-
Notifications
You must be signed in to change notification settings - Fork 5.7k
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
[Serve] [Docs] Update the "Monitoring Ray Serve" Page #27777
[Serve] [Docs] Update the "Monitoring Ray Serve" Page #27777
Conversation
Signed-off-by: Shreyas Krishnaswamy <[email protected]>
Signed-off-by: Shreyas Krishnaswamy <[email protected]>
Signed-off-by: Shreyas Krishnaswamy <[email protected]>
Signed-off-by: Shreyas Krishnaswamy <[email protected]>
Signed-off-by: Shreyas Krishnaswamy <[email protected]>
Signed-off-by: Shreyas Krishnaswamy <[email protected]>
Signed-off-by: Shreyas Krishnaswamy <[email protected]>
Signed-off-by: Shreyas Krishnaswamy <[email protected]>
Signed-off-by: Shreyas Krishnaswamy <[email protected]>
Signed-off-by: Shreyas Krishnaswamy <[email protected]>
Signed-off-by: Shreyas Krishnaswamy <[email protected]>
Signed-off-by: Shreyas Krishnaswamy <[email protected]>
Signed-off-by: Shreyas Krishnaswamy <[email protected]>
@shrekris-anyscale FYI you have a merge conflict here |
Signed-off-by: Shreyas Krishnaswamy <[email protected]>
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Thanks for adding the serve.run
changes, those will be very helpful.
doc/source/serve/monitoring.md
Outdated
This includes details such as: | ||
* the number of deployment replicas currently running | ||
* logs for your Serve controller, deployment replicas, and HTTP proxies | ||
* the Ray nodes (i.e. machines) running in your Ray cluster |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
period to end the last bullet point
doc/source/serve/monitoring.md
Outdated
For example, if you're running Ray Serve on a local Ray cluster, you can access the dashboard by going to this address in your browser: | ||
|
||
``` | ||
https://localhost:8265 | ||
``` |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
For example, if you're running Ray Serve on a local Ray cluster, you can access the dashboard by going to this address in your browser: | |
``` | |
https://localhost:8265 | |
``` | |
For example, if you're running Ray Serve locally, you can access the dashboard by going to `https://localhost:8265` in your browser. |
save some page real estate
doc/source/serve/monitoring.md
Outdated
We can view important information about our application here. | ||
For example, we can inspect our deployment replicas by navigating to the Ray dashboard's "Actors" tab while our Serve application is running: |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
the previous paragraph used "you" and now you're using "we," stick with one
Signed-off-by: Shreyas Krishnaswamy <[email protected]>
@edoakes Thanks for the review! I addressed your comments. |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
With those changes, it should be good to go.
Just a general comment. If possible for 2.1, we should consider providing dashboards for this; there is too much boiler code they have to write. But at least we are collecting stats; matter of accessing and presenting easily.
doc/source/serve/monitoring.md
Outdated
This section should help you understand how to debug and monitor your Serve application. | ||
There are three main ways to do this: | ||
Using the Ray Dashboard, using Ray logging, and using built-in Ray Serve metrics. | ||
This section should help you debug and monitor your Serve applications by: |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
This section helps you debug and monitor your Serve applications by:
## Ray logging | ||
|
||
You can use Ray logging to understand system-level behavior and to surface application-level details during runtime. | ||
|
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
We repeatedly begin a sentence with "You can use XXX..." Perhaps change the pattern: may be use an infinitive phrase or participle phrase :
To comprehend system-level behavior and surface application-level details during runtime, use Ray logging.
doc/source/serve/monitoring.md
Outdated
``` | ||
|
||
Running this code block, we first get some log messages from the controller saying that a new replica of the deployment is being created: | ||
We can run this deployment using the `serve run` CLI command: |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Run this deployment using .....
doc/source/serve/monitoring.md
Outdated
``` | ||
|
||
Then when we query the deployment, we get both a default access log as well as our custom `"Hello world!"` message. | ||
`serve run` prints out a few log messages immediately. Note that a few of these messages start with identifiers such as |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
"out" may be redundant. This without "out" reads equally well.
prints a few log messages ....
@@ -82,18 +111,21 @@ class Silenced: | |||
logger.setLevel(logging.ERROR) | |||
``` | |||
|
|||
This will prevent the replica INFO-level logs from being written to STDOUT or to files on disk. | |||
This controls which logs are written to STDOUT or files on disk. | |||
You can also use your own custom logger, in which case you'll need to configure the behavior to write to STDOUT/STDERR, files on disk, or both. | |||
|
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Apart from standard Python logger, Serve supports custom logging. Custom logging via configuration can control the behavior of what messages are written to STDOUT/STDERR, files on disk, or both.
First, install Loki and Promtail using the instructions on <https://grafana.com>. | ||
It will be convenient to save the Loki and Promtail executables in the same directory, and to navigate to this directory in your terminal before beginning this walkthrough. | ||
You can explore and filter your logs using [Loki](https://grafana.com/oss/loki/). | ||
Setup and configuration is straightforward on Kubernetes, but as a tutorial, let's set up Loki manually. | ||
|
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Is "setup and configuration" considered a singular compound subject (like ham and cheese is great for breakfast) in this context? If not perhaps change to
"Setup and configuration are straightforward on Kubernetes."
@@ -163,11 +198,14 @@ You should see something similar to the following: | |||
:align: center |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
passward "admin."
``` | ||
|
||
The requests will loop and can be canceled with `Ctrl-C`. | ||
The requests will loop and can be canceled with `ctrl-c`. | ||
|
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
The requests will loop until canceled with ctrl-c
.
Signed-off-by: Shreyas Krishnaswamy <[email protected]>
Thanks for the review @dmatrix! I've addressed all your comments.
Yeah that's a good suggestion. I think that would improve Ray Serve's observability. |
) The "Monitoring Ray Serve" page explains how to inspect your Ray Serve applications. This change updates the page to remove outdated metrics that Serve no longer exposes and to upgrade code samples to use 2.0 APIs. It also improves the content's readability and organization. Link to updated "Monitoring Ray Serve" page: https://ray--27777.org.readthedocs.build/en/27777/serve/monitoring.html
) The "Monitoring Ray Serve" page explains how to inspect your Ray Serve applications. This change updates the page to remove outdated metrics that Serve no longer exposes and to upgrade code samples to use 2.0 APIs. It also improves the content's readability and organization. Link to updated "Monitoring Ray Serve" page: https://ray--27777.org.readthedocs.build/en/27777/serve/monitoring.html Signed-off-by: Stefan van der Kleij <[email protected]>
) The "Monitoring Ray Serve" page explains how to inspect your Ray Serve applications. This change updates the page to remove outdated metrics that Serve no longer exposes and to upgrade code samples to use 2.0 APIs. It also improves the content's readability and organization. Link to updated "Monitoring Ray Serve" page: https://ray--27777.org.readthedocs.build/en/27777/serve/monitoring.html
) The "Monitoring Ray Serve" page explains how to inspect your Ray Serve applications. This change updates the page to remove outdated metrics that Serve no longer exposes and to upgrade code samples to use 2.0 APIs. It also improves the content's readability and organization. Link to updated "Monitoring Ray Serve" page: https://ray--27777.org.readthedocs.build/en/27777/serve/monitoring.html
) The "Monitoring Ray Serve" page explains how to inspect your Ray Serve applications. This change updates the page to remove outdated metrics that Serve no longer exposes and to upgrade code samples to use 2.0 APIs. It also improves the content's readability and organization. Link to updated "Monitoring Ray Serve" page: https://ray--27777.org.readthedocs.build/en/27777/serve/monitoring.html
) The "Monitoring Ray Serve" page explains how to inspect your Ray Serve applications. This change updates the page to remove outdated metrics that Serve no longer exposes and to upgrade code samples to use 2.0 APIs. It also improves the content's readability and organization. Link to updated "Monitoring Ray Serve" page: https://ray--27777.org.readthedocs.build/en/27777/serve/monitoring.html Signed-off-by: Artur Niederfahrenhorst <[email protected]>
Why are these changes needed?
The "Monitoring Ray Serve" page explains how to inspect your Ray Serve applications. This change updates the page to remove outdated metrics that Serve no longer exposes and to upgrade code samples to use 2.0 APIs. It also improves the content's readability and organization.
Link to updated "Monitoring Ray Serve" page: https://ray--27777.org.readthedocs.build/en/27777/serve/monitoring.html
Related issue number
Closes #27720.
Checks
git commit -s
) in this PR.scripts/format.sh
to lint the changes in this PR.