[Serve] Fix serve non atomic shutdown #36927
Merged
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
…up and add tests Signed-off-by: Gene Su <e870252314@gmail.com>
Signed-off-by: Gene Su <e870252314@gmail.com>
Signed-off-by: Gene Su <e870252314@gmail.com>
Signed-off-by: Gene Su <e870252314@gmail.com>
Signed-off-by: Gene Su <e870252314@gmail.com>
Signed-off-by: Gene Su <e870252314@gmail.com>
Signed-off-by: Gene Su <e870252314@gmail.com>
Signed-off-by: Gene Su <e870252314@gmail.com>
GeneDer
force-pushed
the
fix-serve-non-atomic-shutdown
branch
from
8b82976 to
7f88970
Compare
GeneDer
commented
Jun 30, 2023
Signed-off-by: Gene Su <e870252314@gmail.com>
Signed-off-by: Gene Su <e870252314@gmail.com>
Signed-off-by: Gene Su <e870252314@gmail.com>
GeneDer
commented
Jun 30, 2023
Signed-off-by: Gene Su <e870252314@gmail.com>
Tested this locally by adding a long sleep in the client and deploy a test app.
Shut off the client with control+c twice
Go back to the dashboard and seeing serve shutdown and all the actors are in DEAD state
|
GeneDer
changed the title
Signed-off-by: Gene Su <e870252314@gmail.com>
There was a problem hiding this comment.
Could you parametrize this test, so it also runs with two SIGINT commands to make sure that the behavior is as expected when the user kills Serve before it shuts down correctly?
edoakes
reviewed
Jul 3, 2023
Signed-off-by: Gene Su <e870252314@gmail.com>
Signed-off-by: Gene Su <e870252314@gmail.com>
Signed-off-by: Gene Su <e870252314@gmail.com>
Signed-off-by: Gene Su <e870252314@gmail.com>
…tdown Signed-off-by: Gene Su <e870252314@gmail.com>
Signed-off-by: Gene Su <e870252314@gmail.com>
Signed-off-by: Gene Su <e870252314@gmail.com>
Signed-off-by: Gene Su <e870252314@gmail.com>
edoakes
reviewed
Jul 6, 2023
Signed-off-by: Gene Su <e870252314@gmail.com>
Signed-off-by: Gene Su <e870252314@gmail.com>
|
@edoakes just address all the comments. Hopefully all tests are still passing 🙏 |
edoakes
reviewed
Jul 7, 2023
… the actor is still alive Signed-off-by: Gene Su <e870252314@gmail.com>
edoakes
approved these changes
Jul 7, 2023
8 tasks
edoakes
pushed a commit
to edoakes/ray
that referenced
this pull request
Jul 7, 2023
Currently we are relying on the client to wait for all the resources before shutting off the controller. This caused the issue for when they interrupt the process and can cause incomplete shutdown. In this PR we moved the shutdown logic into the event loop which would be triggered by a `_shutting_down` flag on the controller. So even if the client interrupted the process, the controller will continue to shutdown all the resources and then kill itself. Signed-off-by: Edward Oakes <ed.nmi.oakes@gmail.com>
bveeramani
pushed a commit
that referenced
this pull request
Jul 7, 2023
Currently we are relying on the client to wait for all the resources before shutting off the controller. This caused the issue for when they interrupt the process and can cause incomplete shutdown. In this PR we moved the shutdown logic into the event loop which would be triggered by a `_shutting_down` flag on the controller. So even if the client interrupted the process, the controller will continue to shutdown all the resources and then kill itself. Signed-off-by: Edward Oakes <ed.nmi.oakes@gmail.com> Co-authored-by: Gene Der Su <e870252314@gmail.com>
8 tasks
arvind-chandra
pushed a commit
to lmco/ray
that referenced
this pull request
Aug 31, 2023
Currently we are relying on the client to wait for all the resources before shutting off the controller. This caused the issue for when they interrupt the process and can cause incomplete shutdown. In this PR we moved the shutdown logic into the event loop which would be triggered by a `_shutting_down` flag on the controller. So even if the client interrupted the process, the controller will continue to shutdown all the resources and then kill itself. Signed-off-by: e428265 <arvind.chandramouli@lmco.com>
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Why are these changes needed?
Currently we are relying on the client to wait for all the resources before shutting off the controller. This caused the issue for when they interrupt the process and can cause incomplete shutdown. In this PR we moved the shutdown logic into the event loop which would be triggered by a
_shutting_downflag on the controller. So even if the client interrupted the process, the controller will continue to shutdown all the resources and then kill itself.Related issue number
Closes: #36325
Checks
git commit -s) in this PR.scripts/format.shto lint the changes in this PR.method in Tune, I've added it in
doc/source/tune/api/under thecorresponding
.rstfile.