cgroups: change memory usage calculation to match Docker by giuseppe · Pull Request #2455 · containers/common

giuseppe · 2025-06-12T08:56:27Z

This changes the memory usage calculation to follow Docker's approach:

For cgroup v2: memory.current - memory.stat['inactive_file']
For cgroup v1: memory.usage_in_bytes - memory.stat['total_inactive_file']

The previous calculation used anonymous memory only, but Docker's calculation provides more accurate memory usage by excluding inactive file cache.

Closes: #2454

Summary by Sourcery

Align cgroup memory usage calculation with Docker by subtracting inactive file caches for both v1 and v2

Enhancements:

Compute cgroup v2 memory usage as memory.current minus memory.stat['inactive_file']
Compute cgroup v1 memory usage as memory.usage_in_bytes minus memory.stat['total_inactive_file']

This changes the memory usage calculation to follow Docker's approach: - For cgroup v2: memory.current - memory.stat['inactive_file'] - For cgroup v1: memory.usage_in_bytes - memory.stat['total_inactive_file'] The previous calculation used anonymous memory only, but Docker's calculation provides more accurate memory usage by excluding inactive file cache. Closes: containers#2454 Signed-off-by: Giuseppe Scrivano <gscrivan@redhat.com>

sourcery-ai · 2025-06-12T08:56:32Z

Reviewer's Guide

This PR refactors the cgroups memory usage calculation to align with Docker’s methodology by subtracting inactive file cache from total memory usage in both cgroup v2 and cgroup v1, replacing the previous anonymous‐only approach.

Sequence diagram: cgroup v2 memory data retrieval

sequenceDiagram
    participant Handler as linuxMemHandler
    participant FS as FileSystem
    Handler->>FS: Read "memory.current"
    FS-->>Handler: current_value
    Handler->>FS: Read "inactive_file" from "memory.stat"
    FS-->>Handler: inactive_file_value

Sequence diagram: cgroup v1 memory data retrieval

sequenceDiagram
    participant Handler as linuxMemHandler
    participant FS as FileSystem
    Handler->>FS: Read "memory.usage_in_bytes"
    FS-->>Handler: usage_in_bytes_value
    Handler->>FS: Read "total_inactive_file" from "memory.stat"
    FS-->>Handler: total_inactive_file_value

Class diagram for modified linuxMemHandler

classDiagram
    class linuxMemHandler {
        +Stat(ctr *CgroupControl, m *cgroups.Stats) error
    }

Flow diagram for updated memory usage calculation logic

graph TD
    A[Start: Stat function] --> B{Is ctr.cgroup2?};
    B -- Yes --> C[memoryRoot = cgroup v2 path];
    C --> D[Read 'memory.current' --> current_val];
    D --> E[Read 'inactive_file' from 'memory.stat' --> inactiveFile_val];
    E --> F{inactiveFile_val < current_val?};
    F -- Yes --> G[memUsage.Usage.Usage = current_val - inactiveFile_val];
    F -- No --> H[memUsage.Usage.Usage = 0];
    G --> Z[Update m.Stats];
    H --> Z;

    B -- No --> I[memoryRoot = cgroup v1 path];
    I --> J[Read 'memory.usage_in_bytes' --> usageInBytes_val];
    J --> K[Read 'total_inactive_file' from 'memory.stat' --> totalInactiveFile_val];
    K --> L{totalInactiveFile_val < usageInBytes_val?};
    L -- Yes --> M[memUsage.Usage.Usage = usageInBytes_val - totalInactiveFile_val];
    L -- No --> N[memUsage.Usage.Usage = 0];
    M --> Z;
    N --> Z;
    Z --> End[Return error status];

File-Level Changes

Change	Details	Files
Switch cgroup v2 usage to memory.current - inactive_file	Read memory.current with readFileAsUint64 Read inactive_file via readFileByKeyAsUint64 Subtract inactive_file from current if less Removed anon-based usage calculation	`pkg/cgroups/memory_linux.go`
Switch cgroup v1 usage to memory.usage_in_bytes - total_inactive_file	Read memory.usage_in_bytes with readFileAsUint64 Read total_inactive_file via readFileByKeyAsUint64 Subtract total_inactive_file from usage_in_bytes if less Removed looping over anon fields	`pkg/cgroups/memory_linux.go`

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.

sourcery-ai

Hey @giuseppe - I've reviewed your changes and they look great!

Here's what I looked at during the review

🟡 General issues: 1 issue found
🟢 Security: all looks good
🟢 Testing: all looks good
🟢 Complexity: all looks good
🟢 Documentation: all looks good

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

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

sourcery-ai · 2025-06-12T08:59:28Z

+			return err
+		}
+
+		// Docker calculation: memory.current - memory.stat['inactive_file']


suggestion: Extract common memory usage logic into a helper

This calculation is repeated in both cgroup v1 and v2. Refactoring it into a shared utility function would improve maintainability and consistency.

Suggested implementation:

// Docker calculation: memory.current - memory.stat['inactive_file'] memUsage.Usage.Usage = calculateDockerMemUsage(current, inactiveFile)

// Read memory.usage_in_bytes usageInBytes, err := readFileAsUint64(filepath.Join(memoryRoot, "memory.usage_in_bytes"))

func calculateDockerMemUsage(usage, inactiveFile uint64) uint64 { if inactiveFile < usage { return usage - inactiveFile } return 0 }

You should also update the cgroup v1 code path to use calculateDockerMemUsage instead of duplicating the logic.
If the v1 code path also does memUsage.Usage.Usage = 0; if inactiveFile < usageInBytes { memUsage.Usage.Usage = usageInBytes - inactiveFile }, replace it with memUsage.Usage.Usage = calculateDockerMemUsage(usageInBytes, inactiveFile).

giuseppe · 2025-06-12T09:08:10Z

the difference with Docker is that when inactive_file is bigger than memory.current then 0 is returned instead of memory.current. Not sure what behavior is more correct in this race condition but the implementation in the current PR avoids the usage value to switch immediately from memory.current to a very small value when inactive_file is close to the usage.

giuseppe · 2025-06-16T12:50:37Z

@containers/podman-maintainers PTAL

Luap99

LGTM

openshift-ci · 2025-06-16T13:25:18Z

[APPROVALNOTIFIER] This PR is APPROVED

This pull-request has been approved by: giuseppe, Luap99, sourcery-ai[bot]

The full list of commands accepted by this bot can be found here.

The pull request process is described here

Details

Needs approval from an approver in each of these files:

~~OWNERS~~ [Luap99,giuseppe]

Approvers can indicate their approval by writing /approve in a comment
Approvers can cancel approval by writing /approve cancel in a comment

giuseppe · 2025-06-18T12:22:10Z

@mheon @flouthoc PTAL

jankaluza · 2025-06-24T09:09:30Z

LGTM

Luap99 · 2025-06-24T09:18:51Z

/lgtm

…up usage The registry container memory panel was reporting cgroup memory.current which includes reclaimable page cache, inflating reported usage far beyond actual container memory consumption. Add a working_set metric computed as memory.current - inactive_file (the standard calculation used by podman stats and Docker stats per containers/common#2455) and display it as the primary series on the bastion dashboard. Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

…up usage (#838) The registry container memory panel was reporting cgroup memory.current which includes reclaimable page cache, inflating reported usage far beyond actual container memory consumption. Add a working_set metric computed as memory.current - inactive_file (the standard calculation used by podman stats and Docker stats per containers/common#2455) and display it as the primary series on the bastion dashboard. Co-authored-by: Claude Opus 4.6 <noreply@anthropic.com>

openshift-ci Bot added the approved label Jun 12, 2025

giuseppe mentioned this pull request Jun 12, 2025

podman stats shows misleading memory usage on cgroup v2 #2454

Closed

sourcery-ai Bot approved these changes Jun 12, 2025

View reviewed changes

Luap99 approved these changes Jun 16, 2025

View reviewed changes

openshift-ci Bot assigned Luap99 Jun 24, 2025

openshift-ci Bot added the lgtm label Jun 24, 2025

openshift-merge-bot Bot merged commit 6769e5b into containers:main Jun 24, 2025
14 checks passed

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

cgroups: change memory usage calculation to match Docker#2455

cgroups: change memory usage calculation to match Docker#2455
openshift-merge-bot[bot] merged 1 commit into
containers:mainfrom
giuseppe:change-calculation-used-memory

giuseppe commented Jun 12, 2025 •

edited by sourcery-ai Bot

Loading

Uh oh!

sourcery-ai Bot commented Jun 12, 2025 •

edited

Loading

Interacting with Sourcery

Customizing Your Experience

Getting Help

Uh oh!

sourcery-ai Bot left a comment

Uh oh!

sourcery-ai Bot Jun 12, 2025

Uh oh!

giuseppe commented Jun 12, 2025

Uh oh!

giuseppe commented Jun 16, 2025

Uh oh!

Luap99 left a comment

Uh oh!

openshift-ci Bot commented Jun 16, 2025

Uh oh!

giuseppe commented Jun 18, 2025

Uh oh!

jankaluza commented Jun 24, 2025

Uh oh!

Luap99 commented Jun 24, 2025

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

3 participants

Conversation

giuseppe commented Jun 12, 2025 • edited by sourcery-ai Bot Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Summary by Sourcery

Uh oh!

sourcery-ai Bot commented Jun 12, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Reviewer's Guide

Sequence diagram: cgroup v2 memory data retrieval

Sequence diagram: cgroup v1 memory data retrieval

Class diagram for modified linuxMemHandler

Flow diagram for updated memory usage calculation logic

File-Level Changes

Interacting with Sourcery

Customizing Your Experience

Getting Help

Uh oh!

sourcery-ai Bot left a comment

Choose a reason for hiding this comment

Uh oh!

sourcery-ai Bot Jun 12, 2025

Choose a reason for hiding this comment

Uh oh!

giuseppe commented Jun 12, 2025

Uh oh!

giuseppe commented Jun 16, 2025

Uh oh!

Luap99 left a comment

Choose a reason for hiding this comment

Uh oh!

openshift-ci Bot commented Jun 16, 2025

Uh oh!

giuseppe commented Jun 18, 2025

Uh oh!

jankaluza commented Jun 24, 2025

Uh oh!

Luap99 commented Jun 24, 2025

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

3 participants

giuseppe commented Jun 12, 2025 •

edited by sourcery-ai Bot

Loading

sourcery-ai Bot commented Jun 12, 2025 •

edited

Loading