Bug 920752
Summary: | cannot open device nodes for writing on RO filesystems | ||||||||||
---|---|---|---|---|---|---|---|---|---|---|---|
Product: | Red Hat Enterprise Linux 6 | Reporter: | Mateusz Guzik <mguzik> | ||||||||
Component: | kernel | Assignee: | Eric Sandeen <esandeen> | ||||||||
Status: | CLOSED ERRATA | QA Contact: | Eryu Guan <eguan> | ||||||||
Severity: | low | Docs Contact: | |||||||||
Priority: | high | ||||||||||
Version: | 6.4 | CC: | aviro, branto, btotty, bugproxy, ctrianta, dhoward, dhowells, d.yu.bolkhovityanov, eguan, esandeen, fs-maint, hannsj_uhl, istoev, james-p, jhunt, msvoboda, pep, rpiddapa, rwheeler, tgummels, wmealing, woodard | ||||||||
Target Milestone: | rc | Keywords: | Regression, ZStream | ||||||||
Target Release: | --- | ||||||||||
Hardware: | All | ||||||||||
OS: | Linux | ||||||||||
Whiteboard: | |||||||||||
Fixed In Version: | kernel-2.6.32-375.el6 | Doc Type: | Bug Fix | ||||||||
Doc Text: |
A bug in the do_filp_open() function caused it to exit early if any write access was requested on a read-only file system. This prevented the opening of device nodes on a read-only file system. With this update, the do_filp_open() has been fixed to no longer exit if a write request is made on a read-only file system.
|
Story Points: | --- | ||||||||
Clone Of: | |||||||||||
: | 1020725 (view as bug list) | Environment: | |||||||||
Last Closed: | 2013-11-21 16:56:45 UTC | Type: | Bug | ||||||||
Regression: | --- | Mount Type: | --- | ||||||||
Documentation: | --- | CRM: | |||||||||
Verified Versions: | Category: | --- | |||||||||
oVirt Team: | --- | RHEL 7.3 requirements from Atomic Host: | |||||||||
Cloudforms Team: | --- | Target Upstream Version: | |||||||||
Embargoed: | |||||||||||
Bug Depends On: | |||||||||||
Bug Blocks: | 832743, 855142, 928683, 961662, 975406, 1020725 | ||||||||||
Attachments: |
|
This behaviour makes some sense for a read-only file system. Are you looking to make a read-only root for security reasons? Is this a real issue in production or just a feature request? Thanks! Ric, we do need to be able to open devices RW even if the device node is on an RO fs. Looks like a regression from the big freeze patchset. I'll take it. Thanks, -Eric Same customer reported broken write support for file bind mounts (mount --bind file /rofs/file). This is also fixed with attached patch. *** Bug 923778 has been marked as a duplicate of this bug. *** *** Bug 928026 has been marked as a duplicate of this bug. *** *** Bug 955580 has been marked as a duplicate of this bug. *** *** Bug 957780 has been marked as a duplicate of this bug. *** Patch(es) available on kernel-2.6.32-375.el6 I've taken the private flag off this bug, several people have run into this, several bugs have been duped to it, and there's nothing sensitive here. *** Bug 962624 has been marked as a duplicate of this bug. *** Created attachment 752878 [details]
the result of "sosreport -a" on NFS server
Created attachment 752879 [details]
the result of "sosreport --batch"
*** Bug 962630 has been marked as a duplicate of this bug. *** ext2/3/4 xfs and btrfs all passed xfstests generic/306 which is the reproducer for this bug. Set to VERIFIED. Since the problem described in this bug report should be resolved in a recent advisory, it has been closed with a resolution of ERRATA. For information on the advisory, and where to find the updated files, follow the link below. If the solution does not work for you, open a new bug report. http://rhn.redhat.com/errata/RHSA-2013-1645.html *** Bug 951311 has been marked as a duplicate of this bug. *** *** Bug 951668 has been marked as a duplicate of this bug. *** |
Created attachment 709070 [details] patch Description of problem: If a filesystem is mounted read-only it is not possible to open device nodes residing on it for writing. Version-Release number of selected component (if applicable): kernel-2.6.32-327.el6 and up, introduced in: 1d3c71fcd66a1204fb43f3f06221732c6df1c7c2 Author: Eric Sandeen <sandeen> Date: Mon Sep 24 21:27:50 2012 -0400 [fs] Push mnt_want_write() outside of i_mutex How reproducible: Always Steps to Reproduce: 1. prepare fs dd if=/dev/zero of=foo bs=1M count=10 losetup -vf foo mkfs.ext3 /dev/loop0 mount /dev/loop0 /mnt mknod /mnt/null c 1 3 umount /mnt 2. mount fs ro: mount -o ro /dev/loop0 /mnt 3. Try to write: echo foo >> /mnt/null Actual results: zsh: read-only file system: /mnt/null Expected results: successful open + write