and KernelStatus.ERROR to dead status sets

  #4384 merged May 13, 2025

• fix(BA-1342): Make BaseAction's entity_type(), operation_type() classmethod

  #4377 merged May 13, 2025

• fix(BA-1334): Add missing KernelStatus.ERROR to dead kernel status set (#4371)

  #4372 merged May 12, 2025

• fix(BA-1334): Add missing KernelStatus.ERROR to dead kernel status set

  #4371 merged May 12, 2025

• refactor: Rename redis target

  #4363 merged May 10, 2025

• doc: Update towncrier command docs

  #4364 merged May 10, 2025

• feat(BA-1324): Refactor event propagation

  #4358 merged May 10, 2025

• fix: Remove wrong shared config

  #4362 merged May 9, 2025

• fix(BA-1330): Fixture populate not working

  #4360 merged May 9, 2025

• feat(BA-1263): Introduce Config Loaders, UnifiedConfig

  #4351 merged May 9, 2025

• feat(BA-1005): Migrate manager config to Pydantic

  #4317 merged May 9, 2025

• feat(BA-986): Add Action Test Code for User

  #4059 merged May 8, 2025

• feat(BA-988): Add Action Test Code for Group

  #4051 merged May 8, 2025

• feat(BA-1279): Add etcd service discovery

  #4343 merged May 7, 2025

• feat(BA-1300): Add error code to API exception message

  #4336 merged May 7, 2025

• fix(BA-1282): Use label's items for making resource info (#4341)

  #4350 merged May 7, 2025

• fix(BA-1302): Add TypeError handling in redis_helper (#4339)

  #4349 merged May 7, 2025

• feat(BA-1301): Add error code to metric

  #4337 merged May 7, 2025

• fix(BA-1302): Add TypeError handling in redis_helper

  #4339 merged May 7, 2025

• fix(BA-1284): Agent retries to retrieve kernel service info when it fails (#4321)

  #4348 merged May 6, 2025

• fix(BA-1281): Add default value of task_info value (#4340)

  #4347 merged May 6, 2025

• fix(BA-1284): Agent retries to retrieve kernel service info when it fails

  #4321 merged May 6, 2025

• fix(BA-1281): Add default value of task_info value

  #4340 merged May 6, 2025

• fix(BA-1282): Use label's items for making resource info

  #4341 merged May 6, 2025

• feat(BA-1241): Add release script

  #4316 merged Apr 30, 2025

3 Pull requests opened by 2 people

• fix: container not created when host network is enabled

  #4394 opened May 15, 2025

• feat(BA-1451): Refactor event dispatchers of idle checker

  #4516 opened May 27, 2025

• fix(BA-1235): Allow String data to set quota limit

  #4527 opened May 28, 2025

95 Issues closed by 5 people

• Fix `AgentWatcher.get_status` API to use proper query parameter

  #4513 closed May 30, 2025

• GET method parameter error in SDK

  #4111 closed May 30, 2025

• Add "*" support subscribed_actions

  #4396 closed May 30, 2025

• Introduce Soft Delete Mechanism for `Image` DB row

  #3606 closed May 30, 2025

• Improve Action Reporter

  #4298 closed May 30, 2025

• Propagate errors when some tasks in BgTask fail

  #3700 closed May 30, 2025

• cannot rename session on session detail panel

  #4407 closed May 30, 2025

• Wrong types in ResourcePolicy GQL modifier

  #4228 closed May 30, 2025

• Deserialization error in `BgtaskPartialSuccessEvent`

  #3876 closed May 30, 2025

• Implement AuditLog table

  #3914 closed May 30, 2025

• Add processorPackage interface

  #4280 closed May 30, 2025

• Add unified config interface

  #4304 closed May 30, 2025

• Add Audit Log Schema GQL API, SDK and CLI

  #4390 closed May 30, 2025

• `message` is missing from `BgTaskFailedEvent`

  #4562 closed May 30, 2025

• Fix `Network.create` SDK implementation to successfully create network

  #4528 closed May 30, 2025

• Remove false alerts from Autoclaim

  #4500 closed May 30, 2025

• orjson.JSONDecodeError in `ComputeSession.stream_pty`

  #4519 closed May 29, 2025

• `Session.get_status_history` API not found error

  #4518 closed May 29, 2025

• Apply audit logs for user login APIs

  #4486 closed May 29, 2025

• `Resource.usage_per_month` input validation error

  #4525 closed May 29, 2025

• `Resource.list` server error

  #4524 closed May 29, 2025

• Fix to avoid missing entity IDs in processor

  #4549 closed May 29, 2025

• Follow up 2FA config changes

  #4545 closed May 29, 2025

• Add anonymous request router to webserver for TOTP key registration

  #4326 closed May 29, 2025

• Make accelerator fragmentation option configurable

  #4478 closed May 29, 2025

• Remove outdated Image SDK

  #4503 closed May 29, 2025

• Enhance log information

  #4539 closed May 29, 2025

• `ScalingGroup.list_available` print result

  #4521 closed May 29, 2025

• Check TOTP status in manager `POST_AUTHORIZE` hook

  #4324 closed May 28, 2025

• Add Action Test Code for `Image`

  #3964 closed May 28, 2025

• Make manager CLI command to check each scheduler's last execution time

  #4506 closed May 28, 2025

• Implement provisioner stage

  #4508 closed May 27, 2025

• Decouple keypair preparation logic from GraphQL

  #4359 closed May 27, 2025

• Separate event dispatcher in manager

  #4490 closed May 26, 2025

• Change hearbeat to register when service is removed

  #4491 closed May 26, 2025

• Add `workflow` event domain

  #4498 closed May 25, 2025

• Inconsistent handling of nullable value for `group_name` in service creation API

  #4417 closed May 23, 2025

• Apply OTEL

  #4476 closed May 23, 2025

• Wrong permission calculation when admins query vfolders in Project scopes

  #4458 closed May 22, 2025

• RPC metric collection errors in agent

  #4474 closed May 22, 2025

• Potential validation error for default service session name generated by client SDK

  #4416 closed May 22, 2025

• Modify the permission check logic in the GQLMutationPrivilegeCheckMiddleware

  #4464 closed May 22, 2025

• Handle `NoItems` exceptions correctly where using our CLI framework

  #4455 closed May 22, 2025

• Cannot upload files to compute session

  #4456 closed May 21, 2025

• Missing default values in BootstrapConfig

  #4454 closed May 21, 2025

• got 500 error when inviting agin with same email

  #4445 closed May 21, 2025

• Develop Event Logging System

  #4238 closed May 21, 2025

• cannot leave an invited folder when I'm an admin.

  #4408 closed May 21, 2025

• Type error when handling nullable `endpoint.environ` in route creation

  #4419 closed May 21, 2025

• Add service discovery to components

  #4344 closed May 20, 2025

• CLI: generate-token occurs `aiohttp.client_exceptions.ContentTypeError: 404` error.

  #4440 closed May 20, 2025

• CLI: cannot use `service get-endpoint` command. (TypeError)

  #4439 closed May 20, 2025

• Remove excessive autoclaim logs

  #4434 closed May 20, 2025

• Prevent creating `ResourcePreset` without intrinsic resource

  #4312 closed May 20, 2025

• `backend.ai admin resource usage-per-period` not working

  #4431 closed May 20, 2025

• Can't get attribute `KernelLifecycleEventReason` on ai.backend.common.events module

  #4435 closed May 20, 2025

• Commit quota exceeded error not passed to response

  #4271 closed May 19, 2025

• Implement ServiceConfig GQL API

  #4305 closed May 19, 2025

• client py `vfolder ls` occurs KeyError: files

  #4424 closed May 19, 2025

• Wrong location of event handling observer

  #4391 closed May 19, 2025

• `vfolder ls` CLI command not working

  #4425 closed May 19, 2025

• Add VFolder force-delete API to client SDK

  #4352 closed May 19, 2025

• cannot rescan registry on environment page

  #4411 closed May 18, 2025

• Add missing `UserBgtaskEvent` implementation

  #4402 closed May 17, 2025

• Add relative path feature to `AutoDirectoryPath`

  #4412 closed May 17, 2025

• Storage host list is not loaded

  #4373 closed May 16, 2025

• Add Logic for Manager to Check 'Last Seen' Status

  #4240 closed May 16, 2025

• Replace Container label names from a hard-coded string to Enum

  #4327 closed May 16, 2025

• Change pyzmq version, compatible with python 3.13

  #4403 closed May 16, 2025

• Use `orjson` in serializing `BackendAIError`

  #4410 closed May 15, 2025

• add zeromq-devel as a required package for RHEL-based distributions

  #4388 closed May 15, 2025

• Remove `subscribed_actions` config, and change AuditLogReporter to AuditLogMonitor

  #4398 closed May 15, 2025

• Introduce ActionSpec

  #4399 closed May 15, 2025

• Revert sane default value update

  #4397 closed May 15, 2025

• Refactor the `alias` of `SharedConfig` into `validation_alias` and `serialization_alias`

  #4367 closed May 15, 2025

• Make all configs of the Manager to share the same Config Loader

  #4368 closed May 15, 2025

• Introduce ProcessorPackage

  #4380 closed May 14, 2025

• Revert addition of `KernelStatus.ERROR` to dead kernel status set

  #4383 closed May 13, 2025

• Make BaseAction's `entity_type()`, `operation_type()` classmethod

  #4378 closed May 13, 2025

• Shared config under Volumes is not loaded properly.

  #4375 closed May 12, 2025

• `KernelStatus.ERROR` missing from dead kernel status set

  #4369 closed May 12, 2025

• Refactor bgtask event

  #4357 closed May 10, 2025

• Fixture populating not working

  #4361 closed May 9, 2025

• Implement config loaders for each type

  #4303 closed May 9, 2025

• Migrate to pydantic based manager configuration schema

  #3993 closed May 9, 2025

• Add Action Test Code for User

  #3966 closed May 8, 2025

• Add Action Test Code for Group

  #3968 closed May 8, 2025

• Pass error codes in the API (REST, GraphQL)

  #4334 closed May 7, 2025

• Add error code to metric

  #4335 closed May 7, 2025

• Handle TypeError in redis.py when client connection is closed.

  #4338 closed May 7, 2025

• Agent does not retry failed kernel creation

  #4320 closed May 6, 2025

• Fix key error when bgtask is already done

  #4318 closed May 6, 2025

• GPU slider always blacked out when creating session

  #4319 closed May 6, 2025

• .

  #4333 closed May 3, 2025

• Create release execution script

  #4282 closed Apr 30, 2025

50 Issues opened by 6 people

• Improvement of Exception Handling for InvalidAPIParameters

  #4568 opened May 30, 2025

• Skipped image log (`no tag`) spamming

  #4567 opened May 30, 2025

• image required resources sync issue

  #4566 opened May 30, 2025

• Prepare 25.9.0 release

  #4565 opened May 30, 2025

• Unable to create SFTP session

  #4534 opened May 28, 2025

• Fix `ComputeSession().upload()` creating malformed filenames when using relative paths with parent directory references

  #4530 opened May 28, 2025

• Fix `KeyError` when calling `ComputeSession.restart`

  #4529 opened May 28, 2025

• `Model.list` implementation is blank

  #4526 opened May 27, 2025

• `session.Auth.login`, `session.Auth.logout` 404 not found error

  #4523 opened May 27, 2025

• `KeyPair.activate`, ``KeyPair.deactivate` NotNullViolationError

  #4522 opened May 27, 2025

• Deprecate `session.ServerLog` SDK and replace `session.ErrorLog`

  #4520 opened May 27, 2025

• Timeout error of `ComputeSession.complete` SDK

  #4517 opened May 27, 2025

• Refactor idle checker dispatchers

  #4515 opened May 27, 2025

• Add `installed` field to ImageNode GQL type

  #4514 opened May 27, 2025

• Image rescanning not working on macOS after upgrade python 3.13

  #4512 opened May 26, 2025

• Implement `Image.PreloadImage` method in SDK

  #4502 opened May 26, 2025

• Update Python Client SDK for Tester

  #4501 opened May 26, 2025

• Add metrics to check that scheduling is working properly

  #4496 opened May 24, 2025

• Fixed an issue where Agent would get stuck when running on a server

  #4495 opened May 24, 2025

• Analyze scheduling stuck

  #4494 opened May 24, 2025

• Improve scheduler stability

  #4493 opened May 24, 2025

• Separate Client layer

  #4489 opened May 23, 2025

• Add auth service & processors

  #4488 opened May 23, 2025

• Develop a tester to verify consistency of server behavior

  #4487 opened May 23, 2025

• Write down test template

  #4485 opened May 23, 2025

• Avoid creating a new ClientSession for every HTTP request

  #4477 opened May 22, 2025

• Process Dangling container event in manager

  #4471 opened May 22, 2025

• Container Abnormal Termination Handling (Eventing)

  #4470 opened May 22, 2025

• Implement Kernel Termination Logic (Normal & Abnormal)

  #4469 opened May 22, 2025

• Refactor Agent's Kernel Creation Flow

  #4468 opened May 22, 2025

• Implement Kernel Runner

  #4467 opened May 22, 2025

• Define Test Scenarios for Verifying Agent Functionality

  #4466 opened May 22, 2025

• Refactor `UnifiedConfig`, `BootstrapConfig` to use `BaseConfigModel`

  #4462 opened May 21, 2025

• Outdated CLI commands

  #4430 opened May 19, 2025

• Define sub-issues and technical specifications for agent refactoring

  #4423 opened May 19, 2025

• Define Test Scenarios for Verifying Agent Functionality

  #4422 opened May 19, 2025

• Refactoring agent to improve stability

  #4421 opened May 19, 2025

• Misc Epic Sprint #9

  #4418 opened May 18, 2025

• Develop component-based subscription management

  #4356 opened May 8, 2025

• Refactor event architecture

  #4355 opened May 8, 2025

• Separate user event hub code with event dispatcher

  #4346 opened May 6, 2025

• Add prometheus sd_configs api to update prometheus targets

  #4345 opened May 5, 2025

• Implement Service Discovery feature

  #4342 opened May 5, 2025

• Apply account manager

  #4332 opened May 3, 2025

• Research potential new infrastructure options

  #4331 opened May 3, 2025

• Evaluate current infrastructure dependencies

  #4330 opened May 3, 2025

• Review Infra dependency

  #4329 opened May 3, 2025

• Add token-based TOTP key registration APIs to TOTP plugin

  #4325 opened May 3, 2025

• Relocate TOTP settings from web server to manager

  #4323 opened May 3, 2025

• Change TOTP mechanism to reduce vulnerability

  #4322 opened May 3, 2025

82 Unresolved conversations

Sometimes conversations happen on old items that aren’t yet closed. Here is a list of all the Issues and Pull Requests with unresolved conversations.

• feat(BA-1213): Add detection and event notifications for kernel/container mismatches

  #4252 commented on May 11, 2025 • 7 new comments

• Implement API for downloading multiple file as a zip file

  #3608 commented on May 23, 2025 • 0 new comments

• Add functionality to provide statistical indicators to Prometheus for frontend use.

  #3802 commented on May 23, 2025 • 0 new comments

• Collect and store `Image` metadata in database

  #3696 commented on May 23, 2025 • 0 new comments

• Design Data Structure for Access Rights

  #3850 commented on May 23, 2025 • 0 new comments

• Set the maximum memory value for session creation to be limited by the smallest node.

  #3801 commented on May 23, 2025 • 0 new comments

• Replace DB `sessions`, `kernels`, `vfolders` table's `status_history` column types with `list`

  #3901 commented on May 23, 2025 • 0 new comments

• Implement Configuration File Validation in CLI

  #3946 commented on May 23, 2025 • 0 new comments

• Introduce a manageable configuration structure

  #3942 commented on May 23, 2025 • 0 new comments

• Passing information such as host and port in the redirect URL information together

  #3953 commented on May 23, 2025 • 0 new comments

• Automate Display of Newly Added Configuration Fields

  #3947 commented on May 23, 2025 • 0 new comments

• Develop CLI Command for Viewing Version-Specific Configuration Changes

  #3945 commented on May 23, 2025 • 0 new comments

• Misc Image Issues

  #3974 commented on May 23, 2025 • 0 new comments

• Add missing ownership attributes to Kernels where these values are null

  #3981 commented on May 23, 2025 • 0 new comments

• CLONE - Migrate to pydantic based agent configuration schema

  #4075 commented on May 23, 2025 • 0 new comments

• Add `image purge` CLI command for hard deleting ImageRow from DB

  #3979 commented on May 23, 2025 • 0 new comments

• Migrate to pydantic based storage-proxy configuration schema

  #4076 commented on May 23, 2025 • 0 new comments

• Expose allocation map to `AbstractComputePlugin.get_hooks()` interface

  #4100 commented on May 23, 2025 • 0 new comments

• `create_user_resource_policy` mutation error due to absense of arguments' default

  #4115 commented on May 23, 2025 • 0 new comments

• Cannot set Storage quota for user/project bigger than 9007199254740991

  #4208 commented on May 23, 2025 • 0 new comments

• Remove 'possibly-undefined' warning/errors of mypy

  #4157 commented on May 23, 2025 • 0 new comments

• feat(BA-1185): Add `service-config.toml` to `VFolder`, `VFolderNode` GQL

  #4220 commented on May 13, 2025 • 0 new comments

• feat(BA-817): Better error log during server context initialization

  #3796 commented on May 6, 2025 • 0 new comments

• feat(BA-466): Replace `vfolder`'s `status_history`'s type `dict` with `list`

  #3205 commented on May 8, 2025 • 0 new comments

• feat(BA-465): Add `dummy_kernels` table for testing `sql_json_merge`

  #3204 commented on May 8, 2025 • 0 new comments

• feat(BA-20): Replace `sessions`, `kernels`'s `status_history`'s type `dict` with `list`

  #3201 commented on May 8, 2025 • 0 new comments

• Make the shared memory setting for new sessions more intuitive

  #1726 commented on May 28, 2025 • 0 new comments

• Add Operational Policy for SMTP Reporter

  #4169 commented on May 26, 2025 • 0 new comments

• Separate setup code in storage proxy

  #3932 commented on May 26, 2025 • 0 new comments

• Store inference session creation params as config file in Model VFolder

  #4211 commented on May 26, 2025 • 0 new comments

• Make scripts for backend.ai

  #4281 commented on May 23, 2025 • 0 new comments

• Create script for generating test environment

  #4285 commented on May 23, 2025 • 0 new comments

• Implement an API to stream a ZIP file based on a JWT from storage-proxy

  #3445 commented on May 23, 2025 • 0 new comments

• Add `dummy_kernels` table for testing `sql_json_merge`

  #3398 commented on May 23, 2025 • 0 new comments

• Change the `status_history` column in the `kernels` and `sessions` tables to list.

  #3200 commented on May 23, 2025 • 0 new comments

• Reduce potential hanging during manager termination

  #3423 commented on May 23, 2025 • 0 new comments

• Implement an API to generate a JWT for downloading multiple files from manager

  #3444 commented on May 23, 2025 • 0 new comments

• Replace `vfolder`'s `status_history`'s type `dict` with `list`

  #3399 commented on May 23, 2025 • 0 new comments

• Enhance usability of runtime variant feature

  #3561 commented on May 23, 2025 • 0 new comments

• Implement an API to generate a JWT for downloading multiple files from storage-proxy

  #3663 commented on May 23, 2025 • 0 new comments

• Manage half-stack config files by collecting them in one place

  #3687 commented on May 23, 2025 • 0 new comments

• Change source data of utilization idle checker to Prometheus

  #4141 commented on May 23, 2025 • 0 new comments

• Tenstorrent Wormhole device support

  #3501 commented on May 21, 2025 • 0 new comments

• Forces user to use node from nvm, even if node is installed locally

  #3678 commented on May 21, 2025 • 0 new comments

• Migrate to pydantic-based logging configuration schema

  #3518 commented on May 21, 2025 • 0 new comments

• Update the pants plugin development settings

  #3343 commented on May 21, 2025 • 0 new comments

• Notarize macOS self-bootstrapping builds

  #1842 commented on May 21, 2025 • 0 new comments

• Migrate to pydantic-based local configuration schema

  #2764 commented on May 21, 2025 • 0 new comments

• Add GPU Monitoring metrics

  #4074 commented on May 21, 2025 • 0 new comments

• Refactor Exception

  #3677 commented on May 21, 2025 • 0 new comments

• Add Action Test code

  #3903 commented on May 21, 2025 • 0 new comments

• set user's timezone information on the Backend.AI

  #3890 commented on May 21, 2025 • 0 new comments

• Implement Kafka as the Event Message Queue

  #3609 commented on May 21, 2025 • 0 new comments

• Separate broadcaster from event producer

  #4250 commented on May 21, 2025 • 0 new comments

• Model service logic improvements

  #4249 commented on May 21, 2025 • 0 new comments

• Implement broadcaster for broadcast events using pubsub

  #4251 commented on May 19, 2025 • 0 new comments

• Replace redis_client with AbstractMessageQueue in EventDispatcher, EventProducer.

  #3887 commented on May 19, 2025 • 0 new comments

• Improve Agent Selector

  #4294 commented on May 13, 2025 • 0 new comments

• Add Action Test Code for `Session`

  #4052 commented on May 12, 2025 • 0 new comments

• Add Action Test Code for `Resource`

  #3970 commented on May 11, 2025 • 0 new comments

• Add scaling group Service & Processors

  #3958 commented on May 5, 2025 • 0 new comments

• Enhance Scheduler Reliability with Exception Handling and Retry Mechanisms

  #3514 commented on May 3, 2025 • 0 new comments

• Update Kernel States Based on Heartbeat Events

  #4241 commented on May 23, 2025 • 0 new comments

• Develop API for Kernel Regeneration

  #4236 commented on May 23, 2025 • 0 new comments

• Trigger Kernel/Container Dangling Events

  #4242 commented on May 23, 2025 • 0 new comments

• Create Recovery Logic for Dangling Events

  #4239 commented on May 23, 2025 • 0 new comments

• Implement Heartbeat Check Mechanism

  #4237 commented on May 23, 2025 • 0 new comments

• Cannot set storage quota values for users/projects larger than JavaScript BigInteger

  #4275 commented on May 23, 2025 • 0 new comments

• Add interface or type for probing feature

  #4270 commented on May 23, 2025 • 0 new comments

• Improper error handling when ZMQ socket connection fails after kernel creation

  #4255 commented on May 23, 2025 • 0 new comments

• Improved manager server stability

  #4279 commented on May 23, 2025 • 0 new comments

• Support for LTS versions

  #4288 commented on May 23, 2025 • 0 new comments

• Fix scheduler error message

  #4311 commented on May 23, 2025 • 0 new comments

• Create Docker compose for new LTS version.

  #4307 commented on May 23, 2025 • 0 new comments

• Make DB as Source of Truth of Sessions

  #4235 commented on May 21, 2025 • 0 new comments

• Issues for PALI integration

  #4300 commented on May 21, 2025 • 0 new comments

• Add default environment variables for PyTorch/TensorFlow distributed training

  #4243 commented on May 21, 2025 • 0 new comments

• Expand the accelerator metadata format

  #3324 commented on May 21, 2025 • 0 new comments

• Inaccurate max value of `cpu_util` in live stat

  #4126 commented on May 21, 2025 • 0 new comments

• Generate /etc/timezone file in the container if the host does not have /etc/timezone

  #3841 commented on May 21, 2025 • 0 new comments

• Add actions to generate new version's necessary information for Backend.ai users at the release point.

  #3741 commented on May 21, 2025 • 0 new comments

• Handle error log when manager fails to connect storage proxy

  #3790 commented on May 21, 2025 • 0 new comments