Re: [PATCH 2/2] imsm: FIX: Be more patient during loading matadata

[Dan Williams <dan.j.williams@intel.com>]

On Tue, Apr 12, 2011 at 5:51 AM, Adam Kwolek <adam.kwolek@intel.com> wrote:
> Sometimes occurs that metadata cannot be loaded e.g. wrong check sum
> It can happen due to metadata update racing with mdmon condition.
> If mpb loading is tried again, it is loaded successfully.
> Try to load metadata again before really giving up.
>
> Signed-off-by: Adam Kwolek <adam.kwolek@intel.com>
> ---
>
>  super-intel.c |   10 ++++++++--
>  1 files changed, 8 insertions(+), 2 deletions(-)
>
> diff --git a/super-intel.c b/super-intel.c
> index dc5e34e..d23267a 100644
> --- a/super-intel.c
> +++ b/super-intel.c
> @@ -2773,8 +2773,14 @@ load_and_parse_mpb(int fd, struct intel_super *super, char *devname, int keep_fd
>        int err;
>
>        err = load_imsm_mpb(fd, super, devname);
> -       if (err)
> -               return err;
> +       if (err) {
> +               /* try to load mpb again,
> +                * in case of mdmon race we could have more luck...
> +                */
> +               err = load_imsm_mpb(fd, super, devname);
> +               if (err)
> +                       return err;
> +       }
>        err = load_imsm_disk(fd, super, devname, keep_fd);
>        if (err)
>                return err;

This is semi-duplicates the check we already do after returning from
load_and_parse_mpb in load_super_imsm_all.  I'm curious, are you
hitting this path from load_super_imsm?  If the container is assembled
we should be loading from the container, if the container is not
available then mdmon can't be running and checksum errors are real.

--
Dan
--
To unsubscribe from this list: send the line "unsubscribe linux-raid" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html