On Mon, Aug 7, 2017 at 10:57 AM, zhangyi (F) <yi.zhang@xxxxxxxxxx> wrote:
> Hi,all:
> I found two problemes about overlayfs, please let me know what you think:
>
> 1. Cannot update upper dir's access time.
> Reproduce:
> # mkdir lower upper worker merger
> # mkdir upper/dir
> # mount -t overlay -o lowerdir=lower,upperdir=upper,workdir=worker overlayfs merger
> # ls mrger/dir
> # stat mrger/dir
> Access: 2017-08-07 11:03:56.364771584 -0400
> Modify: 2017-08-07 11:01:52.319771584 -0400
> Change: 2017-08-07 11:01:52.319771584 -0400
> # touch mrger/dir/aa
> # stat mrger/dir
> Access: 2017-08-07 11:03:56.364771584 -0400
> Modify: 2017-08-07 11:05:33.969771584 -0400
> Change: 2017-08-07 11:05:33.969771584 -0400
> # ls mrger/dir
> # stat mrger/dir
> Access: 2017-08-07 11:03:56.364771584 -0400
> Modify: 2017-08-07 11:05:33.969771584 -0400
> Change: 2017-08-07 11:05:33.969771584 -0400
>
> I find the reason of this problem is in update_ovl_inode_times():
> (598e3c8f7 "vfs: update ovl inode before relatime check")
>
> static void update_ovl_inode_times(struct dentry *dentry, struct inode *inode,
> bool rcu)
> {
> if (!rcu) {
> struct inode *realinode = d_real_inode(dentry);
>
> *d_real_inode() cannot get dir's real inode, so i_mtime always equal to i_ctime*
> *an not updated*
>
> if (unlikely(inode != realinode) &&
> (!timespec_equal(&inode->i_mtime, &realinode->i_mtime) ||
> !timespec_equal(&inode->i_ctime, &realinode->i_ctime))) {
> inode->i_mtime = realinode->i_mtime;
> inode->i_ctime = realinode->i_ctime;
> }
> }
> }
>
> 2. Chattr will modify lower file's attributes directly.
> Reproduce:
> # mkdir lower upper worker merger
> # touch lower/aa
> # lsattr -p lower/aa
> 0 --------------e---- lower/aa
> # mount -t overlay -o lowerdir=lower,upperdir=upper,workdir=worker overlayfs merger
> # chattr -p 123 merger/aa #set project id
> # lsattr -p lower/aa
> 123 --------------e---- lower/aa
>
> If we try to set "immutable" or any other attributes, the result are consistent.
> Because chattr open file in RDONLY mode, so it will not trigger copyup, and then,
> FS_IOC_SETFLAGS ioctl will get the lower inode and modify it. By the way,
> ovl's directory not support ioctl now, so we fail to change dir's attributes.
>
> Do you think these two "problemes" need to fix or should avoid ?
>
ZhangYi,
Now that fixes for the 2 issues you reported are in overlayfs-next,
could you send those 2 reproducers as xfstests?
Thanks,
Amir.
--
To unsubscribe from this list: send the line "unsubscribe linux-unionfs" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
