mirror of https://lore.kernel.org/lkml/
 help / color / mirror / Atom feed
From: Brent Cook <busterb@mail.utexas.edu>
To: Dave Jones <davej@suse.de>
Cc: linux-kernel@vger.kernel.org
Subject: File permission problem with NFSv3 and 2.5.20-dj4
Date: Fri, 14 Jun 2002 17:30:01 -0500 (CDT)	[thread overview]
Message-ID: <20020614171820.A13031-100000@abbey.hauschen> (raw)
In-Reply-To: <Pine.LNX.4.33.0202072243560.26015-100000@Appserv.suse.de>

Hi,

  Looks like there is a problem with NFSv3 and file permissions in the DJ
kernels.

A file that is marked as executable will lose its executable flag whenever
it is written to. I suspect the proble lies in the changes to the NFS file
info cacheing in the DJ kernels at least since 2.5.20-dj1 (I was unable
to find where this change occured in the changelog).

Here is one example:

 Enter NFS mount (in this case, the NFS server is a FreeBSD 4.6 machine)
 Compile a simple program; gcc hello.c -o hello
 Result: hello has the following permissions: -rw-r--r--
 Modify the permissions manually; chmod 755 hello
 Result: hello has the following permissions: -rwxr-xr-x

Here is another:

 Enter NFS mount
 Compile a simple program; gcc hello.c -o hello
 Result: hello has the following permissions: -rw-r--r--
 Unmount the NFS mount; umount /home
 Remount the NFS mount; mount server:/home /home
 Result: hello has the following permissions: -rwxr-xr-x

Here is the final one:

 Enter NFS mount, find a file with executable permissions;
 Edit a file; vi whahoo.sh
 Save and close the file
 Results: file has the following permissions: -rw-r--r--

So, in the process of writing a file, its executable bits are lost. Can
someone help? The problem is not present with vanilla Linux-2.5.20.

Regards,
 - Brent Cook


  parent reply	other threads:[~2002-06-14 22:30 UTC|newest]

Thread overview: 22+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2002-02-06  3:52 Fix for duplicate /proc entries Brent Cook
2002-02-06 18:11 ` Dave Jones
2002-02-07 21:38   ` Brent Cook
2002-02-07 21:45     ` Dave Jones
2002-02-08 16:13       ` Brent Cook
2002-02-08 17:47         ` Mike Fedyk
2002-02-08 17:54           ` Alexander Viro
2002-02-08 18:12           ` Tommy Reynolds
2002-06-14 22:30       ` Brent Cook [this message]
2002-06-15 12:23         ` File permission problem with NFSv3 and 2.5.20-dj4 Dave Jones
2002-06-19 17:45           ` Brent Cook
2002-06-19 17:48             ` Dave Jones
2002-06-19 17:52             ` another sched.c error with athlon Kirk Reiser
2002-06-19 18:27               ` Adrian Bunk
2002-04-10 14:02   ` Mouse interrupts: the death knell of a VP6 Brent Cook
2002-04-10 15:23     ` Oleg Drokin
2002-04-10 16:43       ` Brent Cook
2002-04-10 16:49         ` William Park
2002-04-11  6:39           ` john slee
2002-04-10 20:55         ` Oleg Drokin
2002-04-10 17:16       ` John Adams
2002-04-10 17:52         ` Brent Cook

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=20020614171820.A13031-100000@abbey.hauschen \
    --to=busterb@mail.utexas.edu \
    --cc=davej@suse.de \
    --cc=linux-kernel@vger.kernel.org \
    /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

all inboxes | Powered by JetHome®