Commit graph

15 commits

Author SHA1 Message Date
Archit Kulkarni
df581c584a
[Job] [Dashboard] Add Job Submission data to cluster snapshot (#22225)
The existing Job info in the cluster snapshot uses the old definition of Job, which is a single Ray driver (a single `ray.init()` connection).  

In the new Job Submission protocol, a Job just specifies an entrypoint which can be any shell command.  As such a Job can have zero or multiple Ray drivers.  This means we should add a new snapshot entry corresponding to new jobs.  We'll leave the old snapshot in place for legacy jobs.

- Also fixes `get_all_jobs` by using the appropriate KV namespace, and stripping the job key KV prefix from the job ID.  It wasn't working before.

- This PR also unifies the datatype used by the GET jobs/ endpoint to be the same as the one used by the new jobs cluster snapshot.  For backwards compatibility, the `status` and `message` fields are preserved.
2022-02-18 09:54:37 -06:00
Edward Oakes
5df2a0a6c6
[jobs] Add test condition that job runs w/o CPUs available on head node (#22260) 2022-02-10 10:23:02 -06:00
Archit Kulkarni
50e2bef9d0
[Jobs] Hide dashboard from Job Submission import path (#22223)
For public SDK APIs, change the import path from 

```python
from ray.dashboard.modules.job.common import JobStatus, JobStatusInfo
from ray.dashboard.modules.job.sdk import JobSubmissionClient
```

to 
```python
from ray.job_submission import JobStatus, JobSubmissionClient
```

`JobStatus`, `JobStatusInfo` and `JobSubmissionClient` were the only names referenced in the SDK doc so far, but we can add more later as they appear.
2022-02-09 13:55:32 -06:00
Edward Oakes
8806b2d5c4
[jobs] Monitor jobs in the background to avoid requiring clients to poll (#22180) 2022-02-07 15:25:25 -06:00
Jiao
a692e7d05e
[jobs] Fix restarting local ray cluster with http ray address broke local job submission (#21938)
As titled. We have a corner case on user laptop where user might left RAY_ADDRESS as http address but restarted local ray cluster. In this case we will try to do job submission with an http prefixed address.

Co-authored-by: Archit Kulkarni <architkulkarni@users.noreply.github.com>
Co-authored-by: Jiao Dong <jiaodong@anyscale.com>
2022-02-04 17:51:43 -06:00
Balaji Veeramani
7f1bacc7dc
[CI] Format Python code with Black (#21975)
See #21316 and #21311 for the motivation behind these changes.
2022-01-29 18:41:57 -08:00
Archit Kulkarni
f058a1d342
[Jobs] Stream logs during job instead of only at the end (#21659)
Closes https://github.com/ray-project/ray/issues/21517
2022-01-20 15:21:07 -06:00
Yi Cheng
6dccfbffa9
Revert "Revert "[gcs] turn on grpc pubsub by default"" (#21585)
Reverts ray-project/ray#21584 and turn the flag off
2022-01-13 16:12:03 -08:00
Yi Cheng
bc696212d2
Revert "[gcs] turn on grpc pubsub by default" (#21584)
test-reconnect seems flaky.
Reverts ray-project/ray#21513
2022-01-13 12:34:02 -08:00
Yi Cheng
6194783312
[gcs] turn on grpc pubsub by default (#21513)
Turn on grpc pubsub by default.  This PR also fixed several tests which are failed before.

Co-authored-by: Mingwei Tian <mwtian@anyscale.com>
2022-01-12 22:13:03 -08:00
Jiao
ed34434131
[Jobs] Add log streaming for jobs (#20976)
Current logs API simply returns a str to unblock development and integration. We should add proper log streaming for better UX and external job manager integration.

Co-authored-by: Sven Mika <sven@anyscale.io>
Co-authored-by: sven1977 <svenmika1977@gmail.com>
Co-authored-by: Ed Oakes <ed.nmi.oakes@gmail.com>
Co-authored-by: Richard Liaw <rliaw@berkeley.edu>
Co-authored-by: Simon Mo <simon.mo@hey.com>
Co-authored-by: Avnish Narayan <38871737+avnishn@users.noreply.github.com>
Co-authored-by: Jiao Dong <jiaodong@anyscale.com>
2021-12-14 17:01:53 -08:00
Edward Oakes
d26c9e67e8
[job submission] Add a message to the JobStatus to return more detailed errors (#20491) 2021-11-18 10:15:23 -06:00
Edward Oakes
eae523159f
[job submission] Prefix job ID with raysubmit_ and pass job_name metadata (#20490) 2021-11-17 21:48:22 -06:00
shrekris-anyscale
c0aeb4a236
[runtime_env] Support working_dir and py_modules from HTTPS and Google Cloud Storage (#20280) 2021-11-14 02:16:45 -08:00
Edward Oakes
81f036d078
[job submission] Move job_manager to dashboard module, common parts to common.py (#20209) 2021-11-10 14:14:55 -08:00
Renamed from python/ray/tests/test_job_manager.py (Browse further)