core:services:kraken: Fix container log stream timeout

[joaomariolago]

Closes #3763

Summary by Sourcery

Reduce the frequency of extension log synchronization by integrating it into the existing cleaner task loop instead of running a separate high-frequency background task.

Enhancements:

• Add a reusable helper method to perform extension log synchronization with error handling.
• Remove the dedicated extension log background task in favor of running log synchronization as part of the cleaner task cycle.

Summary by Sourcery

Allow configuring Docker client timeouts for Kraken harbor operations and use a non-timing-out client for container log streaming to prevent premature log stream termination.

Bug Fixes:

• Prevent container log streaming from timing out by using a Docker client configured with no timeout for log retrieval.

Enhancements:

• Extend the Docker context manager to support an optional custom HTTP client timeout via a dedicated aiohttp session.

[sourcery-ai]

Reviewer's guide (collapsed on small PRs)

Reviewer's Guide

Integrates a configurable-timeout Docker client context to prevent container log stream timeouts and uses it specifically for log retrieval, while keeping existing Docker usage unchanged elsewhere.

Sequence diagram for container log retrieval with non expiring timeout

sequenceDiagram
    actor Service
    participant ContainerModel
    participant DockerCtx
    participant Docker
    participant aiohttp_ClientSession
    participant aiohttp_ClientTimeout

    Service->>ContainerModel: get_container_log_by_name(container_name)
    activate ContainerModel
    ContainerModel->>DockerCtx: __init__(timeout=0)
    activate DockerCtx
    DockerCtx->>Docker: Docker(session=True)
    activate Docker
    Docker-->>DockerCtx: docker_instance
    DockerCtx->>aiohttp_ClientTimeout: aiohttp_ClientTimeout(total=None, sock_read=None)
    activate aiohttp_ClientTimeout
    aiohttp_ClientTimeout-->>DockerCtx: timeout_instance
    DockerCtx->>aiohttp_ClientSession: aiohttp_ClientSession(connector=docker_instance.connector, timeout=timeout_instance)
    activate aiohttp_ClientSession
    aiohttp_ClientSession-->>DockerCtx: session_instance
    DockerCtx->>Docker: set session = session_instance
    Docker-->>DockerCtx: session_configured
    DockerCtx-->>ContainerModel: docker_instance as context
    deactivate DockerCtx

    ContainerModel->>Docker: get_container_logs(container_name)
    Docker-->>ContainerModel: Async log stream (no total or sock_read timeout)
    ContainerModel-->>Service: yield log lines
    deactivate ContainerModel

    Service-->>DockerCtx: exit async context
    DockerCtx->>Docker: close()
    Docker-->>DockerCtx: closed
    DockerCtx-->>Service: context closed

Loading

Class diagram for updated DockerCtx and container log retrieval

classDiagram
    class DockerCtx {
        - Docker _client
        + DockerCtx(timeout Optional_int)
        + __aenter__() Docker
        + __aexit__(exc_type type, exc value, traceback object) None
    }

    class Docker {
        + connector aiohttp_BaseConnector
        + session aiohttp_ClientSession
        + Docker(session bool)
    }

    class aiohttp_ClientSession {
        + connector aiohttp_BaseConnector
        + timeout aiohttp_ClientTimeout
        + aiohttp_ClientSession(connector aiohttp_BaseConnector, timeout aiohttp_ClientTimeout)
    }

    class aiohttp_ClientTimeout {
        + total Optional_int
        + sock_read Optional_int
        + aiohttp_ClientTimeout(total Optional_int, sock_read Optional_int)
    }

    class ContainerModel {
        + get_container_log_by_name(container_name str) AsyncGenerator_str_None
        + get_raw_container_by_name(client Docker, container_name str) DockerContainer
    }

    DockerCtx --> Docker : creates
    Docker --> aiohttp_ClientSession : uses
    aiohttp_ClientSession --> aiohttp_ClientTimeout : uses
    ContainerModel ..> DockerCtx : uses_for_log_stream
    ContainerModel ..> Docker : uses_via_DockerCtx

Loading

File-Level Changes

Change	Details	Files
Add optional timeout support to DockerCtx for customizing aiohttp client session timeouts used by aiodocker.	Extend DockerCtx constructor to accept an optional timeout parameter. When no timeout is provided, preserve the previous behavior by constructing Docker with default settings. When a timeout is provided, force creation of an aiohttp session-backed Docker client. Construct a custom aiohttp.ClientSession with connector from Docker and a ClientTimeout that maps timeout=0 to no overall or sock_read timeout, otherwise using the given timeout value.	core/services/kraken/harbor/contexts.py
Use a non-timing-out Docker client when streaming container logs to avoid log stream timeout issues.	Change get_container_log_by_name to create its DockerCtx with timeout=0. Ensure log streaming relies on the custom session timeout behavior while other callers of DockerCtx remain unaffected.	core/services/kraken/harbor/container.py

Assessment against linked issues

Issue	Objective	Addressed	Explanation
#3763	Modify Kraken’s log ingestion/container log handling so it no longer causes long and repeating CPU usage surges in BlueOS 1.5.	✅

---

Tips and commands

Interacting with Sourcery

• Trigger a new review: Comment @sourcery-ai review on the pull request.
• Continue discussions: Reply directly to Sourcery's review comments.
• Generate a GitHub issue from a review comment: Ask Sourcery to create an
  issue from a review comment by replying to it. You can also reply to a
  review comment with @sourcery-ai issue to create an issue from it.
• Generate a pull request title: Write @sourcery-ai anywhere in the pull
  request title to generate a title at any time. You can also comment
  @sourcery-ai title on the pull request to (re-)generate the title at any time.
• Generate a pull request summary: Write @sourcery-ai summary anywhere in
  the pull request body to generate a PR summary at any time exactly where you
  want it. You can also comment @sourcery-ai summary on the pull request to
  (re-)generate the summary at any time.
• Generate reviewer's guide: Comment @sourcery-ai guide on the pull
  request to (re-)generate the reviewer's guide at any time.
• Resolve all Sourcery comments: Comment @sourcery-ai resolve on the
  pull request to resolve all Sourcery comments. Useful if you've already
  addressed all the comments and don't want to see them anymore.
• Dismiss all Sourcery reviews: Comment @sourcery-ai dismiss on the pull
  request to dismiss all existing Sourcery reviews. Especially useful if you
  want to start fresh with a new review - don't forget to comment
  @sourcery-ai review to trigger a new review!

Customizing Your Experience

Access your dashboard to:

• Enable or disable review features such as the Sourcery-generated pull request
  summary, the reviewer's guide, and others.
• Change the review language.
• Add, remove or edit custom review instructions.
• Adjust other review settings.

Getting Help

• Contact our support team for questions or feedback.
• Visit our documentation for detailed guides and information.
• Keep in touch with the Sourcery team by following us on X/Twitter, LinkedIn or GitHub.

[rafaellehmkuhl]

Tried just the harbor/ changes on Tony's Towfish ROV (where the problem was constantly happening) and there's no more CPU usage surges (after 25 minutes of testing).

[rafaellehmkuhl]

@joaomariolago I believe we can link that PR to #3763, right?

[sourcery-ai]

Hey - I've found 1 issue, and left some high level feedback:

• Consider avoiding timeout=0 as a magic value for "no timeout" in DockerCtx (e.g., use a named constant or a dedicated sentinel) and document that behavior in the class docstring so callers understand the semantics.
• The assignment self._client.session = self._client.session = aiohttp.ClientSession(...) in DockerCtx.__init__ looks accidental; simplify this to a single assignment to make the intent clear.

Prompt for AI Agents

Please address the comments from this code review:

## Overall Comments
- Consider avoiding `timeout=0` as a magic value for "no timeout" in `DockerCtx` (e.g., use a named constant or a dedicated sentinel) and document that behavior in the class docstring so callers understand the semantics.
- The assignment `self._client.session = self._client.session = aiohttp.ClientSession(...)` in `DockerCtx.__init__` looks accidental; simplify this to a single assignment to make the intent clear.

## Individual Comments

### Comment 1
<location> `core/services/kraken/harbor/contexts.py:16-20` </location>
<code_context>
+        if timeout is None:
+            self._client: Docker = Docker()
+        else:
+            # aiodocker will not create a session if is different from None
+            self._client: Docker = Docker(session=True)  # type: ignore
+            # We insert a new session with desired timeout
+            self._client.session = self._client.session = aiohttp.ClientSession(
+                connector=self._client.connector,
+                timeout=aiohttp.ClientTimeout(
+                    total=None if timeout == 0 else timeout,
</code_context>

<issue_to_address>
**issue (bug_risk):** Closing or reusing the initially created session should be explicit to avoid potential resource leaks.

Using `Docker(session=True)` creates an internal `ClientSession` that is then replaced with a new one, leaving the original potentially unclosed. To avoid leaking resources, either construct the desired `ClientSession` first and pass it into `Docker`, or explicitly close the initial session before overwriting `self._client.session`.
</issue_to_address>

---

Sourcery is free for open source - if you like our reviews please consider sharing them ✨

• X
• Mastodon
• LinkedIn
• Facebook

_{Help me be more useful! Please click 👍 or 👎 on each comment and I'll use the feedback to improve your reviews.}

[sourcery-ai]

issue (bug_risk): Closing or reusing the initially created session should be explicit to avoid potential resource leaks.

Using Docker(session=True) creates an internal ClientSession that is then replaced with a new one, leaving the original potentially unclosed. To avoid leaking resources, either construct the desired ClientSession first and pass it into Docker, or explicitly close the initial session before overwriting self._client.session.