v0.38.0-rc0

Release Notes

v0.38.0-rc0

Changelog

• d7f0bbf chore: lock published urls to preserve redirects
• e3c31f0 Temporarily disable GitHub Actions credentials.
• 3be954b build: INFENG-938: Update version format in Makefiles (#10142)
• 69b93b0 build: INFENG-940: Fix logic error in CircleCI config make-component job (#10143)
• 00870f5 build: INFENG-937: Publish Helm chart release candidates (#10141)
• 3910426 feat: remove searcher context from harness and master [MD-498] (#10131)
• 27bebdd build: INFENG-938: Tweak version string format (#10139)
• 30ad3c0 feat: add master configurations for access token max and default lifespans [DET-10464] (#10101)
• 782f7a0 revert: "chore: determined_master_host and friends helm support, better defaults" (#10134)
• 233e095 chore: add checkpoint and max slots config policy enforcements in PATCH experiment (#10125)
• b3f928b chore: determined_master_host and friends helm support, better defaults (#10092)
• 6755467 chore: bump Go version used by CI builds to 1.22.8 (#10127)
• 834eeda feat: add actual select all to glide tables [ET-238] (#10081)
• c7e0fb5 docs: add log signal release note and update docs (#10126)
• 02fcc74 test: Add test for filtering user by Role Id (#10095)
• f97fb5a build: INFENG-933: add GitHub action to start a minor release (#10112)
• 685918d docs: Add aurora postgres release note (#10115)
• a84f8c6 chore: SSO improvement feature requires Enterprise Edition. (#10124)
• c71617c feat: Log Signal Exp Config and Monitoring (#9947)
• 06b0b31 chore: fix merge exp flake (#10122)
• 962810a chore: improve messaging when workspace configs conflict with global … (#10121)
• 6158ef7 docs: Update postgres aurora info (#10116)
• 4b0c065 docs: log policies restore exp config (#10120)
• 186962c chore: add config policies to CLI reference docs (#10118)
• 11ea6f4 chore: clarify version overrides during helm installs (#10094)
• 4394f29 chore: standardize status api errors for task config policies (#10119)
• e834302 fix: Add on delete cascade to system_metrics (#10113)
• 3c59233 chore: populate final merged config with defaults when merging invariant configs (#10107)
• deb3772 feat: additional APIs to support "actual select all" functions [ET-238] (#10102)
• fd9cd8a feat: Allow master configuration for ssh key type (#10072)
• 5e9df7c docs: Update release notes (#10114)
• c655f33 docs: fix internal link in multi-rm docs page. (#10074)
• e7186fe docs: Update log policies (#10098)
• 993296b fix: update copy in experiment and trial headers (#10111)
• d74a462 docs: Describe sso improvements (#10110)
• 24d3390 chore: conditionally create VolumeSnapshotClass (#10103)
• f45ebb9 chore: improve documentation surrounding slot caps helm configuration (#10090)
• 0013fd0 ci: shorten test_pending_hpc.py (#10104)
• 22ad457 fix: version upgrade notification bug [CM-411] (#10069)
• 935fa66 fix: Log searche feedbacks (#10088)
• 29a08ec Revert "docs: Describe arbitrary metadata logging" (#10099)
• c6c476c chore: remove e2e_slurm_preemption test series (#10053)
• e6182ed docs: Describe arbitrary metadata logging (#10073)
• 539df5e chore: update CLI commands to work with global APIs (#10089)
• 1f2bea0 feat: update ConfigPolicies with docs link [CM-558] (#10055)
• 4afc15f build: INFENG-926: Fix version.sh version string output (#10085)
• 04861dd chore: return error if workspace config violates global constraints (#10076)
• 912f91e docs: task config policies release note (#10087)
• 6d56101 fix: remove flake-inducing logretention global singleton (#10016)
• b70a622 fix: correct token creation CLI to ensure it works with default expiry (#10084)
• b155332 docs: Describe task config policies (#9969)
• 27a014b fix: Tensorboard broken on unified install [CM-578] (#10080)
• bdb56a4 chore: INFENG-922: use correct gh_team tag for infrastructure (#10077)
• 91e358a INFENG-382: Release redesign (#10002)
• 34e4749 chore: remove redundant rm.ExternalPreemptionPending interface (#10071)
• 28bc072 feat: SSO Improvement - alter user_sessions table to include access token, implement CRUD ops, GET, POST, PATCH APIs and det token CLIs (#9867)
• 472baf9 feat: Add copy task id to task list (#10058)
• 2e822b7 chore: fix update invariant config and constraints (#10078)
• d69f7cc chore(deps): bump google.golang.org/grpc from 1.64.0 to 1.64.1 (#9910)
• e796b92 fix: run checkpoint GC more aggressively to ensure tensorboards are GC'd (#10017)
• a14525f fix: nil deref in usage of incomplete experiment config policies (#10068)
• 6c46a46 refactor: remove annotations requiring search ids in bulk action js (ET-241) (#10062)
• 3ca3418 Docs: describe data files apptainer (#10020)
• 315f65d chore: ntsc config not supported (#10056)
• 2e8de9b test: User Management test updates [CM-468] (#10051)
• 3fc9fed chore: experiment config slots to comply with constraint max slots (#10054)
• 1d5c984 chore: fix slices and maps merge test (#10063)
• 219409b chore: fix helptext for det user (#10060)
• 7d6a1a7 docs: add k8s RP example to the helm values.yaml. (#10027)
• 9efd96d fix: apply config policy constraints to PATCH /experiments/:id (#10048)
• dd6aeda chore: change error code back (#10042)
• 5a39ecb chore: check config policies on 'det notebook set priority' (#10047)
• 2ef2f12 feat: bulk actions matching filters (ET-241) (#9895)
• ac82b3c chore: default priority earlier to ensure constraints are satisfied [CM-553] (#10043)
• 34557ef feat: Extend LogViewer to support scrollable search (#10005)
• dadf75e chore: take invariant_config priority into account with manage job workflow (#10025)
• 2356f91 chore: remove e2e_slurm_misconfigured series tests (#10023)
• b243c26 ci: deflake test_disable_agent_zero_slots (#10040)
• 4e0f1c4 chore: validate global, admin input against task config policies & constraints (#10028)
• 3c1630f test: add e2e tests to the "move project" functionality on the "List View" (#10037)
• 0613cc6 docs: revise postgres permission setup instructions. (#10039)
• 2594d90 chore: remove e2e_slurm_gpu series tests (#10021)
• 1f7ccad chore: exp invariant config silent override during add or update (#10019)
• 30b197d feat: Global Config Policies UI [CM-522] (#10022)
• c27054d feat: add e2e tests for multi-sort filter on experiments lista (#9992)
• 9faa0cb chore: wait_for_task_state shows logs on failure (#10029)
• a166826 fix: Workspace Projects and Tasks test flakes [CM-554] (#10026)
• 33dfdaf test: Workspace Models tests [CM-538] (#9998)
• 7e8dbac fix: Update action bar row layout in UserManagement page (#9862)
• 5b1380c chore: check experiment constraints (#10018)
• f609a2d fix: remove formatDatetime (#10011)
• 9b6f0ac docs: Update release notes date (#9999)
• f5400ea feat: Add regex search to task logs API (#9994)
• ddca766 fix: correct expToWebhookConfig cache locking (#10014)
• 80b29fa feat: Config Policies UI, Workspaces Experiments [CM-521] (#10009)
• 262b4a9 chore: check task config policies against slots and max_slots (#10015)
• a0cc818 ci: replace no_op fixture with a noop api (#9997)
• 987b2a5 test: add e2e experiment list pagination test (#9993)
• 1297899 fix: use UID not username to set HOME dir (#10010)
• 49e72a8 chore: reword jsonschema extension docs (#9965)
• 63d728c fix: display archived column for runs and searches (#9987)
• 83a779e feat: check task config policy constraints before scheduling NTSC wor… (#9991)
• 0083d7e feat: add CLI commands for config policies [CM-423] (#9911)
• ac54cf8 ci: delete pointless test (#10004)
• 7f88390 fix: reset settings not working properly due to url encoding (#10000)
• 25ca6d0 fix: import missing time module (#9985)
• 8ab2145 chore: bump version: 0.37.0-dev0 -> 0.37.1-dev0
• 0760f74 chore: add docs dropdown link for new version
• 23f1f30 docs: add release notes for 0.37.0 (#9995)
• 9989475 test: Workspace Task tests [CM-476] (#9982)
• ad66d3f chore: implement PUT APIs for task config policies (#9983)
• 036336b docs: fix broken links (#9996)
• ac8fbf6 chore: check task config policy priority limit for [CM-490] (#9958)
• 8bc08e5 feat: Read and display log signal from DB (#9959)
• c8b1910 ci: increase datagrid rightclick timeout/ reduce worker count (#9951)
• e92c474 fix: fix default id search for runs (#9988)
• 3ca3d30 test: increase Reactivate test step timeout (#9986)
• bc3b2a6 fix: Reactivate User test flake (#9979)
• f2277f1 fix: fix hf on_save raise exception (#9977)
• dbeea99 fix: Cluster page height (#9975)
• d02495b fix: Deactivate User test flake (#9974)
• a8effe8 fix: show search progress in run table (#9976)
• cf9bdc8 feat: workspace task config policies UI [CM-478] (#9950)
• 924f663 ci: remove default arg from utils.run_command() (#9973)
• a96c5af docs: add docstring for PyTorchContext.current_train_epoch (#9972)
• 66f7a70 fix: grid hp samping ignored empty nests (#9966)
• 8c4f7a0 fix: correct dataPath for hyperparameters (#9971)
• 5c4be96 feat: add database snapshot functionality to Helm chart (#9956)
• 31d9573 fix: show - for empty data in searches table [ET-749] (#9963)