Re: [fuse-devel] mount.fuse command line options

[Hans B. <han...@gm...>]

On Wed, Dec 11, 2013 at 1:56 PM, Miklos Szeredi <mi...@sz...> wrote:
> On Thu, Nov 28, 2013 at 9:56 PM, Arthur van Dongen
> <sou...@av...> wrote:
>> Hello all,
>>
>> In the Unanswered Questions section in the Wiki there is the question to
>> pass fuse helper script options from fstab. I think I've found a way to
>> implement this, without breaking existing configurations.
>>
>> My configuration, which triggered me to solve this, is to automount an
>> encrypted directory using fuse.encfs. The password for the encrypted
>> volume can be supplied by an executable which is indicated by the
>> --extpass option to fuse.encfs. (That the password is stored on this
>> system is not an issue, the encrypted directory is to be protected when
>> stored off-site)
>>
>> The current mount.fuse usage says:
>>   mount.fuse source destination -t type [-o opt[,opt...]]
>> Every option given is added to the -o opt,opts list passed to the
>> mount.type command.
>>
>> I modified the mount.fuse source code so the options provided can be
>> treated in two ways. Options without starting dash are added to the
>> comma-separated options list as normal. Options with a starting dash are
>> added to the command line as separate command line parameters.
>>
>> This way I can provide the necessary command line options for
>> mount.encfs on the options list of the autofs configuration. Haven't
>> tried it, but options in fstab should be possible now too.
>>
>> I've made a fork in sourceforge, see
>> http://sourceforge.net/u/afvandongen/fuse/ci/master/tree/ . The only
>> file I changed is util/mount.fuse.c, the diff can be found under
>> http://sourceforge.net/u/afvandongen/fuse/ci/master/tree/util/mount.fuse.c?diff=e6e7a249c5c70187144dd9cd3767801ed05a1763
>>
>> Please have a look at it, any feedback about this idea and my
>> implementation is welcome.
>
> Why not fix encfs instead to allow the "-oextpass=..." form too?
> Seems less hackish to me.
>
> I see that you are trying to find a generic solution, but I don't
> think this is a problem, except for a handfull of cases.
>
I second Miklos thoughts here. Actually, I received various proposals
on how to make my fuse mount properly through /etc/fstab.
I soon realized that this was not a fuse problem but rather a problem
in my own fs implementation. I did not properly take use of the
argument parsing functions as provided by the fuse API. When I did, I
got all this support for free. Actually, you do not need to add
explicit support for eg. -oextpass in encfs! When using the fuse API
routines for this also constructs like this "-o ro,--extpass,..." work
just as well.  I believe fuse is treating its native switches (like
-f) in the same way. I would report this issue upstream towards encfs
instead and let it improve its argument parsing logic.

Hans


> Thanks,
> Miklos
>
> ------------------------------------------------------------------------------
> Rapidly troubleshoot problems before they affect your business. Most IT
> organizations don't have a clear picture of how application performance
> affects their revenue. With AppDynamics, you get 100% visibility into your
> Java,.NET, & PHP application. Start your 15-day FREE TRIAL of AppDynamics Pro!
> http://pubads.g.doubleclick.net/gampad/clk?id=84349831&iu=/4140/ostg.clktrk
> _______________________________________________
> fuse-devel mailing list
> fus...@li...
> https://lists.sourceforge.net/lists/listinfo/fuse-devel