From: "Fabian Grünbichler" <f.gruenbichler@proxmox.com>
To: Proxmox Backup Server development discussion
<pbs-devel@lists.proxmox.com>
Subject: [pbs-devel] applied: [PATCH proxmox-backup] backup-client: mount: fix read of larger files
Date: Mon, 05 Sep 2022 15:20:43 +0200 [thread overview]
Message-ID: <1662383927.ks7c13wq7m.astroid@nora.none> (raw)
In-Reply-To: <20220902072114.635440-1-d.csapak@proxmox.com>
with a comment added and a slight rewording of the commit message, as
triggering the bug doesn't really truncate files, it overwrites parts of
them with all-zero blocks at different offsets.
On September 2, 2022 9:21 am, Dominik Csapak wrote:
> fuse_lowlevel.h says about read:
>
> Read should send exactly the number of bytes requested except
> on EOF or error, otherwise the rest of the data will be
> substituted with zeroes.
>
> but we simply forwarded the bytes we got from 'read_at'. The result was
> that larger files were truncated as soon as read_at returned not the
> exact number of bytes requested.
>
> To fix that, loop over 'read_at' until our buffer is full, or we read
> 0 bytes, indicating EOF.
>
> reported in the forum:
> https://forum.proxmox.com/threads/proxmox-backup-client-mounting-a-pxar-archive-gives-truncated-files.114447/
>
> Signed-off-by: Dominik Csapak <d.csapak@proxmox.com>
> ---
> tested with some larger files (>100MiB), before the patch, a 'cmp' with the
> original would always fail, after it worked like expected
>
> pbs-client/src/pxar/fuse.rs | 13 +++++++++++--
> 1 file changed, 11 insertions(+), 2 deletions(-)
>
> diff --git a/pbs-client/src/pxar/fuse.rs b/pbs-client/src/pxar/fuse.rs
> index e9e1b6f3..f2ce6a43 100644
> --- a/pbs-client/src/pxar/fuse.rs
> +++ b/pbs-client/src/pxar/fuse.rs
> @@ -566,8 +566,17 @@ impl SessionImpl {
> let file = self.get_lookup(inode)?;
> let content = self.open_content(&file)?;
> let mut buf = vec::undefined(len);
> - let got = content.read_at(&mut buf, offset).await?;
> - buf.truncate(got);
> + let mut pos = 0;
> + loop {
> + let got = content
> + .read_at(&mut buf[pos..], offset + pos as u64)
> + .await?;
> + pos += got;
> + if got == 0 || pos >= len {
> + break;
> + }
> + }
> + buf.truncate(pos);
> Ok(buf)
> }
>
> --
> 2.30.2
>
>
>
> _______________________________________________
> pbs-devel mailing list
> pbs-devel@lists.proxmox.com
> https://lists.proxmox.com/cgi-bin/mailman/listinfo/pbs-devel
>
>
>
prev parent reply other threads:[~2022-09-05 13:21 UTC|newest]
Thread overview: 2+ messages / expand[flat|nested] mbox.gz Atom feed top
2022-09-02 7:21 [pbs-devel] " Dominik Csapak
2022-09-05 13:20 ` Fabian Grünbichler [this message]
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=1662383927.ks7c13wq7m.astroid@nora.none \
--to=f.gruenbichler@proxmox.com \
--cc=pbs-devel@lists.proxmox.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.