Date: Fri, 23 Nov 2007 15:57:21 -0600 From: Jonathan Horne <freebsd@dfwlp.com> To: freebsd-questions@freebsd.org Subject: still fumbling around with poor network speed Message-ID: <20071123155721.n76dj282s4ocw80w@webmail.dfwlp.org>
next in thread | raw e-mail | index | archive | help
ive been chasing this problem around for about a week now. ive got a =20 7.0 betaX box that started out as a beta 1.5. no problems with =20 anything. then, i upgraded it (via cvs) to 2.0, and i instantly noticed a =20 problem with serving the /usr volume via NFS to my laptop. moving my =20 workstation to 3.0 didnt help either. the 7.0beta box showed outbound =20 traffic about 200k. exact same behavior today, accessing it via SMB =20 from an OSX mac. strange thing is, when i use scp to transfer large files, they go at =20 almost 200mbit. as far as i know, scp uses TCP only. but NFS and =20 SMB, do they not use UDP? and if so, what can i do to try to =20 troubleshoot how poorly UDP packets are working? ive tried changing the cable, changing the switch port, and its the =20 same thing each time. i have one other 7.0 server, thats not =20 experiencing this either (but it also has a different ethernet card. as i type this, im wondering if this could be related to the bge =20 driver? i know many people have complained about bge in the past, and =20 i never seemed to have noticed anything wrong with mine... but could =20 that have changed for me as time passes and sourcecode evolves? if anyone can help shed light on this, id really appreciate it. --=20 Jonathan Horne http://dfwlpiki.dfwlp.org freebsd@dfwlp.com ---------------------------------------------------------------- This message was sent using IMP, the Internet Messaging Program.
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20071123155721.n76dj282s4ocw80w>