kernel panic caused by zfs/sa.c

[Greg Bonett]

> > I'm experiencing a kernel panic that appears to be caused by zfs.
> >
> > No errors are making it into /var/log/messages, but here is the
> > error message that appears on my screen after panic (transcribed):
> >
> > panic solaris assert BSWAP_32(sa_hdr_phys->sa_magic) == SA_MAGIC,
> > file: /src/sys/modules/zfs/../../cddl/contrib/opensolaris/uts/common/fs/zfs/sa.c,
> > line 1262
> >
> > I can reproduce this by running "cd /usr/ports/www/webkit-ghtk2;
> > make clean"
> 
> If that's some sort of filesystem corruption, and you use portsnap, do
> you still have a panic after "portsnap extract www/webkit-gtk2 && cd
> /usr/ports/www/webkit-gtk2 && make clean" ?
> 

Thanks for the suggestions, but "portsnap extract www/webkit-gtk2"
causes a panic with the same error. Any other ideas?

Also, in case it's relevant - this zpool is comprised of geli devices.

Thanks.

> > I looked at line 1262 in sa.c and it's part of a bit of code that
> > checks if byte swapping is necessary. The code starting at line
> > 1260:
> >
> >
> > if (IS_SA_BONUSTYPE(bonustype) && sa_hdr_phys->sa_magic != SA_MAGIC
> > && sa_hdr_phys->sa_magic != 0) {
> >                         VERIFY(BSWAP_32(sa_hdr_phys->sa_magic) ==
> > SA_MAGIC); sa_byteswap(hdl, buftype);
> >                 }
> >
> > As far as I can tell, my source is up to date and this code is the
> > same as what is in 9-STABLE.
> >
> > I'm not sure what's causing the panic, does anyone have a
> > suggestion?
> >
> > Thanks for the help.
> >
> > --Greg
> >
> >