Date: Mon, 2 Jan 2012 07:40:19 GMT From: Benjamin Kaduk <kaduk@MIT.EDU> To: freebsd-doc@FreeBSD.org Subject: Re: docs/163742: [patch] document failok mount(8) option Message-ID: <201201020740.q027eJCX005639@freefall.freebsd.org>
next in thread | raw e-mail | index | archive | help
The following reply was made to PR docs/163742; it has been noted by GNATS.
From: Benjamin Kaduk <kaduk@MIT.EDU>
To: Garrett Cooper <yanegomi@gmail.com>
Cc: freebsd-gnats-submit@freebsd.org
Subject: Re: docs/163742: [patch] document failok mount(8) option
Date: Mon, 2 Jan 2012 02:36:45 -0500 (EST)
This message is in MIME format. The first part should be readable text,
while the remaining parts are likely unreadable without MIME-aware tools.
---559023410-854845245-1325489805=:882
Content-Type: TEXT/PLAIN; charset=ISO-8859-1; format=flowed
Content-Transfer-Encoding: QUOTED-PRINTABLE
On Sun, 1 Jan 2012, Garrett Cooper wrote:
> On Sun, Jan 1, 2012 at 10:27 PM, Benjamin Kaduk <kaduk@mit.edu> wrote:
>> On Sat, 31 Dec 2011, Garrett Cooper wrote:
>>
>>>> Description:
>>>
>>> The failok option isn't documented in mount(8). As noted in the
>>> sbin/mount/mount.c:
>>>
>>> 730 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 } else if (strcmp(p=
, "failok") =3D=3D 0) {
>>> 731 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 /*
>>> 732 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0=
* "failok" is used to prevent certain
>>> file
>>> 733 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0=
* systems from being causing the
>>> system to
>>> 734 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0=
* drop into single user mode in the
>>> boot
>>> 735 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0=
* cycle, and is not a real mount
>>> option.
>>> 736 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0=
*/
>>> 737 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 con=
tinue;
>>>
>>
>>
>> r222832 also introduced (some) documentation of failok in fstab.5 -- do =
you
>> feel that this is insufficient and it also needs to be documented in
>> mount.8?
>
> I look at mount*(8) before I look at fstab(5) as fstab(5) is far from
> complete. I'm sure I'm not the only one.
>
> That and I can actually use it via mount -o on the commandline makes
> more sense for it to be in mount(8) than fstab(5).
>
>>> The attached patch documents the missing option.
>
> ...
>
>> It seems hard to phrase this in an unconfusing manner.
>> I would probably say "this option is used to indicate that if an error
>> occurs during the mounting of a file system, the boot process should
>> continue on normally. =A0This is in contrast to the normal behavior wher=
e a
>> mount error will cause the system to drop into single-user mode."
>
> I didn't really wordsmith the original text too much, as you noticed
> above -- and quite frankly I wasn't happy with the text anyhow.
>
> The proposed sentences above get the meaning across, but in a bit
> roundabout manner. What about this instead?
>
> 'In the event that mounting a filesystem via rc fails, standard
> behavior is that the user is dropped into the 'administrative shell'
> to diagnose the mount failure. The failok option allows mount failures
> to pass.
>
> This option should be used on non-boot/-runtime critical filesystems.'
No major objections. The "non-boot/-runtime" is a bit awkward, so I might=
=20
prefer "should be used on file systems which are not critical for boot or=
=20
runtime usage."
Thanks,
Ben
---559023410-854845245-1325489805=:882--
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?201201020740.q027eJCX005639>