[OpenAFS] MR-AFS [was: Re: OpenAFS Project List ]

Derrick J Brashear Derrick J Brashear <shadow@dementia.org>
Thu, 15 Mar 2001 12:55:11 -0500 (EST)

On Thu, 15 Mar 2001, Nathan Rawling wrote:

> > I think to include large file support into OpenAFS servers is easy, if
> > you use the NAMEI-interface instead of abusing inode fields. You then
> > also have the advantage to be completely independent of fsck and you can
> > dump /vicep-partitions. (The only problem is to find a place in the
> > vnode for the upper 32 bits of the file length.)
> I think the other benefits to this make it worthwhile. I hate maintaining
> a seperate fsck (and how many people have been burned that way?), and
> being able to dump /vice partitions would be nice. Even if it meant some
> nasty vos syncserv/syncvldbs at the end, "ufsdump" is WAY faster than "vos
> dump."
> I don't know if the concern is a performance hit, or if it's just storing
> extra info, but it definitely complicates things and makes it that much
> harder to work with.

Well, we can't just switch by default on existing platforms because it
would break compatibility, but it's a desired goal to have the NAMEI
fileserver working on all supported platforms, and for configure to have
an option to build that way (it's rather messy to build both a regular and
a namei fileserver, but it could be done; it's not vastly different than
the whole "t" versions for threaded environments. but... how many variants
do you build? where do you draw the line?)