Date: Mon, 9 Feb 2009 21:45:23 +0100 From: Martin <nakal@web.de> To: Dan Nelson <dnelson@allantgroup.com> Cc: sean.bruno@dsl-only.net, freebsd-current@freebsd.org, sbruno@freebsd.org Subject: Re: UFS Witness LoR + 5 other LoRs Message-ID: <20090209214523.519ee4d6@zelda.local> In-Reply-To: <20090208182023.GF85840@dan.emsphone.com> References: <1233007263.9302.2.camel@localhost.localdomain> <20090129233220.1ed64e6d@zelda.local> <498EB79F.4010905@FreeBSD.org> <20090208130506.267a838d@zelda.local> <20090208182023.GF85840@dan.emsphone.com>
next in thread | previous in thread | raw e-mail | index | archive | help
Am Sun, 8 Feb 2009 12:20:23 -0600 schrieb Dan Nelson <dnelson@allantgroup.com>: > In the last episode (Feb 08), Martin said: > > Btw... it would be very nice if someone finally implements timeouts > > and a detection strategy for NFS packets that don't arrive at their > > destination because of fragmentation and wrong rsize/wsize > > settings. But this is a totally different topic. There is not > > much in the docs about it. > > The solution to that problem is TCP mounts :) Hi Dan. Yes. This could be right, of course. I haven't tried it yet, but I will. There are two points, I want to add: 1) Some people say that UDP mounts have faster transfer rates. I don't know yet, if it's true. 2) I mention the problem with NFS over UDP, because UDP mounts are somehow "fishy", it seems. A robust piece of software should recover from erroneous situations and not get stuck somewhere. The only solution to this situation is an unclean reboot. This is not "nice". -- Martin
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20090209214523.519ee4d6>