On Fri, 19 Mar 1999, Philip Blundell wrote:
> >You cannot do this with NFS 2. The protocol forbids blocksizes > 8K
>
> In fact I think Chris said he was using NBD not NFS anyway. So just ignore me.
I tried both; I was originally trying to configure /dev/nbd0 at the
top of rc.sysinit, but judging by by `echo' command at the top,
rc.sysinit is never being reached, and thus I can't configure nbd0
as swap, and hence I can never use it.
So, reasoning that perhaps the swap was exhausting too fast to even
invoke rc.sysinit; hence I went back to NFS, and tried `swap=/tmp/swap'
on the kernel command-line. However, this is now having its own
problems. I may attempt to incorporate /dev/nbd0 configuration into
sys_swapon(), or somesuch, and switch back to nbd again...
The NFS server is looking *very* dodgy; as I posted earlier, I'm
getting `IP reassembly time exceeded', and cranberry saying `server
not responding', and all sorts of stuff. I'll try to upgrade it.
--
Chris <[EMAIL PROTECTED]> ( http://www.fluff.org/chris )
unsubscribe: body of `unsubscribe linux-arm' to [EMAIL PROTECTED]
