linux-btrfs.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Anand Jain <Anand.Jain@oracle.com>
To: linux-btrfs@vger.kernel.org
Subject: Re: [PATCH] btrfs-progs: replace fails start but in the background v2
Date: Fri, 06 Sep 2013 17:39:35 +0800	[thread overview]
Message-ID: <5229A2D7.10501@oracle.com> (raw)
In-Reply-To: <1378090548-12719-1-git-send-email-anand.jain@oracle.com>


  this there is commit error. Kindly ignore this.
  I have sent out a new patch for this.

Thanks, Anand

On 09/02/2013 10:55 AM, Anand Jain wrote:
> when the balance is running, the replace start ioctl
> fails (for the right reasons). but since the cli has
> put ioctl thread to background (for right reasons)
> the user won't know that cli failed to start.
>
> so before cli goes to the background, it should check
> if mutually_exclusive_operation_running is not held.
>
> this is done by newly introduced ioctl
>   BTRFS_IOC_CHECK_DEV_EXCL_OPS by the following kernel patch:
>
> btrfs: introduce BTRFS_IOC_CHECK_DEV_EXCL_OPS ioctl to check dev excl op
>
> v2:
>   fix checkpatch.pl warnings as spotted by Wang
>
> Signed-off-by: Anand Jain <anand.jain@oracle.com>
> ---
>   cmds-device.c  |    4 ++--
>   cmds-replace.c |   14 ++++++++++++++
>   ioctl.h        |    1 +
>   man/btrfs.8.in |    4 ++--
>   utils.c        |   15 +++++++++++++++
>   utils.h        |    1 +
>   6 files changed, 35 insertions(+), 4 deletions(-)
>
> diff --git a/cmds-device.c b/cmds-device.c
> index 8446502..27f7f84 100644
> --- a/cmds-device.c
> +++ b/cmds-device.c
> @@ -45,7 +45,7 @@ static const char * const device_cmd_group_usage[] = {
>   };
>
>   static const char * const cmd_add_dev_usage[] = {
> -	"btrfs device add <device> [<device>...] <path>",
> +	"btrfs device add [-f] <device> [<device>...] <path>",
>   	"Add a device to a filesystem",
>   	NULL
>   };
> @@ -85,7 +85,7 @@ static int cmd_add_dev(int argc, char **argv)
>   		int	devfd, res;
>   		u64 dev_block_count = 0;
>   		int mixed = 0;
> -
> +printf("asj: add %s\n", argv[i]);
>   		res = test_dev_for_mkfs(argv[i], force, estr);
>   		if (res) {
>   			fprintf(stderr, "%s", estr);
> diff --git a/cmds-replace.c b/cmds-replace.c
> index e3ff695..0b2cbc8 100644
> --- a/cmds-replace.c
> +++ b/cmds-replace.c
> @@ -203,6 +203,20 @@ static int cmd_start_replace(int argc, char **argv)
>   		goto leave_with_error;
>   	}
>
> +	/* check if there is some other device exclusive
> +	 * operation running in the FS which won't let this replace
> +	 * to run. And ENOTTY is when older kernel doesn't support
> +	 * lock checking ioctl
> +	 */
> +	ret = is_dev_excl_op_free(fdmnt);
> +	if (ret && ret != -ENOTTY) {
> +		fprintf(stderr,
> +			"ERROR: replace start failed on \"%s\" - %s\n",
> +			path,
> +			ret > 0 ? btrfs_err_str(ret) : strerror(-ret));
> +		goto leave_with_error;
> +	}
> +
>   	srcdev = argv[optind];
>   	dstdev = argv[optind + 1];
>
> diff --git a/ioctl.h b/ioctl.h
> index e959720..afadcbc 100644
> --- a/ioctl.h
> +++ b/ioctl.h
> @@ -598,6 +598,7 @@ struct btrfs_ioctl_clone_range_args {
>   #define BTRFS_IOC_DEV_REPLACE _IOWR(BTRFS_IOCTL_MAGIC, 53, \
>   				    struct btrfs_ioctl_dev_replace_args)
>   #define BTRFS_IOC_DEDUP_CTL _IOW(BTRFS_IOCTL_MAGIC, 55, int)
> +#define BTRFS_IOC_CHECK_DEV_EXCL_OPS _IO(BTRFS_IOCTL_MAGIC, 56)
>
>   #ifdef __cplusplus
>   }
> diff --git a/man/btrfs.8.in b/man/btrfs.8.in
> index d493d7e..0cca5c3 100644
> --- a/man/btrfs.8.in
> +++ b/man/btrfs.8.in
> @@ -47,7 +47,7 @@ btrfs \- control a btrfs filesystem
>   \fBbtrfs\fP \fB[filesystem] balance status\fP [-v] \fI<path>\fP
>   .PP
>   .PP
> -\fBbtrfs\fP \fBdevice add\fP \fI<device>\fP [\fI<device>...\fP] \fI<path>\fP
> +\fBbtrfs\fP \fBdevice add\fP [-f] \fI<device>\fP [\fI<device>...\fP] \fI<path>\fP
>   .PP
>   \fBbtrfs\fP \fBdevice delete\fP \fI<device>\fP [\fI<device>...\fP] \fI<path>\fP
>   .PP
> @@ -386,7 +386,7 @@ be verbose
>   .RE
>   .TP
>
> -\fBdevice add\fR\fI <dev> \fP[\fI<dev>...\fP] \fI<path>\fR
> +\fBdevice add\fR [-f] \fI <dev> \fP[\fI<dev>...\fP] \fI<path>\fR
>   Add device(s) to the filesystem identified by \fI<path>\fR.
>   .TP
>
> diff --git a/utils.c b/utils.c
> index b039a03..c53b7f1 100644
> --- a/utils.c
> +++ b/utils.c
> @@ -1993,3 +1993,18 @@ int is_vol_small(char *file)
>   		return 0;
>   	}
>   }
> +
> +/* Returns:
> + * BTRFS_ERROR_DEV_EXCL_RUN_IN_PROGRESS:
> + *	If the device is locked to prevent other device operations
> + *	from the user cli like device add remove replace balance etc..
> + * < 0:
> + *	For any other error including if kernel don't support the
> + *	ioctl (-ENOTTY)
> + */
> +int is_dev_excl_op_free(int fd)
> +{
> +	int ret;
> +	ret = ioctl(fd, BTRFS_IOC_CHECK_DEV_EXCL_OPS, NULL);
> +	return ret > 0 ? ret : -errno;
> +}
> diff --git a/utils.h b/utils.h
> index eb6fba3..a1e5d67 100644
> --- a/utils.h
> +++ b/utils.h
> @@ -83,4 +83,5 @@ int test_num_disk_vs_raid(u64 metadata_profile, u64 data_profile,
>   	u64 dev_cnt, int mixed, char *estr);
>   int get_btrfs_mount(const char *dev, char *mp, size_t mp_size);
>   int is_vol_small(char *file);
> +int is_dev_excl_op_free(int fd);
>   #endif
>

  reply	other threads:[~2013-09-06  9:32 UTC|newest]

Thread overview: 9+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2013-08-21 13:10 [PATCH] btrfs: introduce BTRFS_IOC_CHECK_DEV_EXCL_OPS ioctl to check dev excl op Anand Jain
2013-08-21 13:10 ` [PATCH] btrfs-progs: replace fails start but in the background Anand Jain
2013-09-02  2:55   ` [PATCH] btrfs-progs: replace fails start but in the background v2 Anand Jain
2013-09-06  9:39     ` Anand Jain [this message]
2013-09-06  9:38   ` [PATCH] btrfs-progs: replace fails start but in the background Anand Jain
2013-08-21 13:58 ` [PATCH] btrfs: introduce BTRFS_IOC_CHECK_DEV_EXCL_OPS ioctl to check dev excl op Stefan Behrens
2013-08-22  9:51   ` anand jain
2013-08-22 10:36     ` Stefan Behrens
2013-08-23  6:45       ` Anand Jain

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=5229A2D7.10501@oracle.com \
    --to=anand.jain@oracle.com \
    --cc=linux-btrfs@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;
as well as URLs for NNTP newsgroup(s).