public inbox for linux-xfs@vger.kernel.org
 help / color / mirror / Atom feed
From: "Darrick J. Wong" <djwong@kernel.org>
To: Catherine Hoang <catherine.hoang@oracle.com>
Cc: linux-xfs@vger.kernel.org
Subject: Re: [PATCH v2 2/2] xfs_admin: get UUID of mounted filesystem
Date: Tue, 20 Dec 2022 14:29:23 -0800	[thread overview]
Message-ID: <Y6I3Q28TAOv7VVET@magnolia> (raw)
In-Reply-To: <20221219181824.25157-3-catherine.hoang@oracle.com>

On Mon, Dec 19, 2022 at 10:18:24AM -0800, Catherine Hoang wrote:
> Adapt this tool to call xfs_io to retrieve the UUID of a mounted filesystem.
> This is a precursor to enabling xfs_admin to set the UUID of a mounted
> filesystem.
> 
> Signed-off-by: Catherine Hoang <catherine.hoang@oracle.com>
> Reviewed-by: Allison Henderson <allison.henderson@oracle.com>
> ---
>  db/xfs_admin.sh      | 27 +++++++++++++++++++++++----
>  man/man8/xfs_admin.8 |  4 ++++
>  2 files changed, 27 insertions(+), 4 deletions(-)
> 
> diff --git a/db/xfs_admin.sh b/db/xfs_admin.sh
> index 409975b2..cc9a2150 100755
> --- a/db/xfs_admin.sh
> +++ b/db/xfs_admin.sh
> @@ -6,10 +6,12 @@
>  
>  status=0
>  DB_OPTS=""
> +DB_EXTRA_OPTS=""
> +IO_OPTS=""
>  REPAIR_OPTS=""
>  REPAIR_DEV_OPTS=""
>  LOG_OPTS=""
> -USAGE="Usage: xfs_admin [-efjlpuV] [-c 0|1] [-L label] [-O v5_feature] [-r rtdev] [-U uuid] device [logdev]"
> +USAGE="Usage: xfs_admin [-efjlpuV] [-c 0|1] [-L label] [-O v5_feature] [-r rtdev] [-U uuid] [mountpoint|device] [logdev]"

I don't think it's necessary to expand xfs_admin to take a mount point,
since findmnt returns the mount point associated with a block device.

>  
>  while getopts "c:efjlL:O:pr:uU:V" c
>  do
> @@ -23,7 +25,8 @@ do
>  	O)	REPAIR_OPTS=$REPAIR_OPTS" -c $OPTARG";;
>  	p)	DB_OPTS=$DB_OPTS" -c 'version projid32bit'";;
>  	r)	REPAIR_DEV_OPTS=" -r '$OPTARG'";;
> -	u)	DB_OPTS=$DB_OPTS" -r -c uuid";;
> +	u)	DB_EXTRA_OPTS=$DB_EXTRA_OPTS" -r -c uuid";
> +		IO_OPTS=$IO_OPTS" -r -c fsuuid";;
>  	U)	DB_OPTS=$DB_OPTS" -c 'uuid "$OPTARG"'";;
>  	V)	xfs_db -p xfs_admin -V
>  		status=$?
> @@ -38,14 +41,30 @@ set -- extra $@
>  shift $OPTIND
>  case $# in
>  	1|2)
> +		# Target either a device or mountpoint, not both
> +		if [ -n "$(findmnt -t xfs -T $1)" ]; then

Note that xfs_io acts upon a mountpoint, not a block device, so you
need to save the mount point that findmnt returns here.  You might also
want to look at the -o option to constrain its output to only the
information you need.

> +			if [ -n "$DB_OPTS" ] || [ -n "$REPAIR_OPTS" ]; then
> +				echo "Offline options target a device, not mountpoint."
> +				exit 2
> +			fi
> +			DB_EXTRA_OPTS=""
> +		else
> +			IO_OPTS=""

Hmm.  DB_EXTRA_OPS is really the container of xfs_db commands that also
have xfs_io alternatives, whereas DB_OPTS/REPAIR_OPTS are things that
require an unmounted fs.  Yet we can't access OPTIND until after we've
completely finished getopts processing.  I suspect this isn't a great
way to be handling this, because what does "EXTRA" mean, actually?  It's
not immediately obvious from the name.

I can't help but wonder if the behaviors would be clearer if we tracked
explicitly which subcommands require a mounted fs, which ones require an
unmounted fs (nearly all of them), and which ones can go back and forth.
Something like this:

require_offline=""
require_online=""
DB_OPTS=""
REPAIR_OPTS=""
IO_OPTS=""
<more initialization>

while getopts "c:efjlL:O:pr:uU:V" c
do
	case $c in
	c)
		REPAIR_OPTS=$REPAIR_OPTS" -c lazycount="$OPTARG
		require_offline=1
		;;

	<more options here>

	u)
		DB_OPTS=$DB_OPTS" -r -c uuid"
		IO_OPTS=$IO_OPTS" -c uuid"
		;;
	U)
		DB_OPTS=$DB_OPTS" -c 'uuid "$OPTARG"'"
		require_offline=1
		;;

	<rest of options here>
done

...and then later on when it's time to take action...

		if mntpt="$(findmnt -t xfs -f -n -o TARGET "$1" 2>/dev/null)"; then
			# filesystem is mounted
			if [ -n "$require_offline" ]; then
				echo "$1: filesystem is mounted."
				exit 2
			fi

			exec xfs_io -p xfs_admin "$IO_OPTS" "$mntpt"
		fi

		# filesystem is not mounted
		if [ -n "$require_offline" ]; then
			echo "$1: filesystem is not mounted"
			exit 2
		fi

		<regular DB_OPTS/REPAIR_OPTS processing>

--D

> +		fi
> +
>  		# Pick up the log device, if present
>  		if [ -n "$2" ]; then
>  			LOG_OPTS=" -l '$2'"
>  		fi
>  
> -		if [ -n "$DB_OPTS" ]
> +		if [ -n "$DB_OPTS" ] || [ -n "$DB_EXTRA_OPTS" ]
> +		then
> +			eval xfs_db -x -p xfs_admin $LOG_OPTS $DB_OPTS $DB_EXTRA_OPTS "$1"
> +			status=$?
> +		fi
> +		if [ -n "$IO_OPTS" ]
>  		then
> -			eval xfs_db -x -p xfs_admin $LOG_OPTS $DB_OPTS "$1"
> +			eval xfs_io -x -p xfs_admin $IO_OPTS "$1"
>  			status=$?
>  		fi
>  		if [ -n "$REPAIR_OPTS" ]
> diff --git a/man/man8/xfs_admin.8 b/man/man8/xfs_admin.8
> index 4794d677..2c7ddc15 100644
> --- a/man/man8/xfs_admin.8
> +++ b/man/man8/xfs_admin.8
> @@ -19,7 +19,11 @@ xfs_admin \- change parameters of an XFS filesystem
>  .B \-r
>  .I rtdev
>  ]
> +[
> +.I mount-point
> +|
>  .I device
> +]
>  [
>  .I logdev
>  ]
> -- 
> 2.25.1
> 

      reply	other threads:[~2022-12-20 22:29 UTC|newest]

Thread overview: 7+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2022-12-19 18:18 [PATCH v2 0/2] get UUID of mounted filesystems Catherine Hoang
2022-12-19 18:18 ` [PATCH v2 1/2] xfs_io: add fsuuid command Catherine Hoang
2022-12-20 21:55   ` Darrick J. Wong
2022-12-21  7:11     ` Catherine Hoang
2022-12-21 16:24       ` Darrick J. Wong
2022-12-19 18:18 ` [PATCH v2 2/2] xfs_admin: get UUID of mounted filesystem Catherine Hoang
2022-12-20 22:29   ` Darrick J. Wong [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=Y6I3Q28TAOv7VVET@magnolia \
    --to=djwong@kernel.org \
    --cc=catherine.hoang@oracle.com \
    --cc=linux-xfs@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