On Saturday, 5 October 2002 at 4:08:19 +0100, Ian Dowse wrote:
>
> [CCs trimmed]
>
>> The divide by zero problem seems to be caused by an interaction
>> between two bugs: GEOM refuses to return the sector size because
> ...
>> The next failure I get is:
>>
>> Can't write config to /dev/da1s1d, error 45 (EOPNOTSUPP)
>
> This turns out to be vinum doing a DIOCWLABEL to make the label
> writable before writing its configuration, but GEOM does not support
> that ioctl I presume. It should be safe to ignore these DIOCWLABEL
> ioctl return values as the actual writing of the vinum data should
> give a suitable error if making the label writable fails and is
> important. The patch below is Robert's patch with all 3 other issues
> fixed, and together, this seems to be enough to make vinum work
> again.
> @@ -678,20 +615,18 @@
> if ((drive->state != drive_unallocated)
> && (drive->state != drive_referenced)) { /* and it's a real drive
>*/
> wlabel_on = 1; /* enable writing the
>label */
> - error = (*devsw(drive->dev)->d_ioctl) (drive->dev, /* make the
>label writeable */
> + (void) (*devsw(drive->dev)->d_ioctl) (drive->dev, /* make the
>label writeable */
> DIOCWLABEL,
> (caddr_t) & wlabel_on,
> FWRITE,
> curthread);
> - if (error == 0)
> - error = write_drive(drive, (char *) vhdr, VINUMHEADERLEN,
>VINUM_LABEL_OFFSET);
> + error = write_drive(drive, (char *) vhdr, VINUMHEADERLEN,
>VINUM_LABEL_OFFSET);
> if (error == 0)
> error = write_drive(drive, config, MAXCONFIG,
>VINUM_CONFIG_OFFSET); /* first config copy */
> if (error == 0)
> error = write_drive(drive, config, MAXCONFIG,
>VINUM_CONFIG_OFFSET + MAXCONFIG); /* second copy */
> wlabel_on = 0; /* enable writing the
>label */
> - if (error == 0)
> - error = (*devsw(drive->dev)->d_ioctl) (drive->dev, /* make the
>label non-writeable again */
> + (void) (*devsw(drive->dev)->d_ioctl) (drive->dev, /* make the
>label non-writeable again */
> DIOCWLABEL,
> (caddr_t) & wlabel_on,
> FWRITE,
I don't know how GEOM handles disk labels, though we've heard a lot
about changes. It's possible that this code is now completely
redundant. It's preparing to write at offset 8 sectors and on from
the beginning of the partition; it would be worth checking what
happens if you just remove the whole disk label writeable stuff.
Greg
--
See complete headers for address and phone numbers
To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-current" in the body of the message
