[OpenAFS-devel] [OSX] Service startup proposal

Derrick J Brashear shadow@dementia.org
Mon, 13 Mar 2006 11:38:53 -0500 (EST)


afsd could be that daemon, and i even have partially written code 
somewhere

On Mon, 13 Mar 2006, Roland Kuhn wrote:

> Hi Derek!
>
> On 13 Mar 2006, at 16:52, Derek Atkins wrote:
>
>> Roland Kuhn <rkuhn@e18.physik.tu-muenchen.de> 
>> writes:
>> 
>>> Question to the gurus: How difficult would it 
>>> be to make AFS behave
>>> when faced with loss of networking? What I'm 
>>> thinking about is a mode
>>> where you get EIO (or something) for all 
>>> requests concerning AFS
>>> instead of having to wait three minutes for an 
>>> answer which is
>>> guaranteed not to come. This can easily be 
>>> found out using some cocoa
>>> interface (forgot the name of the function). It 
>>> would also be nice to
>>> be able to unmount AFS _after_ the network 
>>> problems have started.
>> 
>> There are some patches in AFS to do this..  In 
>> particular in RX.  But
>> because RX is based on UDP, it requires that the 
>> UDP layer actually
>> return an ICMP Host Unreachable message back to 
>> the caller stack.
>> Most OSes do not return this type of error, so 
>> there's no feedback
>> that the sending failed.
>> 
>> On linux this error IS returned, so AFS does 
>> time out much quicker
>> when the network goes away.
>> 
> This mechanism goes in the right direction. 
> However, I was thinking about MacOS X. It would be 
> nice if OpenAFS could be sent into a dumb 
> 'disconnected' mode (only replying EIO to 
> everything immediately) by a daemon which monitors 
> the corresponding COCOA framework. Of course this 
> catches only the laptop "I pulled my cable" case, 
> but if it would be relatively easy...
>
> Ciao,
>                    Roland
>
> --
> TU Muenchen, Physik-Department E18, 
> James-Franck-Str., 85748 Garching
> Telefon 089/289-12575; Telefax 089/289-12570
> --
> CERN office: 892-1-D23 phone: +41 22 7676540 
> mobile: +41 76 487 4482
> --
> UNIX was not designed to stop you from doing 
> stupid things, because that
> would also stop you from doing clever things.
> 	-Doug Gwyn
> -----BEGIN GEEK CODE BLOCK-----
> Version: 3.12
> GS/CS/M/MU d-(++) s:+ a-> C+++ UL++++ P+++ L+++ 
> E(+) W+ !N K- w--- M+ !V Y+
> PGP++ t+(++) 5 R+ tv-- b+ DI++ e+++>++++ h---- 
> y+++
> ------END GEEK CODE BLOCK------
>
>
>
>
>