From owner-freebsd-current Thu Sep 18 07:58:06 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.7/8.8.7) id HAA10988 for current-outgoing; Thu, 18 Sep 1997 07:58:06 -0700 (PDT) Received: from lamb.sas.com (daemon@lamb.sas.com [192.35.83.8]) by hub.freebsd.org (8.8.7/8.8.7) with SMTP id HAA10980 for ; Thu, 18 Sep 1997 07:58:01 -0700 (PDT) Received: from mozart by lamb.sas.com (5.65c/SAS/Gateway/01-23-95) id AA05585; Thu, 18 Sep 1997 10:57:59 -0400 Received: from iluvatar.unx.sas.com by mozart (5.65c/SAS/Domains/5-6-90) id AA23560; Thu, 18 Sep 1997 10:53:43 -0400 From: "John W. DeBoskey" Received: by iluvatar.unx.sas.com (5.65c/SAS/Generic 9.01/3-26-93) id AA09453; Thu, 18 Sep 1997 10:53:43 -0400 Message-Id: <199709181453.AA09453@iluvatar.unx.sas.com> Subject: vx driver problem at 100T speeds To: freebsd-current@freebsd.org Date: Thu, 18 Sep 1997 10:53:43 -0400 (EDT) X-Mailer: ELM [version 2.4 PL23] Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-freebsd-current@freebsd.org X-Loop: FreeBSD.org Precedence: bulk Hello, I have been experiencing problems with nfs on current system(s) for a while and believe I may have narrowed the problem down to the vx driver... (3.0-970911-SNAP is the latest I've tested). In 100T Full Duplex (3com 905 pci , 959 pci card) 100T Half Duplex (3com 905 pci , 959 pci card) the system exhibits a problem of freezing up momentarily and then continuing, sometimes correctly, sometimes with an error. (Not just with nfs, but I'm using nfs almost exclusively, so it shows up the most there). In 10T Half Duplex (3com 905 pci , 959 pci card) the system works correctly, with no errors. In 100T Full Duplex (Intel 10/100 pci card) 100T Half Duplex (Intel 10/100 pci card) the system works correctly. with no errors. We attached a sniffer to the network, and found that the 3com cards are sending bad packets... Approximately 95% of the time, a mount_nfs command will cause a frame to be sent with an incorrect crc. This happens with the 3com card, but not the intel. The sniffer reports a bad crc, and then says that the frame length doesn't match the length given in the header (ie: short frame by 140 bytes). It is ALWAYS 140 bytes short. I've tested 3 different 3com cards: Motherboard 3c905 Adapter pci 3c905 Adapter pci 3c959 They all exhibit the same behaviour. What is the best way to track this down... I have access to the sniffer for the next week or so... Thanks, John -- jwd@NSunx.sas.com (w) John W. De Boskey (919) 677-8000 x6915