lkml.org 
[lkml]   [1999]   [Sep]   [13]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
    /
    From
    SubjectIDE + SMP Lockup (no OOPS) in 2.2.12, 2.2.10
    Date
    Hello,

    I'm running into a problem running an SMP box with heavy ide disk access.
    If the I/O subsystem is stressed, you can lock the machine up in seconds.
    If it's not stressed it can stay up for hours, but will always crash.

    My system: Abit BP-6 dual 366 celerons, 128MB PC100 RAM, redhat 6.0, 12 ide
    disks attached (to primary PIIX4 and three PCI PDC20246's... note, I am NOT
    using the onboard HPT-366), 3com 3c905b 100mbit ethernet

    I have two other very similar systems, both Abit BP-6 dual celeron 366,
    128MB RAM, 3c905b's, redhat 6.0, 2.12.12 but using Tekram 390F controllers
    and UW SCSI disks + raid, and they are both 100% stable.

    But for the ide system:

    Using 2.2.12 + raid0145-19990824-2.2.11 + 2.2.12.uniform-ide-6.20.hydra,
    with SMP enabled:
    #!/bin/sh
    for i in 1 2 3 4 5
    do
    for i in hdb1 hdc1 hde1 hdg1 hdi1
    do
    dd if=/dev/$i of=/dev/null count=500000 &
    done
    dd if=/dev/hdk1 of=/dev/null count=500000
    done
    [root@music /root]# ./crashme.sh

    [machine locks up quickly, responds to nothing, requires hard reset]

    Using a vanilla 2.2.12 kernel, with SMP enabled (note only first two PDC
    controllers are active in this configuration) - the same crashme.sh script
    locks up the computer just as quickly, no messages.

    Using 2.2.12 + raid0145-19990824-2.2.11 + 2.2.12.uniform-ide-6.20.hydra with
    SMP disabled -or- 2.2.12 with no patches and SMP disabled results in no
    crashes, the system will not crash.

    Using 2.2.10 with or without raid & ide patches results in the same lockup
    if SMP is enabled, but worked for weeks of heavy use when SMP was disabled.

    Using 2.2.6 + raid0145-19990421-2.2.6 + 2.2.7.uniform-ide-6.19 with SMP
    enabled passes the test over and over with flying colors, and no lockups in
    2+ months of service (though it does not work with the last promise
    controller either)

    Thanks for your time!

    Attached is a boot log from the 2.2.12 vanilla kernel:

    Sep 13 09:05:46 music syslogd 1.3-3: restart.
    Sep 13 09:05:46 music syslog: syslogd startup succeeded
    Sep 13 09:05:46 music kernel: klogd 1.3-3, log source = /proc/kmsg started.
    Sep 13 09:05:46 music kernel: Inspecting /boot/System.map
    Sep 13 09:05:46 music syslog: klogd startup succeeded
    Sep 13 09:05:46 music kernel: Loaded 5217 symbols from /boot/System.map.
    Sep 13 09:05:46 music kernel: Symbols match kernel version 2.2.12.
    Sep 13 09:05:46 music kernel: No module symbols loaded.
    Sep 13 09:05:46 music kernel: Linux version 2.2.12 (root@music.volition.org)
    (gcc version
    egcs-2.91.66 19990314/Linux (egcs-1.1.2 release)) #1 SMP Mon Sep 13
    08:51:46 PDT 1999
    Sep 13 09:05:46 music kernel: Intel MultiProcessor Specification v1.1
    Sep 13 09:05:46 music kernel: Virtual Wire compatibility mode.
    Sep 13 09:05:46 music kernel: OEM ID: OEM00000 Product ID: PROD00000000 APIC
    at: 0xFEE000
    00
    Sep 13 09:05:46 music kernel: Processor #0 Pentium(tm) Pro APIC version 17
    Sep 13 09:05:46 music kernel: Processor #1 Pentium(tm) Pro APIC version 17
    Sep 13 09:05:46 music kernel: I/O APIC #2 Version 17 at 0xFEC00000.
    Sep 13 09:05:46 music kernel: Processors: 2
    Sep 13 09:05:46 music kernel: mapped APIC to ffffe000 (fee00000)
    Sep 13 09:05:46 music kernel: mapped IOAPIC to ffffd000 (fec00000)
    Sep 13 09:05:46 music kernel: Detected 367518013 Hz processor.
    Sep 13 09:05:46 music kernel: Console: colour VGA+ 80x25
    Sep 13 09:05:46 music kernel: Calibrating delay loop... 367.00 BogoMIPS
    Sep 13 09:05:46 music kernel: Memory: 128060k/131072k available (972k kernel
    code, 420k r
    eserved, 1556k data, 64k init)
    Sep 13 09:05:46 music kernel: Checking 386/387 coupling... OK, FPU using
    exception 16 err
    or reporting.
    Sep 13 09:05:46 music kernel: Checking 'hlt' instruction... OK.
    Sep 13 09:05:46 music kernel: POSIX conformance testing by UNIFIX
    Sep 13 09:05:46 music kernel: per-CPU timeslice cutoff: 25.03 usecs.
    Sep 13 09:05:46 music kernel: CPU0: Intel Celeron (Mendocino) stepping 05
    Sep 13 09:05:46 music kernel: calibrating APIC timer ...
    Sep 13 09:05:46 music kernel: ..... CPU clock speed is 367.5025 MHz.
    Sep 13 09:05:46 music kernel: ..... system bus clock speed is 66.8185 MHz.
    Sep 13 09:05:46 music kernel: Booting processor 1 eip 2000
    Sep 13 09:05:46 music kernel: Calibrating delay loop... 367.00 BogoMIPS
    Sep 13 09:05:46 music kernel: OK.
    Sep 13 09:05:46 music kernel: CPU1: Intel Celeron (Mendocino) stepping 05
    Sep 13 09:05:46 music kernel: Total of 2 processors activated (734.00
    BogoMIPS).
    Sep 13 09:05:46 music kernel: enabling symmetric IO mode... ...done.
    Sep 13 09:05:46 music kernel: ENABLING IO-APIC IRQs
    Sep 13 09:05:46 music kernel: init IO_APIC IRQs
    Sep 13 09:05:46 music kernel: IO-APIC pin 0, 5, 7, 10, 11, 20, 21, 22, 23
    not connected.

    Sep 13 09:05:46 music kernel: number of MP IRQ sources: 16.
    Sep 13 09:05:46 music kernel: number of IO-APIC registers: 24.
    Sep 13 09:05:46 music kernel: testing the IO APIC.......................
    Sep 13 09:05:46 music kernel: .... register #00: 02000000
    Sep 13 09:05:46 music kernel: ....... : physical APIC id: 02
    Sep 13 09:05:46 music kernel: .... register #01: 00170011
    Sep 13 09:05:46 music kernel: ....... : max redirection entries: 0017
    Sep 13 09:05:46 music kernel: ....... : IO APIC version: 0011
    Sep 13 09:05:46 music kernel: .... register #02: 00000000
    Sep 13 09:05:46 music kernel: ....... : arbitration: 00
    Sep 13 09:05:46 music kernel: .... IRQ redirection table:
    Sep 13 09:05:47 music atd: atd startup succeeded
    Sep 13 09:05:46 music kernel: NR Log Phy Mask Trig IRR Pol Stat Dest Deli
    Vect:
    Sep 13 09:05:46 music kernel: 00 000 00 1 0 0 0 0 0 0
    00
    Sep 13 09:05:46 music kernel: 01 000 00 0 0 0 0 0 1 1
    59
    Sep 13 09:05:46 music kernel: 02 0FF 0F 0 0 0 0 0 1 1
    51
    Sep 13 09:05:46 music kernel: 03 000 00 0 0 0 0 0 1 1
    61
    Sep 13 09:05:46 music kernel: 04 000 00 0 0 0 0 0 1 1
    69
    Sep 13 09:05:46 music kernel: 05 000 00 1 0 0 0 0 0 0
    00
    Sep 13 09:05:46 music kernel: 06 000 00 0 0 0 0 0 1 1
    71
    Sep 13 09:05:46 music kernel: 07 000 00 1 0 0 0 0 0 0
    00
    Sep 13 09:05:46 music kernel: 08 000 00 0 0 0 0 0 1 1
    79
    Sep 13 09:05:46 music kernel: 09 000 00 0 0 0 0 0 1 1
    81
    Sep 13 09:05:46 music kernel: 0a 000 00 1 0 0 0 0 0 0
    00
    Sep 13 09:05:46 music kernel: 0b 000 00 1 0 0 0 0 0 0
    00
    Sep 13 09:05:47 music crond: crond startup succeeded
    Sep 13 09:05:46 music kernel: 0c 000 00 0 0 0 0 0 1 1
    89
    Sep 13 09:05:46 music kernel: 0d 000 00 1 0 0 0 0 0 0
    00
    Sep 13 09:05:46 music kernel: 0e 000 00 0 0 0 0 0 1 1
    91
    Sep 13 09:05:46 music kernel: 0f 000 00 0 0 0 0 0 1 1
    99
    Sep 13 09:05:46 music kernel: 10 0FF 0F 1 1 0 1 0 1 1
    A1
    Sep 13 09:05:46 music kernel: 11 0FF 0F 1 1 0 1 0 1 1
    A9
    Sep 13 09:05:46 music kernel: 12 0FF 0F 1 1 0 1 0 1 1
    B1
    Sep 13 09:05:46 music kernel: 13 0FF 0F 1 1 0 1 0 1 1
    B9
    Sep 13 09:05:46 music kernel: 14 000 00 1 0 0 0 0 0 0
    00
    Sep 13 09:05:46 music kernel: 15 000 00 1 0 0 0 0 0 0
    00
    Sep 13 09:05:46 music kernel: 16 000 00 1 0 0 0 0 0 0
    00
    Sep 13 09:05:46 music kernel: 17 000 00 1 0 0 0 0 0 0
    00
    Sep 13 02:05:40 music rc.sysinit: Loading default keymap succeeded
    Sep 13 02:05:40 music rc.sysinit: Setting default font succeeded
    Sep 13 02:05:40 music swapon: swapon: warning: /dev/hda5 has insecure
    permissions 0660, 0
    600 suggested
    Sep 13 02:05:40 music swapon: swapon: warning: /dev/hda6 has insecure
    permissions 0660, 0
    600 suggested
    Sep 13 02:05:40 music rc.sysinit: Activating swap partitions succeeded
    Sep 13 02:05:40 music rc.sysinit: Setting hostname music.volition.org
    succeeded
    Sep 13 02:05:40 music fsck: /dev/hda7: clean, 88550/514000 files,
    1864522/2048001 blocks
    Sep 13 02:05:40 music rc.sysinit: Checking root filesystem succeeded
    Sep 13 02:05:40 music rc.sysinit: Remounting root filesystem in read-write
    mode succeeded

    Sep 13 02:05:40 music depmod: can't open /lib/modules/2.2.12/modules.dep
    Sep 13 02:05:40 music rc.sysinit: Finding module dependencies failed
    Sep 13 02:05:40 music fsck: /dev/hda1: clean, 62/52208 files, 28756/208813
    blocks
    Sep 13 09:05:46 music kernel: .................................... done.
    Sep 13 09:05:46 music kernel: PCI: PCI BIOS revision 2.10 entry at 0xfb420
    Sep 13 09:05:46 music kernel: PCI: Using configuration type 1
    Sep 13 09:05:46 music kernel: PCI: Probing PCI hardware
    Sep 13 09:05:46 music kernel: Linux NET4.0 for Linux 2.2
    Sep 13 09:05:46 music kernel: Based upon Swansea University Computer Society
    NET3.039
    Sep 13 09:05:46 music kernel: NET4: Unix domain sockets 1.0 for Linux
    NET4.0.
    Sep 13 09:05:46 music kernel: NET4: Linux TCP/IP 1.0 for NET4.0
    Sep 13 02:05:40 music rc.sysinit: Checking filesystems succeeded
    Sep 13 02:05:41 music rc.sysinit: Mounting local filesystems succeeded
    Sep 13 02:05:41 music rc.sysinit: Turning on user and group quotas for local
    filesystems
    succeeded
    Sep 13 09:05:43 music date: Mon Sep 13 09:05:43 PDT 1999
    Sep 13 09:05:43 music rc.sysinit: Setting clock : Mon Sep 13 09:05:43 PDT
    1999 succeeded
    Sep 13 09:05:43 music rc.sysinit: Enabling swap space succeeded
    Sep 13 09:05:43 music init: Entering runlevel: 3
    Sep 13 09:05:45 music network: Bringing up interface lo succeeded
    Sep 13 09:05:45 music network: Bringing up interface eth0 succeeded
    Sep 13 09:05:46 music portmap: portmap startup succeeded
    Sep 13 09:05:46 music random: Initializing random number generator succeeded
    Sep 13 09:05:47 music inet: inetd startup succeeded
    Sep 13 09:05:46 music kernel: IP Protocols: ICMP, UDP, TCP
    Sep 13 09:05:46 music kernel: Starting kswapd v 1.5
    Sep 13 09:05:46 music kernel: Detected PS/2 Mouse Port.
    Sep 13 09:05:46 music kernel: Serial driver version 4.27 with no serial
    options enabled
    Sep 13 09:05:46 music kernel: pty: 256 Unix98 ptys configured
    Sep 13 09:05:46 music kernel: PIIX4: IDE controller on PCI bus 00 dev 39
    Sep 13 09:05:46 music kernel: PIIX4: not 100% native mode: will probe irqs
    later
    Sep 13 09:05:46 music kernel: ide0: BM-DMA at 0xf000-0xf007, BIOS
    settings: hda:DMA,
    hdb:DMA
    Sep 13 09:05:46 music kernel: ide1: BM-DMA at 0xf008-0xf00f, BIOS
    settings: hdc:DMA,
    hdd:DMA
    Sep 13 09:05:46 music kernel: PDC20246: IDE controller on PCI bus 00 dev 58
    Sep 13 09:05:46 music kernel: PDC20246: not 100% native mode: will probe
    irqs later
    Sep 13 09:05:46 music kernel: PDC20246: ROM enabled at 0xe6000000
    Sep 13 09:05:46 music kernel: ide2: BM-DMA at 0xa400-0xa407, BIOS
    settings: hde:DMA,
    hdf:DMA
    Sep 13 09:05:48 music named[324]: starting. named 8.2 Wed Mar 31 10:57:12
    EST 1999 ^Iroo
    t@porky.devel.redhat.com:/usr/src/bs/BUILD/bind-8.2/src/bin/named
    Sep 13 09:05:48 music named[324]: cache zone "" (IN) loaded (serial 0)
    Sep 13 09:05:48 music named[324]: Zone "0.0.127.in-addr.arpa" (file
    named.local): No defa
    ult TTL set using SOA minimum instead
    Sep 13 09:05:48 music named[324]: master zone "0.0.127.in-addr.arpa" (IN)
    loaded (serial
    1997022700)
    Sep 13 09:05:48 music named[324]: listening on [127.0.0.1].53 (lo)
    Sep 13 09:05:48 music named[324]: listening on [10.10.10.199].53 (eth0)
    Sep 13 09:05:48 music named[324]: Forwarding source address is
    [0.0.0.0].1024
    Sep 13 09:05:46 music kernel: ide3: BM-DMA at 0xa408-0xa40f, BIOS
    settings: hdg:DMA,
    hdh:DMA
    Sep 13 09:05:46 music kernel: PDC20246: IDE controller on PCI bus 00 dev 68
    Sep 13 09:05:46 music kernel: PDC20246: not 100% native mode: will probe
    irqs later
    Sep 13 09:05:46 music kernel: PDC20246: ROM enabled at 0xe7000000
    Sep 13 09:05:46 music kernel: ide4: BM-DMA at 0xb800-0xb807, BIOS
    settings: hdi:DMA,
    hdj:pio
    Sep 13 09:05:46 music kernel: ide5: BM-DMA at 0xb808-0xb80f, BIOS
    settings: hdk:DMA,
    hdl:pio
    Sep 13 09:05:46 music kernel: PDC20246: IDE controller on PCI bus 00 dev 78
    Sep 13 09:05:46 music kernel: PDC20246: not 100% native mode: will probe
    irqs later
    Sep 13 09:05:46 music kernel: PDC20246: ROM enabled at 0xe8000000
    Sep 13 09:05:46 music kernel: PDC20246: too many IDE interfaces, no room in
    table
    Sep 13 09:05:46 music kernel: PDC20246: too many IDE interfaces, no room in
    table
    Sep 13 09:05:46 music kernel: hda: QUANTUM FIREBALL ST4.3A, ATA DISK drive
    Sep 13 09:05:46 music kernel: hdb: Maxtor 90845D4, ATA DISK drive
    Sep 13 09:05:48 music named: named startup succeeded
    Sep 13 09:05:48 music named[325]: Ready to answer queries.
    Sep 13 09:05:46 music kernel: hdc: Maxtor 90845D4, ATA DISK drive
    Sep 13 09:05:46 music kernel: hdd: Maxtor 90845D4, ATA DISK drive
    Sep 13 09:05:46 music kernel: hde: Maxtor 90845D4, ATA DISK drive
    Sep 13 09:05:46 music kernel: hdf: Maxtor 90840D6, ATA DISK drive
    Sep 13 09:05:46 music kernel: hdg: Maxtor 90840D5, ATA DISK drive
    Sep 13 09:05:46 music kernel: hdi: Maxtor 90840D6, ATA DISK drive
    Sep 13 09:05:46 music kernel: hdk: Maxtor 90845D4, ATA DISK drive
    Sep 13 09:05:46 music kernel: ide0 at 0x1f0-0x1f7,0x3f6 on irq 14
    Sep 13 09:05:46 music kernel: ide1 at 0x170-0x177,0x376 on irq 15
    Sep 13 09:05:46 music kernel: ide2 at 0x9400-0x9407,0x9806 on irq 11
    Sep 13 09:05:46 music kernel: ide3 at 0x9c00-0x9c07,0xa006 on irq 11
    Sep 13 09:05:46 music kernel: ide4 at 0xa800-0xa807,0xac06 on irq 7
    Sep 13 09:05:46 music kernel: ide5 at 0xb000-0xb007,0xb406 on irq 7
    Sep 13 09:05:46 music kernel: hda: QUANTUM FIREBALL ST4.3A, 4110MB w/81kB
    Cache, CHS=524/
    255/63, UDMA
    Sep 13 09:05:48 music keytable: Loading keymap:
    Sep 13 09:05:46 music kernel: hdb: Maxtor 90845D4, 8063MB w/256kB Cache,
    CHS=16383/16/63,
    UDMA
    Sep 13 09:05:46 music kernel: hdc: Maxtor 90845D4, 8063MB w/256kB Cache,
    CHS=16383/16/63,
    UDMA
    Sep 13 09:05:46 music kernel: hdd: Maxtor 90845D4, 8063MB w/256kB Cache,
    CHS=16383/16/63,
    UDMA
    Sep 13 09:05:46 music kernel: hde: Maxtor 90845D4, 8063MB w/256kB Cache,
    CHS=16383/16/63,
    UDMA
    Sep 13 09:05:46 music kernel: hdf: Maxtor 90840D6, 8010MB w/256kB Cache,
    CHS=16276/16/63,
    UDMA
    Sep 13 09:05:46 music kernel: hdg: Maxtor 90840D5, 8047MB w/256kB Cache,
    CHS=16351/16/63,
    UDMA
    Sep 13 09:05:46 music kernel: hdi: Maxtor 90840D6, 8010MB w/256kB Cache,
    CHS=16276/16/63,
    UDMA
    Sep 13 09:05:46 music kernel: hdk: Maxtor 90845D4, 8063MB w/256kB Cache,
    CHS=16383/16/63,
    UDMA
    Sep 13 09:05:46 music kernel: Floppy drive(s): fd0 is 1.44M
    Sep 13 09:05:46 music kernel: FDC 0 is a post-1991 82077
    Sep 13 09:05:46 music kernel: scsi : 0 hosts.
    Sep 13 09:05:46 music kernel: scsi : detected total.
    Sep 13 09:05:46 music kernel: 3c59x.c:v0.99H 11/17/98 Donald Becker
    http://cesdis.gsfc.na
    sa.gov/linux/drivers/vortex.html
    Sep 13 09:05:49 music keytable: Loading
    /usr/lib/kbd/keymaps/i386/qwerty/us.kmap.gz
    Sep 13 09:05:46 music kernel: eth0: 3Com 3c905 Boomerang 100baseTx at
    0xd000, 00:60:08:b
    0:2e:3c, IRQ 10
    Sep 13 09:05:46 music kernel: 8K word-wide RAM 3:5 Rx:Tx split,
    autoselect/MII interfac
    e.
    Sep 13 09:05:46 music kernel: MII transceiver found at address 24, status
    786f.
    Sep 13 09:05:46 music kernel: Enabling bus-master transmits and
    whole-frame receives.
    Sep 13 09:05:46 music kernel: Partition check:
    Sep 13 09:05:46 music kernel: hda: hda1 hda2 < hda5 hda6 hda7 >
    Sep 13 09:05:46 music kernel: hdb: hdb1
    Sep 13 09:05:46 music kernel: hdc: hdc1
    Sep 13 09:05:46 music kernel: hdd: hdd1
    Sep 13 09:05:46 music kernel: hde: hde1
    Sep 13 09:05:46 music kernel: hdf: hdf1
    Sep 13 09:05:46 music kernel: hdg: hdg1
    Sep 13 09:05:46 music kernel: hdi: hdi1
    Sep 13 09:05:46 music kernel: hdk: hdk1








    -
    To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
    the body of a message to majordomo@vger.rutgers.edu
    Please read the FAQ at http://www.tux.org/lkml/

    \
     
     \ /
      Last update: 2005-03-22 13:53    [W:0.045 / U:0.276 seconds]
    ©2003-2016 Jasper Spaans. hosted at Digital OceanAdvertise on this site