Re: dk for z/lvm volumes?

[Chavdar Ivanov <ci4ic4%gmail.com@localhost>]

On Mon, 29 Jun 2020 at 17:08, Staffan Thomén <duck%shangtai.net@localhost> wrote:
>
> Chavdar Ivanov wrote:
> >> # ccdconfig -c ccd0 0 /dev/zvol/dsk/internal1/photos@backup
> >
> > You are mapping a snapshot, it will be r/o.
> >
> > You have to promote it first.
>
> While it's true that the snapshot is r/o, surely it shouldn't crash the system.

Sure,

>
> And is there a reason why ccd would require r/w?
>
> Staffan
>
Well, I got the same crash:

uvm_fault(0xffff847d4c69ae58, 0x0, 1) -> e
fatal page fault in supervisor mode
trap type 6 code 0 rip 0xffffffff80d648cc cs 0x8 rflags 0x10246 cr2
0x18 ilevel 0 rsp 0xffff94819c763d80
curlwp 0xffff847d2ec79940 pid 10964.10964 lowest kstack 0xffff94819c7602c0
panic: trap
cpu0: Begin traceback...
vpanic() at netbsd:vpanic+0x152
snprintf() at netbsd:snprintf
startlwp() at netbsd:startlwp
alltraps() at netbsd:alltraps+0xc3
sysctl_dispatch() at netbsd:sysctl_dispatch+0xa5
sys___sysctl() at netbsd:sys___sysctl+0xc5
syscall() at netbsd:syscall+0x26e
--- syscall (number 202) ---
netbsd:syscall+0x26e:
cpu0: End traceback...

when I tried ccdconfig onto the snapshot; when I cloned the zvol (not
promote, my mistake), it works:

#  zfs clone  pail/nbsd1@testa  pail/nbsd2
#  ccdconfig -c ccd0 0 /dev/zvol/dsk/pail/nbsd2
# crash  ccdconfig -g
ccd0         0       0x0     4294967296      /dev/zvol/dsk/pail/nbsd2

I'd say the zvol snapshot is perhaps only relevant to ZFS internal
operations; you have to have a real zvol to be able to access it with
any external to ZFS tools, like istgt or ccdconfig. As far as I
remember,  even under Solaris you couldn't do much with a snapshot of
a zvol, other than clone it or send it.

But the panic should be examined, of course.

-- 
----