[OpenAFS-devel] Volume id range

[Derrick J Brashear]

On Thu, 19 Feb 2004, Kris Van Hees wrote:

> Hi,
>
> 	Is there any historic reason for why volume ids start at 0x20000000
> rather than simply starting from 0?  It seems to be a waste of a pretty large
> part of the id space, for no apparent reason?  Some backwards compatibility
> issue?

I'd *guess* because at some point the ITC was keeping volumes separated
based on generation of code, but that's a really wild, out-there guess.
The andrew.cmu.edu has low-numbered volumes currently in use. There are
basically different epochs in the ID space; I can easily discern at least
4:
0 (Craig Everhart's volume is id 125, root.cell is 48720, for instance)
something around 2000000 (user.fk03 is 33586502)
something around 3000000 (user.diskin is 50331921)
something around 70000000 (my volume is 1970660716)

Why? I don't know, and I'm not sure any of the people who might could
easily be asked.