feat: Improve client perf and error handling#247
Open
LukeAVanDrie wants to merge 1 commit intokubernetes-sigs:mainfrom
Open
feat: Improve client perf and error handling#247LukeAVanDrie wants to merge 1 commit intokubernetes-sigs:mainfrom
LukeAVanDrie wants to merge 1 commit intokubernetes-sigs:mainfrom
Conversation
Reuses aiohttp.ClientSession across requests in openAIModelServerClient to reduce connection overhead. This change improves client-side throughput and latency. Additional improvements: - Refines error handling to distinguish between network errors (like aiohttp.ClientError), non-200 HTTP status codes, and errors during response processing. - Ensures non-200 responses with text bodies are captured. - Guarantees response body is always consumed to release connections.
k8s-ci-robot
added
the
cncf-cla: yes
Contributor
|
[APPROVALNOTIFIER] This PR is NOT APPROVED This pull-request has been approved by: LukeAVanDrie The full list of commands accepted by this bot can be found here. DetailsNeeds approval from an approver in each of these files:Approvers can indicate their approval by writing |
k8s-ci-robot
added
the
size/M
jjk-g
reviewed
Oct 7, 2025
| ) | ||
| ) | ||
|
|
||
| end_time = time.perf_counter() |
jjk-g
reviewed
Oct 7, 2025
Collaborator
jjk-g
left a comment
jjk-g
left a comment
There was a problem hiding this comment.
Thanks for adding! One nit
/lgtm
k8s-ci-robot
added
the
lgtm
Contributor
achandrasekar
left a comment
achandrasekar
left a comment
There was a problem hiding this comment.
Can you add how the change was tested and if you have any numbers on improvements that'd be great too?
Contributor
|
Please address the linting and type check issue above |
k8s-ci-robot
added
the
needs-rebase
Contributor
|
PR needs rebase. DetailsInstructions for interacting with me using PR comments are available here. If you have questions or suggestions related to my behavior, please file an issue against the kubernetes-sigs/prow repository. |
Collaborator
|
@LukeAVanDrie friendly ping for linting and type check errors |
| elif not tokenizer_config: | ||
| tokenizer_config = CustomTokenizerConfig(pretrained_model_name_or_path=self.model_name) | ||
| self.tokenizer = CustomTokenizer(tokenizer_config) | ||
| self.session = aiohttp.ClientSession(connector=aiohttp.TCPConnector(limit=self.max_tcp_connections)) |
Contributor
There was a problem hiding this comment.
Please correct me if I'm wrong, but isn't openAIModelServerClient shared across multiple asyncio event loops because of multiprocessing? Creating a single ClientSession here might cause issues if the same instance is also being shared to all the multiprocessing workers.
Relevant link: https://stackoverflow.com/questions/62707369/one-aiohttp-clientsession-per-thread
diamondburned
added a commit
to diamondburned/inference-perf
that referenced
this pull request
Nov 11, 2025
Slightly refactor `openAIModelServerClient` to accept a custom
`aiohttp.ClientSession` per request, which allows us to use exactly 1
client session per worker.
Prior to this commit, a new `aiohttp.ClientSession` is created for each
request. Not only is this inefficient and lowers throughput, on certain
environments, it also leads to inotify watch issues:
aiodns - WARNING - Failed to create DNS resolver channel with
automatic monitoring of resolver configuration changes. This usually
means the system ran out of inotify watches. Falling back to socket
state callback. Consider increasing the system inotify watch limit:
Failed to initialize c-ares channel
Indeed, because each DNS resolver is created for a new `ClientSession`,
creating tons of new `ClientSession`s causes eventual inotify watch
exhaustion. Sharing `ClientSession`s solves this issue.
Relevant links:
- https://docs.aiohttp.org/en/stable/http_request_lifecycle.html
- https://stackoverflow.com/questions/62707369/one-aiohttp-clientsession-per-thread
- home-assistant/core#144457 (comment)
Relevant PR: kubernetes-sigs#247
(doesn't address the issue of worker sharing).
diamondburned
added a commit
to diamondburned/inference-perf
that referenced
this pull request
Nov 12, 2025
Slightly refactor `openAIModelServerClient` to accept a custom
`aiohttp.ClientSession` per request, which allows us to use exactly 1
client session per worker.
Prior to this commit, a new `aiohttp.ClientSession` is created for each
request. Not only is this inefficient and lowers throughput, on certain
environments, it also leads to inotify watch issues:
aiodns - WARNING - Failed to create DNS resolver channel with
automatic monitoring of resolver configuration changes. This usually
means the system ran out of inotify watches. Falling back to socket
state callback. Consider increasing the system inotify watch limit:
Failed to initialize c-ares channel
Indeed, because each DNS resolver is created for a new `ClientSession`,
creating tons of new `ClientSession`s causes eventual inotify watch
exhaustion. Sharing `ClientSession`s solves this issue.
Relevant links:
- https://docs.aiohttp.org/en/stable/http_request_lifecycle.html
- https://stackoverflow.com/questions/62707369/one-aiohttp-clientsession-per-thread
- home-assistant/core#144457 (comment)
Relevant PR: kubernetes-sigs#247
(doesn't address the issue of worker sharing).
diamondburned
added a commit
to diamondburned/inference-perf
that referenced
this pull request
Nov 13, 2025
Slightly refactor `openAIModelServerClient` to accept a custom
`aiohttp.ClientSession` per request, which allows us to use exactly 1
client session per worker.
Prior to this commit, a new `aiohttp.ClientSession` is created for each
request. Not only is this inefficient and lowers throughput, on certain
environments, it also leads to inotify watch issues:
aiodns - WARNING - Failed to create DNS resolver channel with
automatic monitoring of resolver configuration changes. This usually
means the system ran out of inotify watches. Falling back to socket
state callback. Consider increasing the system inotify watch limit:
Failed to initialize c-ares channel
Indeed, because each DNS resolver is created for a new `ClientSession`,
creating tons of new `ClientSession`s causes eventual inotify watch
exhaustion. Sharing `ClientSession`s solves this issue.
Relevant links:
- https://docs.aiohttp.org/en/stable/http_request_lifecycle.html
- https://stackoverflow.com/questions/62707369/one-aiohttp-clientsession-per-thread
- home-assistant/core#144457 (comment)
Relevant PR: kubernetes-sigs#247
(doesn't address the issue of worker sharing).
diamondburned
added a commit
to diamondburned/inference-perf
that referenced
this pull request
Nov 17, 2025
Slightly refactor `openAIModelServerClient` to accept a custom
`aiohttp.ClientSession` per request, which allows us to use exactly 1
client session per worker.
Prior to this commit, a new `aiohttp.ClientSession` is created for each
request. Not only is this inefficient and lowers throughput, on certain
environments, it also leads to inotify watch issues:
aiodns - WARNING - Failed to create DNS resolver channel with
automatic monitoring of resolver configuration changes. This usually
means the system ran out of inotify watches. Falling back to socket
state callback. Consider increasing the system inotify watch limit:
Failed to initialize c-ares channel
Indeed, because each DNS resolver is created for a new `ClientSession`,
creating tons of new `ClientSession`s causes eventual inotify watch
exhaustion. Sharing `ClientSession`s solves this issue.
Relevant links:
- https://docs.aiohttp.org/en/stable/http_request_lifecycle.html
- https://stackoverflow.com/questions/62707369/one-aiohttp-clientsession-per-thread
- home-assistant/core#144457 (comment)
Relevant PR: kubernetes-sigs#247
(doesn't address the issue of worker sharing).
diamondburned
added a commit
to diamondburned/inference-perf
that referenced
this pull request
Nov 17, 2025
Slightly refactor `openAIModelServerClient` to add a new method,
`process_request_with_session`, that accepts a custom
`ReusableHTTPClientSession` per request, which allows the caller to
reuse an HTTP client session per worker.
The previous method, `process_request`, is made to create a fresh HTTP
client session then call `process_request_with_session`, preserving the
previous behavior.
Prior to this commit, a new `aiohttp.ClientSession` is created for each
request. Not only is this inefficient and lowers throughput, on certain
environments, it also leads to inotify watch issues:
aiodns - WARNING - Failed to create DNS resolver channel with
automatic monitoring of resolver configuration changes. This usually
means the system ran out of inotify watches. Falling back to socket
state callback. Consider increasing the system inotify watch limit:
Failed to initialize c-ares channel
Indeed, because each DNS resolver is created for a new `ClientSession`,
creating tons of new `ClientSession`s causes eventual inotify watch
exhaustion. Sharing `ClientSession`s solves this issue.
Relevant links:
- https://docs.aiohttp.org/en/stable/http_request_lifecycle.html
- https://stackoverflow.com/questions/62707369/one-aiohttp-clientsession-per-thread
- home-assistant/core#144457 (comment)
Relevant PR: kubernetes-sigs#247
(doesn't address the issue of worker sharing).
diamondburned
added a commit
to diamondburned/inference-perf
that referenced
this pull request
Nov 17, 2025
Slightly refactor `openAIModelServerClient` to add a new method,
`process_request_with_session`, that accepts a custom
`ReusableHTTPClientSession` per request, which allows the caller to
reuse an HTTP client session per worker.
The previous method, `process_request`, is made to create a fresh HTTP
client session then call `process_request_with_session`, preserving the
previous behavior.
Prior to this commit, a new `aiohttp.ClientSession` is created for each
request. Not only is this inefficient and lowers throughput, on certain
environments, it also leads to inotify watch issues:
aiodns - WARNING - Failed to create DNS resolver channel with
automatic monitoring of resolver configuration changes. This usually
means the system ran out of inotify watches. Falling back to socket
state callback. Consider increasing the system inotify watch limit:
Failed to initialize c-ares channel
Indeed, because each DNS resolver is created for a new `ClientSession`,
creating tons of new `ClientSession`s causes eventual inotify watch
exhaustion. Sharing `ClientSession`s solves this issue.
Relevant links:
- https://docs.aiohttp.org/en/stable/http_request_lifecycle.html
- https://stackoverflow.com/questions/62707369/one-aiohttp-clientsession-per-thread
- home-assistant/core#144457 (comment)
Relevant PR: kubernetes-sigs#247
(doesn't address the issue of worker sharing).
diamondburned
added a commit
to diamondburned/inference-perf
that referenced
this pull request
Nov 20, 2025
Slightly refactor `openAIModelServerClient` to add a new method,
`process_request_with_session`, that accepts a custom
`ReusableHTTPClientSession` per request, which allows the caller to
reuse an HTTP client session per worker.
The previous method, `process_request`, is made to create a fresh HTTP
client session then call `process_request_with_session`, preserving the
previous behavior.
Prior to this commit, a new `aiohttp.ClientSession` is created for each
request. Not only is this inefficient and lowers throughput, on certain
environments, it also leads to inotify watch issues:
aiodns - WARNING - Failed to create DNS resolver channel with
automatic monitoring of resolver configuration changes. This usually
means the system ran out of inotify watches. Falling back to socket
state callback. Consider increasing the system inotify watch limit:
Failed to initialize c-ares channel
Indeed, because each DNS resolver is created for a new `ClientSession`,
creating tons of new `ClientSession`s causes eventual inotify watch
exhaustion. Sharing `ClientSession`s solves this issue.
Relevant links:
- https://docs.aiohttp.org/en/stable/http_request_lifecycle.html
- https://stackoverflow.com/questions/62707369/one-aiohttp-clientsession-per-thread
- home-assistant/core#144457 (comment)
Relevant PR: kubernetes-sigs#247
(doesn't address the issue of worker sharing).
diamondburned
added a commit
to diamondburned/inference-perf
that referenced
this pull request
Nov 20, 2025
Slightly refactor `openAIModelServerClient` to add a new method,
`process_request_with_session`, that accepts a custom
`ReusableHTTPClientSession` per request, which allows the caller to
reuse an HTTP client session per worker.
The previous method, `process_request`, is made to create a fresh HTTP
client session then call `process_request_with_session`, preserving the
previous behavior.
Prior to this commit, a new `aiohttp.ClientSession` is created for each
request. Not only is this inefficient and lowers throughput, on certain
environments, it also leads to inotify watch issues:
aiodns - WARNING - Failed to create DNS resolver channel with
automatic monitoring of resolver configuration changes. This usually
means the system ran out of inotify watches. Falling back to socket
state callback. Consider increasing the system inotify watch limit:
Failed to initialize c-ares channel
Indeed, because each DNS resolver is created for a new `ClientSession`,
creating tons of new `ClientSession`s causes eventual inotify watch
exhaustion. Sharing `ClientSession`s solves this issue.
Relevant links:
- https://docs.aiohttp.org/en/stable/http_request_lifecycle.html
- https://stackoverflow.com/questions/62707369/one-aiohttp-clientsession-per-thread
- home-assistant/core#144457 (comment)
Relevant PR: kubernetes-sigs#247
(doesn't address the issue of worker sharing).
Contributor
|
@LukeAVanDrie Thanks for the contribution. Can you please rebase this PR? |
Contributor
Author
Yes, apologies for the long delay here. I will make sure to update the description with my testing results and verify @diamondburned's concern regarding multiprocessing. |
diamondburned
added a commit
to diamondburned/inference-perf
that referenced
this pull request
Nov 21, 2025
Slightly refactor `openAIModelServerClient` to add a new method,
`process_request_with_session`, that accepts a custom
`ReusableHTTPClientSession` per request, which allows the caller to
reuse an HTTP client session per worker.
The previous method, `process_request`, is made to create a fresh HTTP
client session then call `process_request_with_session`, preserving the
previous behavior.
Prior to this commit, a new `aiohttp.ClientSession` is created for each
request. Not only is this inefficient and lowers throughput, on certain
environments, it also leads to inotify watch issues:
aiodns - WARNING - Failed to create DNS resolver channel with
automatic monitoring of resolver configuration changes. This usually
means the system ran out of inotify watches. Falling back to socket
state callback. Consider increasing the system inotify watch limit:
Failed to initialize c-ares channel
Indeed, because each DNS resolver is created for a new `ClientSession`,
creating tons of new `ClientSession`s causes eventual inotify watch
exhaustion. Sharing `ClientSession`s solves this issue.
Relevant links:
- https://docs.aiohttp.org/en/stable/http_request_lifecycle.html
- https://stackoverflow.com/questions/62707369/one-aiohttp-clientsession-per-thread
- home-assistant/core#144457 (comment)
Relevant PR: kubernetes-sigs#247
(doesn't address the issue of worker sharing).
diamondburned
added a commit
to diamondburned/inference-perf
that referenced
this pull request
Nov 26, 2025
Slightly refactor `openAIModelServerClient` to add a new method,
`process_request_with_session`, that accepts a custom
`ReusableHTTPClientSession` per request, which allows the caller to
reuse an HTTP client session per worker.
The previous method, `process_request`, is made to create a fresh HTTP
client session then call `process_request_with_session`, preserving the
previous behavior.
Prior to this commit, a new `aiohttp.ClientSession` is created for each
request. Not only is this inefficient and lowers throughput, on certain
environments, it also leads to inotify watch issues:
aiodns - WARNING - Failed to create DNS resolver channel with
automatic monitoring of resolver configuration changes. This usually
means the system ran out of inotify watches. Falling back to socket
state callback. Consider increasing the system inotify watch limit:
Failed to initialize c-ares channel
Indeed, because each DNS resolver is created for a new `ClientSession`,
creating tons of new `ClientSession`s causes eventual inotify watch
exhaustion. Sharing `ClientSession`s solves this issue.
Relevant links:
- https://docs.aiohttp.org/en/stable/http_request_lifecycle.html
- https://stackoverflow.com/questions/62707369/one-aiohttp-clientsession-per-thread
- home-assistant/core#144457 (comment)
Relevant PR: kubernetes-sigs#247
(doesn't address the issue of worker sharing).
diamondburned
added a commit
to diamondburned/inference-perf
that referenced
this pull request
Nov 26, 2025
Slightly refactor `openAIModelServerClient` to add a new method,
`process_request_with_session`, that accepts a custom
`ReusableHTTPClientSession` per request, which allows the caller to
reuse an HTTP client session per worker.
The previous method, `process_request`, is made to create a fresh HTTP
client session then call `process_request_with_session`, preserving the
previous behavior.
Prior to this commit, a new `aiohttp.ClientSession` is created for each
request. Not only is this inefficient and lowers throughput, on certain
environments, it also leads to inotify watch issues:
aiodns - WARNING - Failed to create DNS resolver channel with
automatic monitoring of resolver configuration changes. This usually
means the system ran out of inotify watches. Falling back to socket
state callback. Consider increasing the system inotify watch limit:
Failed to initialize c-ares channel
Indeed, because each DNS resolver is created for a new `ClientSession`,
creating tons of new `ClientSession`s causes eventual inotify watch
exhaustion. Sharing `ClientSession`s solves this issue.
Relevant links:
- https://docs.aiohttp.org/en/stable/http_request_lifecycle.html
- https://stackoverflow.com/questions/62707369/one-aiohttp-clientsession-per-thread
- home-assistant/core#144457 (comment)
Relevant PR: kubernetes-sigs#247
(doesn't address the issue of worker sharing).
jjk-g
pushed a commit
that referenced
this pull request
Nov 26, 2025
Slightly refactor `openAIModelServerClient` to add a new method,
`process_request_with_session`, that accepts a custom
`ReusableHTTPClientSession` per request, which allows the caller to
reuse an HTTP client session per worker.
The previous method, `process_request`, is made to create a fresh HTTP
client session then call `process_request_with_session`, preserving the
previous behavior.
Prior to this commit, a new `aiohttp.ClientSession` is created for each
request. Not only is this inefficient and lowers throughput, on certain
environments, it also leads to inotify watch issues:
aiodns - WARNING - Failed to create DNS resolver channel with
automatic monitoring of resolver configuration changes. This usually
means the system ran out of inotify watches. Falling back to socket
state callback. Consider increasing the system inotify watch limit:
Failed to initialize c-ares channel
Indeed, because each DNS resolver is created for a new `ClientSession`,
creating tons of new `ClientSession`s causes eventual inotify watch
exhaustion. Sharing `ClientSession`s solves this issue.
Relevant links:
- https://docs.aiohttp.org/en/stable/http_request_lifecycle.html
-
https://stackoverflow.com/questions/62707369/one-aiohttp-clientsession-per-thread
-
home-assistant/core#144457 (comment)
Relevant PR: #247
(doesn't address the issue of worker sharing).
Collaborator
|
@LukeAVanDrie any updates on this PR? |
This file contains hidden or 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
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
6 participants
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.
Reuses aiohttp.ClientSession across requests in openAIModelServerClient to reduce connection overhead. This change improves client-side throughput and latency.
Additional improvements: