Re: ZFS errors with two HEX numbers.

From: Zaphod Beeblebrox <zbeeble_at_gmail.com>
Date: Sun, 26 Sep 2021 21:31:18 UTC
 Entertaining.  That (even with only two --> zdb -dd vr1/0x57b2) gives ...
core dump.  SIGSEGV in strlen.  Dereferencing a NULL.

(gdb) bt
#0  strlen (str=0xd <error: Cannot access memory at address 0xd>) at
/usr/src/lib/libc/string/strlen.c:101
#1  0x0000000800f6ea60 in __vfprintf (fp=<optimized out>,
locale=0x80102eed8 <__xlocale_global_locale>,
    fmt0=<optimized out>, ap=<optimized out>) at
/usr/src/lib/libc/stdio/vfprintf.c:854
#2  0x0000000800f6b8df in snprintf (str=0x801644400 "", n=<optimized out>,
fmt=0x8009ad9b9 "%s/%s")
    at /usr/src/lib/libc/stdio/snprintf.c:74
#3  0x0000000800a58b76 in zfs_file_open (path=path@entry=0x801644000
"/etc/zfs/zpool.cache", flags=<optimized out>,
    flags@entry=0, mode=<optimized out>, mode@entry=0, fpp=fpp@entry
=0x7fffffe58788)
    at /usr/src/sys/contrib/openzfs/lib/libzpool/kernel.c:1075
#4  0x0000000800b3e622 in spa_config_load () at
/usr/src/sys/contrib/openzfs/module/zfs/spa_config.c:101
#5  0x0000000800b47b39 in spa_init (mode=mode@entry=SPA_MODE_READ)
    at /usr/src/sys/contrib/openzfs/module/zfs/spa_misc.c:2393
#6  0x0000000800a586ae in kernel_init (mode=1) at
/usr/src/sys/contrib/openzfs/lib/libzpool/kernel.c:844
#7  0x0000000000405984 in ?? ()
#8  0x00000000004052df in ?? ()
#9  0x0000000800637008 in ?? ()
#10 0x0000000000000000 in ?? ()

On Sun, Sep 26, 2021 at 12:53 AM Rich <rincebrain@gmail.com> wrote:

> I don't have a testpool handy to confirm it, but I would expect you to
> be able to do something like
> zdb -ddddd [pool]/[objsetid in hex] [objectid in hex]
> e.g.
> zdb -ddddd vr1/0x57b2 0x73ab46
>
> Or failing that, drop the object id (and then probably some of the -d,
> that gets real verbose).
>
> - Rich
>
> On Sun, Sep 26, 2021 at 12:27 AM Zaphod Beeblebrox <zbeeble@gmail.com>
> wrote:
> >
> > How do I find which snapshot it's in?
> >
> > On Sat, Sep 25, 2021 at 6:01 PM Miroslav Lachman <000.fbsd@quip.cz>
> wrote:
> >
> > >
> > >
> > > I have seen this in the past. It was corruption on some deleted file
> > > preserved on snapshot. Once I deleted snapshots this error disappeared.
> > >
> > > [...]
> > >
> > > Miroslav Lachman
> > >
>