[Bug 216775] New: fanotify reports parent PPID insted of PID for FAN_MODIFY events

[Bug 216775] New: fanotify reports parent PPID insted of PID for FAN_MODIFY events

[bugzilla-daemon]

https://bugzilla.kernel.org/show_bug.cgi?id=216775

            Bug ID: 216775
           Summary: fanotify reports parent PPID insted of PID for
                    FAN_MODIFY events
           Product: File System
           Version: 2.5
    Kernel Version: 5.15.0
          Hardware: Intel
                OS: Linux
              Tree: Mainline
            Status: NEW
          Severity: normal
          Priority: P1
         Component: ext4
          Assignee: fs_ext4@kernel-bugs.osdl.org
          Reporter: saikiran.gummaraj@icloud.com
        Regression: No

Hello,

While I've been developing a library around fanotify in Go, I noticed that
fanotify subsystem reports the parent process ID in fanotify_event_metadata.pid
instead of the Process ID when mask is set to FAN_MODIFY. I was able to confirm
the error through a test and also manually verifying the PIDs in the audit log.
I did not observe this behaviour for FAN_ACCESS bit.

I've been able to reproduce this on -

Ubuntu 20.04.5 - 5.15.0-53-generic
Ubuntu 22.10 - 5.19.0-23-generic

It can be reproduced by -

git clone git@github.com:opcoder0/fanotify.git
cd fanotify
sudo go test -v

The test "TestWithCapSysAdmFanotifyFileModified" fails reporting pid mismatch.

-- 
You may reply to this email to add a comment.

You are receiving this mail because:
You are watching the assignee of the bug.

Re: [Bug 216775] New: fanotify reports parent PPID insted of PID for FAN_MODIFY events

[Amir Goldstein]

On Mon, Dec 5, 2022 at 7:02 AM <bugzilla-daemon@kernel.org> wrote:
>
> https://bugzilla.kernel.org/show_bug.cgi?id=216775
>
>             Bug ID: 216775
>            Summary: fanotify reports parent PPID insted of PID for
>                     FAN_MODIFY events
>            Product: File System
>            Version: 2.5
>     Kernel Version: 5.15.0
>           Hardware: Intel
>                 OS: Linux
>               Tree: Mainline
>             Status: NEW
>           Severity: normal
>           Priority: P1
>          Component: ext4
>           Assignee: fs_ext4@kernel-bugs.osdl.org
>           Reporter: saikiran.gummaraj@icloud.com
>         Regression: No
>
> Hello,
>
> While I've been developing a library around fanotify in Go, I noticed that
> fanotify subsystem reports the parent process ID in fanotify_event_metadata.pid
> instead of the Process ID when mask is set to FAN_MODIFY. I was able to confirm
> the error through a test and also manually verifying the PIDs in the audit log.
> I did not observe this behaviour for FAN_ACCESS bit.
>
> I've been able to reproduce this on -
>
> Ubuntu 20.04.5 - 5.15.0-53-generic
> Ubuntu 22.10 - 5.19.0-23-generic
>
> It can be reproduced by -
>
> git clone git@github.com:opcoder0/fanotify.git
> cd fanotify
> sudo go test -v
>
> The test "TestWithCapSysAdmFanotifyFileModified" fails reporting pid mismatch.
>

It's a test bug.
The modify event with self pid is generated by os.WriteFile()
Either change test to expect modify event with self pid or move
test file creation before starting the listener.

Thanks,
Amir.

[Bug 216775] fanotify reports parent PPID insted of PID for FAN_MODIFY events

[bugzilla-daemon]

https://bugzilla.kernel.org/show_bug.cgi?id=216775

--- Comment #1 from Amir Goldstein (amir73il@gmail.com) ---
On Mon, Dec 5, 2022 at 7:02 AM <bugzilla-daemon@kernel.org> wrote:
>
> https://bugzilla.kernel.org/show_bug.cgi?id=216775
>
>             Bug ID: 216775
>            Summary: fanotify reports parent PPID insted of PID for
>                     FAN_MODIFY events
>            Product: File System
>            Version: 2.5
>     Kernel Version: 5.15.0
>           Hardware: Intel
>                 OS: Linux
>               Tree: Mainline
>             Status: NEW
>           Severity: normal
>           Priority: P1
>          Component: ext4
>           Assignee: fs_ext4@kernel-bugs.osdl.org
>           Reporter: saikiran.gummaraj@icloud.com
>         Regression: No
>
> Hello,
>
> While I've been developing a library around fanotify in Go, I noticed that
> fanotify subsystem reports the parent process ID in
> fanotify_event_metadata.pid
> instead of the Process ID when mask is set to FAN_MODIFY. I was able to
> confirm
> the error through a test and also manually verifying the PIDs in the audit
> log.
> I did not observe this behaviour for FAN_ACCESS bit.
>
> I've been able to reproduce this on -
>
> Ubuntu 20.04.5 - 5.15.0-53-generic
> Ubuntu 22.10 - 5.19.0-23-generic
>
> It can be reproduced by -
>
> git clone git@github.com:opcoder0/fanotify.git
> cd fanotify
> sudo go test -v
>
> The test "TestWithCapSysAdmFanotifyFileModified" fails reporting pid
> mismatch.
>

It's a test bug.
The modify event with self pid is generated by os.WriteFile()
Either change test to expect modify event with self pid or move
test file creation before starting the listener.

Thanks,
Amir.

-- 
You may reply to this email to add a comment.

You are receiving this mail because:
You are watching the assignee of the bug.

[Bug 216775] fanotify reports parent PPID insted of PID for FAN_MODIFY events

[bugzilla-daemon]

https://bugzilla.kernel.org/show_bug.cgi?id=216775

--- Comment #2 from opcoder0 (saikiran.gummaraj@icloud.com) ---
Sorry missed that. Thanks a lot!

-- 
You may reply to this email to add a comment.

You are receiving this mail because:
You are watching the assignee of the bug.

[Bug 216775] fanotify reports parent PPID insted of PID for FAN_MODIFY events

[bugzilla-daemon]

https://bugzilla.kernel.org/show_bug.cgi?id=216775

opcoder0 (saikiran.gummaraj@icloud.com) changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|---                         |INVALID

--- Comment #3 from opcoder0 (saikiran.gummaraj@icloud.com) ---
Test issue.

-- 
You may reply to this email to add a comment.

You are receiving this mail because:
You are watching the assignee of the bug.