linux-xfs.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: "Darrick J. Wong" <darrick.wong@oracle.com>
To: Eric Sandeen <sandeen@sandeen.net>
Cc: linux-xfs@vger.kernel.org
Subject: Re: [PATCH 1/5] xfs_spaceman: always report sick metadata, checked or not
Date: Fri, 1 Nov 2019 11:40:10 -0700	[thread overview]
Message-ID: <20191101184010.GY15222@magnolia> (raw)
In-Reply-To: <877be5a9-d19c-cecf-10d7-e4ea3626d0f5@sandeen.net>

On Fri, Nov 01, 2019 at 01:17:11PM -0500, Eric Sandeen wrote:
> On 10/22/19 1:46 PM, Darrick J. Wong wrote:
> > From: Darrick J. Wong <darrick.wong@oracle.com>
> > 
> > If the kernel thinks a piece of metadata is bad, we must always report
> > it.  This will happen with an upcoming series to mark things sick
> > whenever we return EFSCORRUPTED at runtime.
> > 
> > Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com>
> 
> I gotta say, I find this all really hard to read - something I should
> have commented on earlier.  Masks, maps, and functions oh my.  bad and
> sick and checked ... reported++ with no actual reporting ....

Hm, yeah, the "reported" variable here counts the number of things the
kernel reported to us as having been checked or marked sick at some point.
The whole point of that is that if the kernel hasn't marked anything
checked or sick then we really can't say much about the state of the
filesystem and maybe the user should run xfs_scrub.

> I'll try to think about what would make my poor brain happier later.
> Comments, for one I think.  Maybe some bikeshedding over variable names.
> 
> I guess the upshot here is that if it's marked sick due to the kernel
> sumbling over corruption, report it whether or not we ever explicitly
> /asked/ for a check via the scrub interfaces?

Correct.

--D

> Reviewed-by: Eric Sandeen <sandeen@redhat.com>
> 
> > ---
> >  spaceman/health.c |    4 ++--
> >  1 file changed, 2 insertions(+), 2 deletions(-)
> > 
> > 
> > diff --git a/spaceman/health.c b/spaceman/health.c
> > index 8fd985a2..0d3aa243 100644
> > --- a/spaceman/health.c
> > +++ b/spaceman/health.c
> > @@ -171,10 +171,10 @@ report_sick(
> >  	for (f = maps; f->mask != 0; f++) {
> >  		if (f->has_fn && !f->has_fn(&file->xfd.fsgeom))
> >  			continue;
> > -		if (!(checked & f->mask))
> > +		bad = sick & f->mask;
> > +		if (!bad && !(checked & f->mask))
> >  			continue;
> >  		reported++;
> > -		bad = sick & f->mask;
> >  		if (!bad && quiet)
> >  			continue;
> >  		printf("%s %s: %s\n", descr, _(f->descr),
> > 

  reply	other threads:[~2019-11-01 18:40 UTC|newest]

Thread overview: 20+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2019-10-22 18:46 [PATCH 0/5] xfsprogs: random fixes Darrick J. Wong
2019-10-22 18:46 ` [PATCH 1/5] xfs_spaceman: always report sick metadata, checked or not Darrick J. Wong
2019-11-01 18:17   ` Eric Sandeen
2019-11-01 18:40     ` Darrick J. Wong [this message]
2019-11-01 18:44       ` Darrick J. Wong
2019-10-22 18:46 ` [PATCH 2/5] xfs_db: btheight should check geometry more carefully Darrick J. Wong
2019-11-01 18:21   ` Eric Sandeen
2019-10-22 18:46 ` [PATCH 3/5] xfs_scrub: report repair activities on stdout, not stderr Darrick J. Wong
2019-11-01 18:26   ` Eric Sandeen
2019-11-01 18:42     ` Darrick J. Wong
2019-10-22 18:46 ` [PATCH 4/5] xfs_scrub: don't allow error or negative error injection interval Darrick J. Wong
2019-11-01 18:31   ` Eric Sandeen
2019-10-22 18:47 ` [PATCH 5/5] libfrog: fix workqueue_add error out Darrick J. Wong
2019-11-01 18:33   ` Eric Sandeen
2019-10-30 17:52 ` [PATCH 6/5] xfs_repair: print better information when metadata updates fail Darrick J. Wong
2019-11-01 18:42   ` Eric Sandeen
2019-10-30 17:53 ` [PATCH 7/5] libxfs: fix typo in message about write verifier Darrick J. Wong
2019-11-01 18:44   ` Eric Sandeen
2019-11-01 18:52 ` [PATCH 8/5] mkfs: fix incorrect error message Darrick J. Wong
2019-11-01 19:28   ` Eric Sandeen

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=20191101184010.GY15222@magnolia \
    --to=darrick.wong@oracle.com \
    --cc=linux-xfs@vger.kernel.org \
    --cc=sandeen@sandeen.net \
    /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).