linux-raid.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: NeilBrown <neil@brown.name>
To: linux-raid@vger.kernel.org
Cc: Zhao Heming <heming.zhao@suse.com>,
	neilb@suse.com, jes@trained-monkey.org
Subject: Re: [PATCH] mdadm/Detail: show correct state for cluster-md array
Date: Mon, 20 Jul 2020 10:17:18 +1000	[thread overview]
Message-ID: <87blkbvy41.fsf@notabene.neil.brown.name> (raw)
In-Reply-To: <1595009398-5069-1-git-send-email-heming.zhao@suse.com>

[-- Attachment #1: Type: text/plain, Size: 3357 bytes --]

On Sat, Jul 18 2020, Zhao Heming wrote:

> After kernel md module commit 480523feae581, in md-cluster env,
> mddev->in_sync always zero, it will make array.state never set
> up MD_SB_CLEAN. it causes "mdadm -D /dev/mdX" show state 'active'
> all the time.
>
> bitmap.c: add a new API IsBitmapDirty() to support inquiry bitmap
> dirty or clean.
>
> Signed-off-by: Zhao Heming <heming.zhao@suse.com>
> ---
>  Detail.c | 21 ++++++++++++++++++++-
>  bitmap.c | 27 +++++++++++++++++++++++++++
>  mdadm.h  |  1 +
>  3 files changed, 48 insertions(+), 1 deletion(-)
>
> diff --git a/Detail.c b/Detail.c
> index 24eeba0..fd580a3 100644
> --- a/Detail.c
> +++ b/Detail.c
> @@ -495,8 +495,27 @@ int Detail(char *dev, struct context *c)
>  							  sra->array_state);
>  				else
>  					arrayst = "clean";
> -			} else
> +			} else {
>  				arrayst = "active";
> +				if (array.state & (1<<MD_SB_CLUSTERED)) {
> +					int dirty = 0;
> +					for (d = 0; d < max_disks * 2; d++) {
> +						char *dv;
> +						mdu_disk_info_t disk = disks[d];
> +
> +						if (d >= array.raid_disks * 2 &&
> +								disk.major == 0 && disk.minor == 0)
> +							continue;
> +						if ((d & 1) && disk.major == 0 && disk.minor == 0)
> +							continue;
> +						dv = map_dev_preferred(disk.major, disk.minor, 0, c->prefer);
> +						if (dv != NULL)
> +							if ((dirty = IsBitmapDirty(dv))) break;

You don't need to read the bitmap from every device - they should all be
the same.  Just reading one is sufficient.

However ....


> +					}
> +					if (dirty == 0)
> +						arrayst = "clean";
> +				}
> +			}
>  
>  			printf("             State : %s%s%s%s%s%s%s \n",
>  			       arrayst, st,
> diff --git a/bitmap.c b/bitmap.c
> index e38cb96..90ad932 100644
> --- a/bitmap.c
> +++ b/bitmap.c
> @@ -368,6 +368,33 @@ free_info:
>  	return rv;
>  }
>  
> +int IsBitmapDirty(char *filename)
> +{
> +	/*
> +	 * Read the bitmap file
> +	 * return: 1(dirty), 0 (clean), -1(error)
> +	 */
> +
> +	struct supertype *st = NULL;
> +	bitmap_info_t *info;
> +	int fd = -1, rv = -1;
> +
> +	fd = bitmap_file_open(filename, &st, 0);
> +	if (fd < 0)
> +		return rv;
> +
> +	info = bitmap_fd_read(fd, 0);

This only examines the first bitmap.  For a cluster-array there is a
separate bitmap for each node in the cluster.
See the
		for (i = 0; i < (int)sb->nodes; i++) {
loop in ExamineBitmap().  You need to check for active bits in every
bitmap on the selected device.

NeilBrown


> +	if (!info) {
> +		goto out;
> +	}
> +	rv = info->dirty_bits ? 1 : 0;
> +	free(info);
> +out:
> +	close(fd);
> +	free(st);
> +	return rv;
> +}
> +
>  int CreateBitmap(char *filename, int force, char uuid[16],
>  		 unsigned long chunksize, unsigned long daemon_sleep,
>  		 unsigned long write_behind,
> diff --git a/mdadm.h b/mdadm.h
> index 399478b..ba8ba91 100644
> --- a/mdadm.h
> +++ b/mdadm.h
> @@ -1447,6 +1447,7 @@ extern int CreateBitmap(char *filename, int force, char uuid[16],
>  			unsigned long long array_size,
>  			int major);
>  extern int ExamineBitmap(char *filename, int brief, struct supertype *st);
> +extern int IsBitmapDirty(char *filename);
>  extern int Write_rules(char *rule_name);
>  extern int bitmap_update_uuid(int fd, int *uuid, int swap);
>  
> -- 
> 2.25.0

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 832 bytes --]

  reply	other threads:[~2020-07-20  0:17 UTC|newest]

Thread overview: 4+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2020-07-17 18:09 [PATCH] mdadm/Detail: show correct state for cluster-md array Zhao Heming
2020-07-20  0:17 ` NeilBrown [this message]
  -- strict thread matches above, loose matches on Subject: below --
2020-07-17 17:51 Zhao Heming
2020-07-17 17:55 ` heming.zhao

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=87blkbvy41.fsf@notabene.neil.brown.name \
    --to=neil@brown.name \
    --cc=heming.zhao@suse.com \
    --cc=jes@trained-monkey.org \
    --cc=linux-raid@vger.kernel.org \
    --cc=neilb@suse.com \
    /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).