Date: Wed, 24 Jul 2002 16:13:37 -0700 From: Terry Lambert <tlambert2@mindspring.com> To: Bruce Evans <bde@zeta.org.au> Cc: Alfred Perlstein <bright@mu.org>, fs@FreeBSD.ORG Subject: Re: rename hardlinks "works" on FreeBSD, but no-op on others Message-ID: <3D3F34A1.5BD85141@mindspring.com> References: <20020724234432.P33567-100000@gamplex.bde.org>
next in thread | previous in thread | raw e-mail | index | archive | help
Bruce Evans wrote: > NFSv4 is fairly convincing. I'll clean up my patch and commit something > later this week. No, it's not. The *single* most important comment I made on 4 occasions was ignored. Section 2.2: ... An adjunct time maintenance protocol is recommended to reduce client and server time skew. ... 8.12. Clocks and Calculating Lease Expiration ... To avoid the need for synchronized clocks, lease times are granted by the server as a time delta. However, there is a requirement that the client and server clocks do not drift excessively over the duration of the lock. Jackasses. All they had to do was include the localtime on any request whose response could include a timestamp, and then send deltas instead of absolutes over the wire in all responses: LET local time since epoch RET remote time since epoch LFT local file time RFT remote filetime delta = LET - RET + LFT RFT = RET + delta And then all of the time synchronization requirements between nodes totally disappear. The assinine synchronization requirement was *the biggest issue* for NFSv2, remained *the biggest issue* for NFSv3, and they happily carried the albatross on their necks into the future with them, so they didn't have to disturb their "NFSv4 implementations" that they made before the standard was even a gleam in most people's eyes. And don't even get me started on character set normalization... Grrrr. 8-(. -- Terry To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-fs" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?3D3F34A1.5BD85141>