All of lore.kernel.org
 help / color / mirror / Atom feed
From: Steve French <smfrench@gmail.com>
To: Xiaoli Feng <xifeng@redhat.com>,
	samba-technical <samba-technical@lists.samba.org>
Cc: CIFS <linux-cifs@vger.kernel.org>
Subject: Re: fallocate can't change the cifs disk space usage
Date: Mon, 3 Aug 2020 13:36:36 -0500	[thread overview]
Message-ID: <CAH2r5muYQB-OaX8Njhr60LVy58LRcp2bSHMK1abA4Mqg4uFdQw@mail.gmail.com> (raw)
In-Reply-To: <CAH2r5muKciTKUjMa6o62J28Tvq=qaQryOGvfC8NRWkV29yiDaQ@mail.gmail.com>

One of the things that Samba server appears to be doing wrong is that
Samba is creating the file as a sparse file and returning that it is
not a sparse file - this obviously creates a problem because an
fallocate on a non-sparse file can be a no op ... but the server
didn't tell us it was a sparse file ...

On Mon, Aug 3, 2020 at 10:59 AM Steve French <smfrench@gmail.com> wrote:
>
> So locally on xfs this works, but Samba when receiving the request
> remotely apparently isn't doing the fallocate.  Would be curious if
> this works with vfs_btrfs and btrfs filesystem on the share - or a
> Samba server bug.
>
> On Sun, Aug 2, 2020 at 9:55 PM Xiaoli Feng <xifeng@redhat.com> wrote:
> >
> > Hello all,
> >
> > Recently when I'm investigating the xfstests generic/213 generic/228 failures for cifs.
> > Found that fallocate can't change the cifs disk space usage. Comparing xfs fileystem,
> > fallocate can update space usage.
> >
> > My tests is in 5.8.0-rc7+. I also file a bug for this issue.
> >   https://bugzilla.kernel.org/show_bug.cgi?id=208775
> >
> > # cat /etc/samba/smb.conf
> > [cifs]
> > path=/mnt/cifs
> > writeable=yes
> > # mount //localhost/cifs cifs -o user=root,password=redhat,cache=none,actimeo=0
> > # df -h cifs
> > Filesystem        Size  Used Avail Use% Mounted on
> > //localhost/cifs   36G   23G   13G  66% /root/cifs
> > #  fallocate -o 0 -l 2g /root/cifs/file1
> > # df -h cifs
> > Filesystem        Size  Used Avail Use% Mounted on
> > //localhost/cifs   36G   23G   13G  66% /root/cifs
> > ]# ls -l cifs
> > total 1
> > -rwxr-xr-x. 1 root root 2147483648 Aug  2 21:57 file1
> >
> > Thanks.
> >
> > --
> > Best regards!
> > XiaoLi Feng 冯小丽
> >
>
>
> --
> Thanks,
>
> Steve



-- 
Thanks,

Steve

  reply	other threads:[~2020-08-03 18:36 UTC|newest]

Thread overview: 7+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
     [not found] <1119714633.44793917.1596421202774.JavaMail.zimbra@redhat.com>
2020-08-03  2:48 ` fallocate can't change the cifs disk space usage Xiaoli Feng
2020-08-03  4:42   ` Steve French
2020-08-03  8:46     ` Xiaoli Feng
2020-08-03 15:59   ` Steve French
2020-08-03 18:36     ` Steve French [this message]
2020-08-04 17:39   ` Steve French
2020-08-06  2:35     ` Xiaoli Feng

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=CAH2r5muYQB-OaX8Njhr60LVy58LRcp2bSHMK1abA4Mqg4uFdQw@mail.gmail.com \
    --to=smfrench@gmail.com \
    --cc=linux-cifs@vger.kernel.org \
    --cc=samba-technical@lists.samba.org \
    --cc=xifeng@redhat.com \
    /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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.