At Tue, 21 Jan 2003 23:29:16 +0000 (UTC),
Masafumi NAKANE wrote:
> Well, not so quick. I took another -CURRENT box and the problem was
> not reproduceable on this box. I can't quite figure out what the
> differences between these two boxes are, though.
I think I found the solution.
On HEAD and RELENG_5_0 as of now
On RELENG_4 and old (e.g. last year's) HEAD, refering remote
filesystem via AMD retrieves export list from server and trying to
mount like this:
c->s MOUNT V1 EXPORT Call
s->c MOUNT V1 EXPORT Reply
c->s MOUNT V1 MNT Call
s->c MOUNT V1 MNT Reply
But on latest HEAD and RELENG_5_0, packets show like this:
c->s MOUNT V1 EXPORT Call
s->c MOUNT V1 EXPORT Reply
c->s MOUNT V1 MNT Call
s->c MOUNT V1 EXPORT Reply (duplicate)
So, problem may be in mountd. I tested with reverting the commit at
2003/01/15, and it works fine.
mbr 2003/01/15 23:27:31 PST
Modified files:
usr.sbin/mountd mountd.c
Log:
Implement nonblocking tpc-connections. rpcgen -m does still
produce backcompatible code.
I don't know the problem is in amd or mountd. I fear other programs
are broken with this commit...
--
Jun Kuriyama <[EMAIL PROTECTED]> // IMG SRC, Inc.
<[EMAIL PROTECTED]> // FreeBSD Project
To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-current" in the body of the message
