From owner-freebsd-stable Thu Apr 13 16:43:10 2000 Delivered-To: freebsd-stable@freebsd.org Received: from evil.2y.net (ztown1-1-38.adsl.one.net [207.78.254.38]) by hub.freebsd.org (Postfix) with ESMTP id 64EA137BE26; Thu, 13 Apr 2000 16:42:46 -0700 (PDT) (envelope-from cokane@evil.2y.net) Received: (from cokane@localhost) by evil.2y.net (8.9.3/8.9.3) id TAA07707; Thu, 13 Apr 2000 19:47:39 -0400 (EDT) (envelope-from cokane) Date: Thu, 13 Apr 2000 19:47:39 -0400 From: Coleman Kane To: Chris Wasser Cc: questions@FreeBSD.ORG, stable@FreeBSD.ORG Subject: Re: Serious NETWORK PROBLEMS Message-ID: <20000413194739.B7655@cokane.yi.org> References: <20000413101951.B5040@area51.v-wave.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Mailer: Mutt 1.0.1i In-Reply-To: <20000413101951.B5040@area51.v-wave.com>; from cwasser@v-wave.com on Thu, Apr 13, 2000 at 12:25:38PM -0400 X-Vim: vim:tw=70:ts=4:sw=4 Sender: owner-freebsd-stable@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.ORG I have had trouble with the PNIC cards before, in any OS. I dunno why. -- cokane Chris Wasser had the audacity to say: > Gonna keep reposting this until someone answers I guess. > > This was previously a AMD-K7 machine (4.0-STABLE) but has been moved over > to a Intel based machine for the time being while I get a new motherboard > (the K7 motherboard is shot -- the other K7 box runs like a charm.) > > Running 4.0-STABLE from Wednesday, April 12th. There are two Macronix PNIC > II's in the machine, one at 10Mbps/Half-Duplex(dc0) and the other at > 100Mbps/Full-Duplex(dc1) (both NICs use the dc? driver). On the 100Base-TX > side of things, during any sort of heavy network load I get: > > Apr 12 13:00:45 cmdmicro /kernel: dc1: TX underrun -- increasing TX > threshold > Apr 12 13:29:17 cmdmicro /kernel: dc1: TX underrun -- increasing TX > threshold > Apr 12 17:25:50 cmdmicro /kernel: dc1: TX underrun -- increasing TX > threshold > Apr 12 17:25:50 cmdmicro /kernel: dc1: TX underrun -- using store and > forward mode > Apr 12 17:25:50 cmdmicro /kernel: dc1: TX underrun -- using store and > forward mode > > Now from what I was reading on the older archives, these messages aren't > all that big a deal if appearing in sparse amounts, however I have a > 3.3-STABLE machine with the same make/model of network cards on the same > machine with no such messages popping up (slower speed processor as > well). > > On the 3.3-S machine I'm able to sustain 11MB/s transfer rates to and from > the network (it's switched) with no problems, however the 4.0-S machine > barely breaks 8.0MB/s on the same test. I originally had 3c905C-TX's in > the 4.0 box but they ended up shitting oatmeal under heavy network loads > (no big surprise there) > > The 4.0-STABLE machine has only been up a few hours now: > > 8:33PM up 7:43, 3 users, load averages: 0.00, 0.00, 0.00 > > It should be noted that net.inet.tcp.delayed_ack is at 1 on the 4.0-STABLE > machine and 0 on the 3.3-STABLE (I've recently just set it to 0 to see if > it's made any difference) > > Machine specs: > -------------- > 3.3-STABLE (1999/09/16) > P3-450 / 128MB > 1 x LNE100TX v2.0 Macronix PNIC II (mx? driver) @ 10Mbps > 1 x LNE100TX v2.0 Macronix PNIC II (mx? driver) @ 100Mbps Full-Duplex > > 4.0-STABLE (2000/04/12) > P3-550 / 256MB > 1 x LNE100TX v2.0 Macronix PNIC II (dc? driver) @ 10Mbps > 1 x LNE100TX v2.0 Macronix PNIC II (dc? driver) @ 100Mbps Full-Duplex > > 100Mbps side of the network is switched (LAN). The "test" was conducted by > xferring a 7GB tarball between the two machines via ftp (see ftp notes > below) Except for processor and ram, these two machines are identical > hardware-wise including BIOS setup/version# (and I do mean identical, all > the hardware except processor and ram are identical make & model right > down to the motherboard revision) > > One of the first things I checked was if there was shared PCI irqs, this > is not the case, each device has it's own IRQ. USB is disabled. > > * FTP NOTES: tests were conducted with stock ftp client in 4.0-STABLE in > active/PORT mode (not PASSIVE). There seems to be a problem with the stock > ftp client with large file sizes. It reported the 7GB tarball as -11TB > (when I did a ls of the file) and ofcourse messes up the bar graph > display. It also stalls for a second when it reaches the end of the bar > graph, but continues without error. It also doesn't show the correct file > size during transfer (as in total bytes to xfer) but it does xfer the file > properly nevertheless, I tested the tarball afterwards. I checked my own > 4.0-STABLE machine at home, the stock ftp client exhibits the same > behavior. > > I appriciate _any_ response (good or bad) as I need to get this worked out > as soon as possible (I'd like to replace our 3.3-STABLE machine) > > 4.0-STABLE information (space-adjusted for word-wrap purposes): > > Copyright (c) 1992-2000 The FreeBSD Project. > Copyright (c) 1982, 1986, 1989, 1991, 1993 > The Regents of the University of California. All rights reserved. > FreeBSD 4.0-STABLE #0: Wed Apr 12 12:38:04 MDT 2000 > root@:/usr/src/sys/compile/DEEPTHOUGHT > Timecounter "i8254" frequency 1193182 Hz > Timecounter "TSC" frequency 551252571 Hz > CPU: Pentium III/Pentium III Xeon (551.25-MHz 686-class CPU) > Origin = "GenuineIntel" Id = 0x673 Stepping = 3 > Features=0x383f9ff ,PGE,MCA,CMOV,PAT,PSE36,MMX,FXSR,XMM> > real memory = 268435456 (262144K bytes) > config> q > avail memory = 258334720 (252280K bytes) > Preloaded elf kernel "kernel" at 0xc0296000. > Preloaded userconfig_script "/boot/kernel.conf" at 0xc029609c. > VESA: v2.0, 8192k memory, flags:0x1, mode table:0xc0249ba2 (1000022) > VESA: ATI RAGE128 > Pentium Pro MTRR support enabled > npx0: on motherboard > npx0: INT 16 interface > pcib0: on motherboard > pci0: on pcib0 > pcib1: at device 1.0 on pci0 > pci1: on pcib1 > pci1: at 0.0 irq 5 > isab0: at device 7.0 on pci0 > isa0: on isab0 > atapci0: port 0xf000-0xf00f at device 7.1 > on pci0 > ata0: at 0x1f0 irq 14 on atapci0 > pci0: at 7.2 > chip1: port 0x5000-0x500f > at device 7.3 on pci0 > dc0: port 0xa400-0xa4ff > mem 0xe5000000-0xe50000ff irq 10 at device 11.0 on pci0 > dc0: Ethernet address: 00:a0:cc:32:eb:7c > miibus0: on dc0 > dcphy0: on miibus0 > dcphy0: 10baseT, 10baseT-FDX, 100baseTX, 100baseTX-FDX, auto > dc1: port 0xa800-0xa8ff > mem 0xe5001000-0xe50010ff irq 11 at device 15.0 on pci0 > dc1: Ethernet address: 00:a0:cc:33:7b:ab > miibus1: on dc1 > dcphy1: on miibus1 > dcphy1: 10baseT, 10baseT-FDX, 100baseTX, 100baseTX-FDX, auto > atapci1: > port 0xb400-0xb4ff,0xb000-0xb003,0xac00-0xac07 irq 15 at device 19.0 on pci0 > ata2: at 0xac00 on atapci1 > atapci2: > port 0xc000-0xc0ff,0xbc00-0xbc03,0xb800-0xb807 irq 15 at device 19.1 on pci0 > ata3: at 0xb800 on atapci2 > fdc0: at port 0x3f0-0x3f5,0x3f7 irq 6 drq 2 on isa0 > fdc0: FIFO enabled, 8 bytes threshold > fd0: <1440-KB 3.5" drive> on fdc0 drive 0 > atkbdc0: at port 0x60,0x64 on isa0 > atkbd0: irq 1 on atkbdc0 > psm0: irq 12 on atkbdc0 > psm0: model MouseMan+, device ID 0 > vga0: at port 0x3c0-0x3df iomem 0xa0000-0xbffff on isa0 > sc0: on isa0 > sc0: VGA <16 virtual consoles, flags=0x200> > sio0 at port 0x3f8-0x3ff irq 4 flags 0x10 on isa0 > sio0: type 16550A > ppc0: at port 0x378-0x37f irq 7 on isa0 > ppc0: Generic chipset (EPP/NIBBLE) in COMPATIBLE mode > ppbus0: IEEE1284 device found /NIBBLE/PS2 > Probing for PnP devices on ppbus0: > ppbus0: NPAP,PJL,PCL > lpt0: on ppbus0 > lpt0: Interrupt-driven port > ppc1: at port 0x278-0x27f irq 5 on isa0 > ppc1: Generic chipset (NIBBLE-only) in COMPATIBLE mode > lpt1: on ppbus1 > lpt1: Interrupt-driven port > IP Filter: initialized. Default = pass all, Logging = enabled > IP Filter: v3.3.8 > ad4: 19609MB [39842/16/63] at ata2-master > using UDMA66 > ad6: 19609MB [39842/16/63] at ata3-master > using UDMA66 > acd0: CDROM at ata0-master using UDMA33 > > ifconfig -a (dc0 ip address blanked out) > > dc0: flags=8843 mtu 1500 > inet 24.108.x.x netmask 0xffffff00 broadcast 24.108.89.255 > ether 00:a0:cc:32:eb:7c > media: autoselect (10baseT/UTP) status: active > supported media: autoselect 100baseTX 100baseTX > 10baseT/UTP 10baseT/UTP 100baseTX none > dc1: flags=8843 mtu 1500 > inet 192.168.1.1 netmask 0xffffff00 broadcast 192.168.1.255 > ether 00:a0:cc:33:7b:ab > media: autoselect (100baseTX ) status: active > supported media: autoselect 100baseTX 100baseTX > 10baseT/UTP 10baseT/UTP 100baseTX none > lo0: flags=8049 mtu 16384 > inet 127.0.0.1 netmask 0xff000000 > > netstat -bi (again, formatted, I apologize if it's hard to read, I've > tried to make it easier) > > Name Mtu Network Address Ipkts Ierrs Ibytes > dc0 1500 00:a0:cc:32:eb:7c 39288 0 4923890 > Opkts Oerrs Obytes Coll > 4767 0 672170 549 > Name Mtu Network Address Ipkts Ierrs Ibytes > dc0 1500 24.108.89/24 cmdmicro 39288 0 4923890 > Opkts Oerrs Obytes Coll > 4767 0 672170 549 > Name Mtu Network Address Ipkts Ierrs Ibytes > dc1 1500 00:a0:cc:33:7b:ab 180147 0 114224991 > Opkts Oerrs Obytes Coll > 178687 5 165873105 0 > Name Mtu Network Address Ipkts Ierrs Ibytes > dc1 1500 192.168.1 cmdmicro 180147 0 114224991 > Opkts Oerrs Obytes Coll > 178687 5 165873105 0 > Name Mtu Network Address Ipkts Ierrs Ibytes > lo0 16384 3217 0 191261 > Opkts Oerrs Obytes Coll > 3217 0 191261 0 > Name Mtu Network Address Ipkts Ierrs Ibytes > lo0 16384 127 cmdmicro 3217 0 191261 > Opkts Oerrs Obytes Coll > 3217 0 191261 0 > > kernel config: > > machine i386 > cpu I686_CPU > ident DEEPTHOUGHT > maxusers 128 > > options INET #InterNETworking > options FFS #Berkeley Fast Filesystem > options FFS_ROOT #FFS usable as root device [keep this!] > options CD9660 #ISO 9660 Filesystem > options PROCFS #Process filesystem > options COMPAT_43 #Compatible with BSD 4.3 [KEEP THIS!] > options UCONSOLE #Allow users to grab the console > options USERCONFIG #boot -c editor > options VISUAL_USERCONFIG #visual boot -c editor > options KTRACE #ktrace(1) support > options P1003_1B #Posix P1003_1B real-time extentions > options _KPOSIX_PRIORITY_SCHEDULING > options _KPOSIX_VERSION=199309L > options ICMP_BANDLIM #Rate limit bad replies > options IPFILTER #ipfilter support > options IPFILTER_LOG #ipfilter logging > options QUOTA > options SOFTUPDATES > #options NMBCLUSTERS=16384 > #options PQ_HUGECACHE > options PANIC_REBOOT_WAIT_TIME=60 > options MD5 > options VESA > options DDB > options DDB_UNATTENDED > options PMAP_SHPGPERPROC=400 > options MAXCONS=16 > options MAXDSIZ=(256*1024*1024) > options DFLDSIZ=(256*1024*1024) > options SYSVSHM #SYSV-style shared memory > options SYSVMSG #SYSV-style message queues > options SYSVSEM #SYSV-style semaphores > > device isa > device pci > > # Floppy drives > device fdc0 at isa? port IO_FD1 irq 6 drq 2 > device fd0 at fdc0 drive 0 > > # ATA and ATAPI devices > device ata0 at isa? port IO_WD1 irq 14 > device ata1 at isa? port IO_WD2 irq 15 > device ata > device atadisk # ATA disk drives > device atapicd # ATAPI CDROM drives > options ATA_STATIC_ID #Static device numbering > options ATA_ENABLE_ATAPI_DMA #Enable DMA on ATAPI devices > > # atkbdc0 controls both the keyboard and the PS/2 mouse > device atkbdc0 at isa? port IO_KBD > device atkbd0 at atkbdc? irq 1 > device psm0 at atkbdc? irq 12 > > device vga0 at isa? > > # splash screen/screen saver > pseudo-device splash > > # syscons is the default console driver, resembling an SCO console > device sc0 at isa? > > # Floating point support - do not disable. > device npx0 at nexus? port IO_NPX irq 13 > > # Power management support (see LINT for more options) > # device apm0 at nexus? # Advanced Power Management > > # Serial (COM) ports > device sio0 at isa? port IO_COM1 flags 0x10 irq 4 > #device sio1 at isa? port IO_COM2 irq 3 > > # Parallel port > device ppc0 at isa? irq 7 > device ppc1 at isa? irq 5 > device ppbus # Parallel port bus (required) > device lpt # Printer > > # PCI Ethernet NICs that use the common MII bus controller code. > device miibus # MII bus support > device dc # DEC/Intel 21143 and various workalikes > > # Pseudo devices - the number indicates how many units to allocated. > pseudo-device loop # Network loopback > pseudo-device ether # Ethernet support > pseudo-device pty # Pseudo-ttys (telnet etc) > > # The `bpf' pseudo-device enables the Berkeley Packet Filter. > # Be aware of the administrative consequences of enabling this! > pseudo-device bpf #Berkeley packet filter > > # pc speaker > pseudo-device speaker > > > To Unsubscribe: send mail to majordomo@FreeBSD.org > with "unsubscribe freebsd-stable" in the body of the message To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-stable" in the body of the message