All of lore.kernel.org
 help / color / mirror / Atom feed
From: OpenBMC Patches <openbmc-patches@stwcx.xyz>
To: openbmc@lists.ozlabs.org
Subject: [PATCH openbmc 5/7] obmc-initfs: run fsck on read/write file system
Date: Sat,  6 Feb 2016 18:00:19 -0600	[thread overview]
Message-ID: <1454803221-12014-6-git-send-email-openbmc-patches@stwcx.xyz> (raw)
In-Reply-To: <1454803221-12014-1-git-send-email-openbmc-patches@stwcx.xyz>

From: Milton Miller <miltonm@us.ibm.com>

Good practice is to run fsck to repair minor damage and make a
filesystem consistent before mounting it.  We don't have space
in the initramfs to add fsck, but we can mount the read-only
reference partition first and run fsck from there.

Also a test that init appears to be an executable file in the
combined target filesystem in addition to /bin/sh actually
executing and a minor consolidation of establishing the overly
directories in the read-write filesystem.

Signed-off-by: Milton Miller <miltonm@us.ibm.com>
---
 .../obmc-phosphor-initfs/files/obmc-init.sh        | 29 +++++++++++++++++++---
 1 file changed, 25 insertions(+), 4 deletions(-)

diff --git a/meta-phosphor/common/recipes-phosphor/obmc-phosphor-initfs/files/obmc-init.sh b/meta-phosphor/common/recipes-phosphor/obmc-phosphor-initfs/files/obmc-init.sh
index daa4f5c..b7ef09f 100644
--- a/meta-phosphor/common/recipes-phosphor/obmc-phosphor-initfs/files/obmc-init.sh
+++ b/meta-phosphor/common/recipes-phosphor/obmc-phosphor-initfs/files/obmc-init.sh
@@ -75,6 +75,8 @@ roopts=ro
 rwopts=rw
 
 init=/sbin/init
+fsck=/sbin/fsck.$rwfst
+fsckopts=-a
 
 echo rofs = $rofs $rofst   rwfs = $rwfs $rwfst
 
@@ -83,6 +85,28 @@ then
 	debug_takeover "Debug initial shell requested by command line."
 fi
 
+mount $rodev $rodir -t $rofst -o $roopts
+
+if test -x $rodir$fsck
+then
+	for fs in $fslist
+	do
+		mount --bind $fs $rodir/$fs
+	done
+	chroot $rodir $fsck $fsckopts $rwdev
+	rc=$?
+	for fs in $fslist
+	do
+		umount $rodir/$fs
+	done
+	if test $rc -gt 1
+	then
+		debug_takeover "fsck of read-write fs on $rwdev failed (rc=$rc)"
+	fi
+else
+	echo "No '$fsck' in read only fs, skipping fsck."
+fi
+
 if ! mount $rwdev $rwdir -t $rwfst -o $rwopts
 then
 	msg="$(cat)" << HERE
@@ -93,11 +117,8 @@ HERE
 	debug_takeover "$msg"
 fi
 
-mount $rodev $rodir -t $rofst -o $roopts
-
 rm -rf $work
-mkdir -p $upper
-mkdir -p $work
+mkdir -p $upper $work
 
 mount -t overlay -o lowerdir=$rodir,upperdir=$upper,workdir=$work cow /root
 
-- 
2.6.4

  parent reply	other threads:[~2016-02-07  0:00 UTC|newest]

Thread overview: 18+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2016-02-07  0:00 [PATCH openbmc 0/7] initfs: check and auto-repair read/write file system OpenBMC Patches
2016-02-07  0:00 ` [PATCH openbmc 1/7] obmc-initfs: minor updates OpenBMC Patches
2016-02-08 16:25   ` Brad Bishop
2016-02-09  0:55   ` Andrew Jeffery
2016-02-07  0:00 ` [PATCH openbmc 2/7] obmc-initfs: look for images in /run/initramfs/image- OpenBMC Patches
2016-02-08 16:25   ` Brad Bishop
2016-02-07  0:00 ` [PATCH openbmc 3/7] obmc-initfs: use varables for paths and mount arguments OpenBMC Patches
2016-02-08 16:37   ` Brad Bishop
2016-02-09  0:46   ` Andrew Jeffery
2016-02-07  0:00 ` [PATCH openbmc 4/7] obmc-initfs: factor debug and takeover OpenBMC Patches
2016-02-08 16:58   ` Brad Bishop
2016-02-07  0:00 ` OpenBMC Patches [this message]
2016-02-08 17:03   ` [PATCH openbmc 5/7] obmc-initfs: run fsck on read/write file system Brad Bishop
2016-02-07  0:00 ` [PATCH openbmc 6/7] obmc-initfs: run update as a sub-script OpenBMC Patches
2016-02-08 17:05   ` Brad Bishop
2016-02-07  0:00 ` [PATCH openbmc 7/7] obmc-initfs: shutdown when rofs is a loop mounted image in /run OpenBMC Patches
2016-02-08 17:07   ` Brad Bishop
  -- strict thread matches above, loose matches on Subject: below --
2016-02-06  0:20 [PATCH openbmc 0/7] initfs: check and auto-repair read/write file system OpenBMC Patches
2016-02-06  0:20 ` [PATCH openbmc 5/7] obmc-initfs: run fsck on " OpenBMC Patches

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=1454803221-12014-6-git-send-email-openbmc-patches@stwcx.xyz \
    --to=openbmc-patches@stwcx.xyz \
    --cc=openbmc@lists.ozlabs.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 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.