GitHub / dstackai/dstack / commits
dstack is an open-source alternative to Kubernetes and Slurm, designed to simplify GPU allocation and AI workload orchestration for ML teams across top clouds, on-prem clusters, and accelerators.
| SHA | Message | Author | Date | Stats |
|---|---|---|---|---|
| fb4a4da8 | Optimize create instance on AWS (#3556) |
Victor Skvortsov <v****3@g****m>
Committed by: GitHub <n****y@g****m> |
4 months ago | |
| cfea44e0 | Update SKILL.md to standardize run name formatting and add permissions guardr... |
Andrey Cheptsov <5****5@u****m>
Committed by: GitHub <n****y@g****m> |
4 months ago | |
| 8ff914b1 | Update SKILL.md with authentication details and OpenAI model usage instructio... |
Andrey Cheptsov <5****5@u****m>
Committed by: GitHub <n****y@g****m> |
4 months ago | |
| fa001e58 | Disable autoflush (#3553) |
Victor Skvortsov <v****3@g****m>
Committed by: GitHub <n****y@g****m> |
4 months ago | |
| 2fd45cc9 | [runner] Write termination_{reason,message} to the log (#3550) |
Dmitry Meyer <m****e@u****m>
Committed by: GitHub <n****y@g****m> |
4 months ago | |
| 3ec1a6de | Events UI #3309 (#3532) |
Oleg <v****k@g****m>
Committed by: GitHub <n****y@g****m> |
4 months ago | |
| 3149be8f | Fix `probes=None` server incompatibility (#3543) |
jvstme <3****e@u****m>
Committed by: GitHub <n****y@g****m> |
4 months ago | |
| 78678c14 | Fix `probes=None` client incompatibility (#3544) |
jvstme <3****e@u****m>
Committed by: GitHub <n****y@g****m> |
4 months ago | |
| 685e0cd8 | [Docs] Update SKILL.md (#3547) |
Andrey Cheptsov <5****5@u****m>
Committed by: GitHub <n****y@g****m> |
4 months ago | |
| 409da83c | [Bug]: Run doesn't show Waiting runner limit exceeded in Error #3545 (#3546) |
Andrey Cheptsov <5****5@u****m>
Committed by: GitHub <n****y@g****m> |
4 months ago | |
| 0182756b | [UX] Remove creation_policy from Concept #3527 (#3542) |
Andrey Cheptsov <5****5@u****m>
Committed by: GitHub <n****y@g****m> |
4 months ago | |
| ebfd8952 | [UX] Improve `dstack fleet` output layout (#3529) |
Andrey Cheptsov <5****5@u****m>
Committed by: GitHub <n****y@g****m> |
4 months ago | |
| b2936728 | [Docs] Unlisted `cudo` backend (#3539) |
Andrey Cheptsov <5****5@u****m>
Committed by: GitHub <n****y@g****m> |
4 months ago | |
| 5ccbf0f8 | CLI crashes with 'Operation not permitted' when log file is not writable (#3538) |
Andrey Cheptsov <5****5@u****m>
Committed by: GitHub <n****y@g****m> |
4 months ago | |
| 93bb51c7 | Add probe `until_ready` configuration option (#3530) |
jvstme <3****e@u****m>
Committed by: GitHub <n****y@g****m> |
4 months ago | |
| 1b65d8f0 | Add job in-place update event (#3541) |
jvstme <3****e@u****m>
Committed by: GitHub <n****y@g****m> |
4 months ago | |
| d01fcfd2 | Add run in-place update event (#3540) |
jvstme <3****e@u****m>
Committed by: GitHub <n****y@g****m> |
4 months ago | |
| 813aed50 | Add `/api/project/{project_name}/instances/get` (#3535) |
jvstme <3****e@u****m>
Committed by: GitHub <n****y@g****m> |
4 months ago | |
| 652c8d8d | [CLI]: `dstack event --watch` (#3533) |
jvstme <3****e@u****m>
Committed by: GitHub <n****y@g****m> |
4 months ago | |
| 2410c47f | [Docs] Update SKILL.md (#3536) |
Andrey Cheptsov <5****5@u****m>
Committed by: GitHub <n****y@g****m> |
4 months ago | |
| 851d8a6f |
[Services] Add default probes if model is set (#3524)
Co-authored-by: jvstme <3****e@u****m>, jvstme <3****e@u****m>, jvstme <3****e@u****m> |
Andrey Cheptsov <5****5@u****m>
Committed by: GitHub <n****y@g****m> |
4 months ago | |
| c2d4655c | [Runpod] Make Community Cloud an "opt-in" (disable by default) #3531 (#3534) |
Andrey Cheptsov <5****5@u****m>
Committed by: GitHub <n****y@g****m> |
4 months ago | |
| 2acb6dee | [Feature]: Show probe statuses in the UI (#3521) |
Oleg <v****k@g****m>
Committed by: GitHub <n****y@g****m> |
4 months ago | |
| 0c3e5648 | [UI] Add Spot policy configuration option to the fleet wizard #3513 (#3520) |
Oleg <v****k@g****m>
Committed by: GitHub <n****y@g****m> |
4 months ago | |
| c46205bc | Add service and replica registration events (#3516) |
jvstme <3****e@u****m>
Committed by: GitHub <n****y@g****m> |
4 months ago | |
| 14ef341c |
[Docs] Remove the mention of the gateway endpoint #3514 (#3518)
Co-authored-by: jvstme <3****e@u****m>, jvstme <3****e@u****m>, jvstme <3****e@u****m> |
Andrey Cheptsov <5****5@u****m>
Committed by: GitHub <n****y@g****m> |
4 months ago | |
| 1dc61213 | [Docs] Add dstack skill (#3525) |
Andrey Cheptsov <5****5@u****m>
Committed by: GitHub <n****y@g****m> |
4 months ago | |
| dbdbf4d5 | Rename event target filters in UI (#3517) |
jvstme <3****e@u****m>
Committed by: GitHub <n****y@g****m> |
4 months ago | |
| c8d6d1c2 | [UI] Add Spot policy configuration option to the fleet wizard (#3519) |
Oleg <v****k@g****m>
Committed by: GitHub <n****y@g****m> |
4 months ago | |
| 9c81898b | Switch UI to pagination-based projects and users API (#3503) |
Oleg <v****k@g****m>
Committed by: GitHub <n****y@g****m> |
4 months ago | |
| 916b94e3 | [Docs] Added `Spot policy` (#3512) |
Andrey Cheptsov <5****5@u****m>
Committed by: GitHub <n****y@g****m> |
4 months ago | |
| 0d58216d | Showed counters for Project and User List | Oleg Vavilov <v****k@g****m> | 4 months ago | |
| bb027888 |
[Docs] Replica groups (#3511)
Co-authored-by: peterschmidt85 <a****v@g****m> |
Bihan Rana <s****n@g****m>
Committed by: GitHub <n****y@g****m> |
4 months ago | |
| 763092d9 | Fix scaling during update to replica groups (#3510) |
jvstme <3****e@u****m>
Committed by: GitHub <n****y@g****m> |
4 months ago | |
| 7ba2f3c1 | Fix `dstack event` compat. with older servers (#3509) |
jvstme <3****e@u****m>
Committed by: GitHub <n****y@g****m> |
4 months ago | |
| 6da15540 | [UI] Minor tweaks (#3508) |
Andrey Cheptsov <5****5@u****m>
Committed by: GitHub <n****y@g****m> |
4 months ago | |
| b7f637b1 | Fix apply plan compatibility with old servers (#3507) |
jvstme <3****e@u****m>
Committed by: GitHub <n****y@g****m> |
4 months ago | |
| bef08a59 | Add secret lifecycle events (#3505) |
jvstme <3****e@u****m>
Committed by: GitHub <n****y@g****m> |
4 months ago | |
| a1f0d58d | [UX] Extend `dstack login` with interactive selection of `url` and default pr... |
Andrey Cheptsov <5****5@u****m>
Committed by: GitHub <n****y@g****m> |
4 months ago | |
| ca56b3b9 |
[Docs] Events #3397 (#3506)
Co-authored-by: jvstme <3****e@u****m>, jvstme <3****e@u****m>, jvstme <3****e@u****m>, jvstme <3****e@u****m>, jvstme <3****e@u****m>, jvstme <3****e@u****m>, jvstme <3****e@u****m> |
Andrey Cheptsov <5****5@u****m>
Committed by: GitHub <n****y@g****m> |
4 months ago | |
| b93476e1 | Support secret events in API, CLI, and UI (#3504) |
jvstme <3****e@u****m>
Committed by: GitHub <n****y@g****m> |
4 months ago | |
| dde4a070 | Docs minor improvements (#3501) |
Andrey Cheptsov <5****5@u****m>
Committed by: GitHub <n****y@g****m> |
4 months ago | |
| b4c6f178 | Add gateway lifecycle events (#3500) |
jvstme <3****e@u****m>
Committed by: GitHub <n****y@g****m> |
4 months ago | |
| 5efca70e | Use numeric replica-group names (#3502) |
Bihan Rana <s****n@g****m>
Committed by: GitHub <n****y@g****m> |
4 months ago | |
| b4b69fb4 | Support gateway events in API, CLI, and UI (#3499) |
jvstme <3****e@u****m>
Committed by: GitHub <n****y@g****m> |
4 months ago | |
| 19853410 | Set JobTerminationReason.INSTANCE_UNREACHABLE for unreachable on-demand insta... |
Victor Skvortsov <v****3@g****m>
Committed by: GitHub <n****y@g****m> |
4 months ago | |
| 90b05792 | Volume events (#3494) |
jvstme <3****e@u****m>
Committed by: GitHub <n****y@g****m> |
4 months ago | |
| ca2172cd | Events: instance/job reachability and health (#3482) |
jvstme <3****e@u****m>
Committed by: GitHub <n****y@g****m> |
4 months ago | |
| 883b4555 | Move ruff.toml to pyproject.toml (#3496) |
Victor Skvortsov <v****3@g****m>
Committed by: GitHub <n****y@g****m> |
4 months ago | |
| 5205c891 | [Docs]: Fix k8s backend config example (#3495) |
jvstme <3****e@u****m>
Committed by: GitHub <n****y@g****m> |
4 months ago | |
| be788642 | [chore]: Add `list_events` utility for unit tests (#3493) |
jvstme <3****e@u****m>
Committed by: GitHub <n****y@g****m> |
4 months ago | |
| bd2d485f | Add replica groups in dstack-service (#3408) |
Bihan Rana <s****n@g****m>
Committed by: GitHub <n****y@g****m> |
4 months ago | |
| 802c450a | [UX] Make `dstack project` and `dstack project set-default` interactive for d... |
Andrey Cheptsov <5****5@u****m>
Committed by: GitHub <n****y@g****m> |
4 months ago | |
| 330eb1f0 | Move pytest.ini options to pyproject.toml (#3491) |
Victor Skvortsov <v****3@g****m>
Committed by: GitHub <n****y@g****m> |
4 months ago | |
| 1c3b7f80 | Update dstack server CLI logo (#3438) |
Victor Skvortsov <v****3@g****m>
Committed by: GitHub <n****y@g****m> |
4 months ago | |
| d9598289 | Implement pagination for `/api/project/list` and `/api/users/list` (#3489) |
Victor Skvortsov <v****3@g****m>
Committed by: GitHub <n****y@g****m> |
4 months ago | |
| 28797f65 | Fix CI errors | Oleg Vavilov <v****k@g****m> | 4 months ago | |
| d3beebc9 | Added react rules for eslint | Oleg Vavilov <v****k@g****m> | 4 months ago | |
| 2ad526cd | Small fix | Oleg Vavilov <v****k@g****m> | 4 months ago | |
| f09d0618 | Hotfix. Fixed generation fleet fields in project forms (#3486) |
Oleg <v****k@g****m>
Committed by: GitHub <n****y@g****m> |
4 months ago | |
| 6d14aadc | Add missing Box imports (#3485) |
Victor Skvortsov <v****3@g****m>
Committed by: GitHub <n****y@g****m> |
4 months ago | |
| 32fbc028 | [UI] Minor re-order in the sidebar (#3484) |
Andrey Cheptsov <5****5@u****m>
Committed by: GitHub <n****y@g****m> |
4 months ago | |
| 19645913 | Support shared AWS compute caches (#3483) |
Victor Skvortsov <v****3@g****m>
Committed by: GitHub <n****y@g****m> |
4 months ago | |
| 65eacc77 |
[UI] Default fleet in project wizard (#3464)
Co-authored-by: peterschmidt85 <a****v@g****m> |
Oleg <v****k@g****m>
Committed by: GitHub <n****y@g****m> |
4 months ago | |
| c01b022a | [runner] Restore `--home-dir` option as no-op (#3480) |
Dmitry Meyer <m****e@u****m>
Committed by: GitHub <n****y@g****m> |
4 months ago | |
| 54d2d0aa | Emit events for instance status changes (#3477) |
jvstme <3****e@u****m>
Committed by: GitHub <n****y@g****m> |
4 months ago | |
| 29076ba1 | Adjust fluent-bit logging integration (#3478) |
Victor Skvortsov <v****3@g****m>
Committed by: GitHub <n****y@g****m> |
4 months ago | |
| 811643f5 | feat(logging): add fluent-bit log shipping (#3431) |
Alexander <4****f@u****m>
Committed by: GitHub <n****y@g****m> |
4 months ago | |
| a07ef352 | Optimize list and get fleets (#3472) |
Victor Skvortsov <v****3@g****m>
Committed by: GitHub <n****y@g****m> |
5 months ago | |
| 628bb8b2 | Fix missing instance lock in delete_fleets (#3471) |
Victor Skvortsov <v****3@g****m>
Committed by: GitHub <n****y@g****m> |
5 months ago | |
| 104834e2 | Fix `find_optimal_fleet_with_offers` log message (#3470) |
Dmitry Meyer <m****e@u****m>
Committed by: GitHub <n****y@g****m> |
5 months ago | |
| 395ccb75 | Add missing job status change event for scaling (#3465) |
jvstme <3****e@u****m>
Committed by: GitHub <n****y@g****m> |
5 months ago | |
| 71c12ad7 | Kubernetes: adjust offer GPU count (#3469) |
Dmitry Meyer <m****e@u****m>
Committed by: GitHub <n****y@g****m> |
5 months ago | |
| d0b4cc3d | Optimize fleet instances db queries (#3467) |
Victor Skvortsov <v****3@g****m>
Committed by: GitHub <n****y@g****m> |
5 months ago | |
| a26c67b3 | [runner] Rework and fix user processing (#3456) |
Dmitry Meyer <m****e@u****m>
Committed by: GitHub <n****y@g****m> |
5 months ago | |
| 8b383ba6 | [CLI] Add `--memory` option to `apply` and `offer` (#3461) |
Dmitry Meyer <m****e@u****m>
Committed by: GitHub <n****y@g****m> |
5 months ago | |
| ad6423df | Optimize job submissions loading (#3466) |
Victor Skvortsov <v****3@g****m>
Committed by: GitHub <n****y@g****m> |
5 months ago | |
| 4432cdfe | Do not return `NO_BALANCE` to older clients (#3462) |
jvstme <3****e@u****m>
Committed by: GitHub <n****y@g****m> |
5 months ago | |
| c90cdf10 | Display `InstanceAvailability.NO_BALANCE` in CLI (#3460) |
jvstme <3****e@u****m>
Committed by: GitHub <n****y@g****m> |
5 months ago | |
| a36577f9 | [Internal]: Handle GitHub API errors in `release_notes.py` (#3463) |
jvstme <3****e@u****m>
Committed by: GitHub <n****y@g****m> |
5 months ago | |
| 22296d6e | Linter fix | peterschmidt85 <a****v@g****m> | 5 months ago | |
| 9e5b3b32 | Migrate from Slurm (#3454) |
Andrey Cheptsov <5****5@u****m>
Committed by: GitHub <n****y@g****m> |
5 months ago | |
| fae73ce0 | Refactoring Inspect page (#3457) |
Oleg <v****k@g****m>
Committed by: GitHub <n****y@g****m> |
5 months ago | |
| 3be819be | Use the same metrics endpoint label for 404 requests (#3455) |
Victor Skvortsov <v****3@g****m>
Committed by: GitHub <n****y@g****m> |
5 months ago | |
| d48b15fb | [Feature] Allow to see JSON state of runs/volumes/fleets/gateways via CLI/UI ... |
Andrey Cheptsov <5****5@u****m>
Committed by: GitHub <n****y@g****m> |
5 months ago | |
| 2a4c0e17 | [runner] Decouple Server and Executor (#3447) |
Dmitry Meyer <m****e@u****m>
Committed by: GitHub <n****y@g****m> |
5 months ago | |
| dd907bf1 | Add `processing instance` debug log message (#3450) |
jvstme <3****e@u****m>
Committed by: GitHub <n****y@g****m> |
5 months ago | |
| d4680c99 | [Dev environments] Support windsurf IDE (#3444) |
Andrey Cheptsov <5****5@u****m>
Committed by: GitHub <n****y@g****m> |
5 months ago | |
| f174dff2 | [Crusoe] Minor edits (#3448) |
Andrey Cheptsov <5****5@u****m>
Committed by: GitHub <n****y@g****m> |
5 months ago | |
| 78d26ee2 | [runner] Fix MPI hostfile (#3441) |
Dmitry Meyer <m****e@u****m>
Committed by: GitHub <n****y@g****m> |
5 months ago | |
| b78851b3 | Adjust kubernetes gpu matching for RTX5090 (#3440) |
Victor Skvortsov <v****3@g****m>
Committed by: GitHub <n****y@g****m> |
5 months ago | |
| 3e931d90 | Make no fleet notifications dismissible (#3439) |
Victor Skvortsov <v****3@g****m>
Committed by: GitHub <n****y@g****m> |
5 months ago | |
| 6adffcae | [UX] Add an API that returns projects that lack active fleets (#3425) |
Andrey Cheptsov <5****5@u****m>
Committed by: GitHub <n****y@g****m> |
5 months ago | |
| 4893cb30 | Change /dstack/venv ownership to the current user (#3437) |
Dmitry Meyer <m****e@u****m>
Committed by: GitHub <n****y@g****m> |
5 months ago | |
| dde66b56 | [runner] Streamline authorized_keys management (#3435) |
Dmitry Meyer <m****e@u****m>
Committed by: GitHub <n****y@g****m> |
5 months ago | |
| d472448a | [UX] Better "No fleets" messages; plus updated `Troubleshooting` guide (#3428) |
Andrey Cheptsov <5****5@u****m>
Committed by: GitHub <n****y@g****m> |
5 months ago | |
| 0697f2d3 | Remove httpx duplicated in dev deps (#3433) |
Victor Skvortsov <v****3@g****m>
Committed by: GitHub <n****y@g****m> |
5 months ago | |
| f9763555 | [shim] Fix DockerRunner tests (#3429) |
Dmitry Meyer <m****e@u****m>
Committed by: GitHub <n****y@g****m> |
5 months ago | |
| ef0d8a73 | Resolve url for dstack login (#3427) |
Victor Skvortsov <v****3@g****m>
Committed by: GitHub <n****y@g****m> |
5 months ago | |
| de7170ba | Updated README.md | peterschmidt85 <a****v@g****m> | 5 months ago |