Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Setting "fast_ino=1" results in wrong directory entries #376

Open
Kawashima-Azumi opened this issue Sep 27, 2022 · 0 comments
Open

Setting "fast_ino=1" results in wrong directory entries #376

Kawashima-Azumi opened this issue Sep 27, 2022 · 0 comments

Comments

@Kawashima-Azumi
Copy link

Kawashima-Azumi commented Sep 27, 2022

Hello,

In a use case where fast_ino is set to 1, wrong entries are found in some of the union directories.

"/data-old" and "/data-new" are two independent nfs3 filesystems. Two fuse-overlayfs are mounted with the following commands:

# fuse-overlayfs -o ro,lowerdir=/data-new:/data-old,timeout=1,noacl,noatime,static_nlink,threaded=1,fast_ino=1 /data-fast_ino

# fuse-overlayfs -o ro,lowerdir=/data-new:/data-old,timeout=1,noacl,noatime,static_nlink,threaded=1 /data

Under /data-fast_ino, there are some subdirectories that come from /data-new show entries that are directly under /data-old. i.e.,

/data-new/a/b/c/{some files}
/data-old/{x, y, z, etc., but no dir 'a', which is only in /data-new}

under /data/a/b/c/ there are the dirs x, y, and z, while missing some files from /data-new/a/b/c/.

A check with "stat -c $i" shows that the inode numbers of the entries are the same under "/data/a/b/c/" and "/data-fast_ino/a/b/c/", even though some entry names are wrong, and their types are shown as directories when they are actually regular files under "/data-fast_ino/a/b/c/".

Under "/data/a/b/c/" everything is correct.

fuse-overlayfs versions: 1.9 and 1.10-dev (20220927).
FUSE library version 3.11.0
using FUSE kernel interface version 7.31
fusermount3 version: 3.11.0

Is this a bug or expected bahevior? If expected, please document it. Thanks!

Kawashima

Edit:
Not sure if this is related to #359
(didn't have time to check).

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

1 participant