Skip site navigation (1)Skip section navigation (2)
Date:      Fri, 21 Jun 2019 14:27:09 -0600
From:      Scott Long <scottl@samsco.org>
To:        Alan Somers <asomers@freebsd.org>, Chuck Silvers <chs@netflix.com>, Kirk McKusick <mckusick@mckusick.com>
Cc:        FreeBSD CURRENT <freebsd-current@freebsd.org>
Subject:   Re: Reducing UFS corruption from unclean shutdowns?
Message-ID:  <F93827F6-1B99-4BDD-B245-C9594AD28ED7@samsco.org>
In-Reply-To: <CAOtMX2jXiaOWpVdEg3_nBYinJWd=iwN_38hQ4eMOocgs8dMWhQ@mail.gmail.com>
References:  <CAOtMX2jPut4ve-Tr7DyikxXqnmqycyjEUpNmAiwUSXbQrK3iCA@mail.gmail.com> <C3016BDF-4B51-4A59-94F2-CCBD0DC4562E@samsco.org> <CAOtMX2jXiaOWpVdEg3_nBYinJWd=iwN_38hQ4eMOocgs8dMWhQ@mail.gmail.com>

next in thread | previous in thread | raw e-mail | index | archive | help

> On Jun 21, 2019, at 2:09 PM, Alan Somers <asomers@freebsd.org> wrote:
>=20
> On Fri, Jun 21, 2019 at 1:56 PM Scott Long <scottl@samsco.org> wrote:
>>=20
>>=20
>>=20
>>> On Jun 21, 2019, at 1:49 PM, Alan Somers <asomers@freebsd.org> =
wrote:
>>>=20
>>> I panic my development VM regularly.  Each time, I need to fsck the
>>> file system.  Even if I had run sync(8) just before the panic, I
>>> frequently find corruption.  What should I change to make sync(8)
>>> work, or at least to make corruption rare?  It looks like my root =
file
>>> system is using soft-updates+journal.  Should I disable those?
>>>=20
>>=20
>> What corruption do you regularly see?
>>=20
>> Scott
>=20
> fsck reports various types of errors, all repairable, like "INODE
> CHECK-HASH FAILED", "FREE BLK COUNT(S) WRONG IN SUPERBLK", "SUMMARY
> INFORMATION BAD", "BLK(S) MISSING IN BIT MAPS", and "UNREF FILE".  If
> I don't run fsck, then I get errors when I try to access files.  Like
> "inode XXX: check-hash failed" and "such and such is marked as an
> executable file but could not be run by the operating system".
> -Alan

The freeblk count and summary information messages are normal and =
expected.  I
don=E2=80=99t think that the blks missing message is expected, and the =
unref file message is
definitely a red flag of something that should have been handed with =
journal
recovery.  Kirk and Chuck, do you have any insight here?

Thanks,
Scott




Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?F93827F6-1B99-4BDD-B245-C9594AD28ED7>