Linux Btrfs filesystem development
 help / color / mirror / Atom feed
From: Vyacheslav Kovalevsky <slava.kovalevskiy.2014@gmail.com>
To: clm@fb.com, dsterba@suse.com
Cc: linux-btrfs@vger.kernel.org, linux-kernel@vger.kernel.org
Subject: Directory is not persisted after creating 100s of files inside, writing to another file and renaming it if system crashes.
Date: Wed, 3 Dec 2025 14:26:22 +0300	[thread overview]
Message-ID: <84c4e713-85d6-42b9-8dcf-0722ed26cb05@gmail.com> (raw)

Directory entry is not persisted after creating 100s (hundreds) of files inside, writing to another file (with `O_SYNC` flag) and renaming it if system crashes.


Detailed description
====================

Hello, we have found another issue with btrfs crash behavior.

In short:

1. Create and sync an empty file in root directory.
2. Make new directory in root directory.
3. Open the file with `O_SYNC` flag and write some data (of specific size).
4. Fill directory with specific number of empty files.
5. Rename the previously written file.
6. Sync root directory.

After system crash directory will be missing, although it was synced in the last step.


System info
===========

Linux version 6.18, also tested on 6.14.11.


How to reproduce
================

```
#include <errno.h>
#include <fcntl.h>
#include <stdio.h>
#include <string.h>
#include <sys/stat.h>
#include <sys/types.h>
#include <unistd.h>

#define BUFFER_LEN 1024 // should be at least ~ 528
#define FILE_N 256      // should be at least ~ 128

int main() {
   int status;
   int file_fd;
   int root_fd;

   int buffer[BUFFER_LEN + 1] = {};
   for (int i = 0; i < BUFFER_LEN; ++i) {
     buffer[i] = i;
   }

   status = creat("file1", S_IRWXU | S_IRWXG | S_IROTH | S_IXOTH);
   printf("CREAT: %d\n", status);
   close(status);

   // persist `file1`
   sync();

   status = mkdir("dir", S_IRWXU | S_IRWXG | S_IROTH | S_IXOTH);
   printf("MKDIR: %d\n", status);

   // `O_SYNC` is important
   status = open("file1", O_WRONLY | O_SYNC);
   printf("OPEN: %d\n", status);
   file_fd = status;

   status = write(file_fd, buffer, BUFFER_LEN);
   printf("WRITE: %d\n", status);

   char path[100];
   // fill directory with a lot of empty files
   for (int i = 0; i < FILE_N; ++i) {
     sprintf(path, "dir/%d", i);
     status = creat(path, S_IRWXU | S_IRWXG | S_IROTH | S_IXOTH);
     close(status);
   }

   status = rename("file1", "file2");
   printf("RENAME: %d\n", status);

   status = open(".", O_RDONLY | O_DIRECTORY);
   printf("OPEN: %d\n", status);
   root_fd = status;

   // persist `dir`
   status = fsync(root_fd);
   printf("FSYNC: %d\n", status);
}
// after the crash `dir` is missing
```

Steps:

1. Create and mount new btrfs file system in default configuration.
2. Change directory to root of the file system and run the compiled test.
3. Cause hard system crash (e.g. QEMU `system_reset` command).
4. Remount file system after crash.
5. Observe that `dir` directory is missing.


             reply	other threads:[~2025-12-03 11:26 UTC|newest]

Thread overview: 2+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2025-12-03 11:26 Vyacheslav Kovalevsky [this message]
2025-12-03 17:42 ` Directory is not persisted after creating 100s of files inside, writing to another file and renaming it if system crashes Filipe Manana

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=84c4e713-85d6-42b9-8dcf-0722ed26cb05@gmail.com \
    --to=slava.kovalevskiy.2014@gmail.com \
    --cc=clm@fb.com \
    --cc=dsterba@suse.com \
    --cc=linux-btrfs@vger.kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox