public inbox for pbs-devel@lists.proxmox.com
 help / color / mirror / Atom feed
From: Christian Ebner <c.ebner@proxmox.com>
To: Gabriel Goller <g.goller@proxmox.com>
Cc: pbs-devel@lists.proxmox.com
Subject: Re: [pbs-devel] [PATCH proxmox-backup v2] client: mount: flush output before exiting
Date: Tue, 18 Jun 2024 15:00:53 +0200	[thread overview]
Message-ID: <52d3ae5f-ed06-42cc-a189-82326e735234@proxmox.com> (raw)
In-Reply-To: <20240618122916.l4cvteseewltnjht@luna.proxmox.com>

On 6/18/24 14:29, Gabriel Goller wrote:
> 
> Hmm I think I know what happens...
> This patch is flaky for me as well, sometimes it works, sometimes it
> doesn't. The problem is that we move the `pw` file descriptor to
> `mount_do`, which means that after it returns, the file_descriptor gets
> dropped.
> 
>      let (pr, pw) = proxmox_sys::pipe()?;
>      let pr: OwnedFd = pr.into(); // until next sys bump
>      let pw: OwnedFd = pw.into();
>      match unsafe { fork() } {
>          Ok(ForkResult::Parent { .. }) => {
>              drop(pw);
>              // Blocks the parent process until we are ready to go in 
> the child
>              let _res = nix::unistd::read(pr.as_raw_fd(), &mut 
> [0]).unwrap();
>              Ok(Value::Null)
>          }
>          Ok(ForkResult::Child) => {
>              drop(pr);
>              nix::unistd::setsid().unwrap();
>              let result = proxmox_async::runtime::main(mount_do(param, 
> Some(pw)));
>                                                                             ^^^
> pw gets dropped here, which means the fd is closed.
> 
>              io::stdout().flush()?;
>              io::stderr().flush()?;
>              result
>          }
>          Err(_) => bail!("failed to daemonize process"),
>      }
> 
> 
> In the parent process we read from the fd—which is then closed—meaning
> that the parent exits at the same time as we do the stderr/stdout flush.
> 
> And actually this patch was stupid to begin with, because we don't even
> print the error in the mound_do function, but later in the api-handler
> :(.

That is actually true, no point in flushing here. So my suggestion for 
the fix was wrong, as I was primed on the error location :(

> 
> I think we need to do something like this:
> 
>      if let Err(e) = nix::unistd::read(pr.as_raw_fd(), &mut [0]) {
>          // if the fd is closed, or some other error happens, wait so
>          // that the child can return the error safely, then exit 
>          nix::sys::wait::wait().unwrap();
>      }
>      Ok(Value::Null)
> 
> 


_______________________________________________
pbs-devel mailing list
pbs-devel@lists.proxmox.com
https://lists.proxmox.com/cgi-bin/mailman/listinfo/pbs-devel

  reply	other threads:[~2024-06-18 13:00 UTC|newest]

Thread overview: 10+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2024-06-14  9:39 Gabriel Goller
2024-06-14  9:54 ` Christian Ebner
2024-06-18 11:20 ` Friedrich Weber
2024-06-18 11:41   ` Christian Ebner
2024-06-18 12:29     ` Gabriel Goller
2024-06-18 13:00       ` Christian Ebner [this message]
2024-06-18 13:10         ` Gabriel Goller
2024-06-18 12:32     ` Friedrich Weber
2024-06-18 12:42       ` Christian Ebner
2024-06-18 12:53       ` Gabriel Goller

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=52d3ae5f-ed06-42cc-a189-82326e735234@proxmox.com \
    --to=c.ebner@proxmox.com \
    --cc=g.goller@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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox
Service provided by Proxmox Server Solutions GmbH | Privacy | Legal