Date: Mon, 17 Nov 1997 17:27:10 +0200 (EET) From: archer@lucky.net To: FreeBSD-gnats-submit@FreeBSD.ORG Subject: kern/5071: 2.2.5-STABLE crashes under heavy network loads Message-ID: <199711171527.RAA01984@burka.carrier.kiev.ua> Resent-Message-ID: <199711171530.HAA17798@hub.freebsd.org>
next in thread | raw e-mail | index | archive | help
>Number: 5071
>Category: kern
>Synopsis: 2.2.5-STABLE crashes under heavy network loads
>Confidential: no
>Severity: serious
>Priority: medium
>Responsible: freebsd-bugs
>State: open
>Class: sw-bug
>Submitter-Id: current-users
>Arrival-Date: Mon Nov 17 07:30:01 PST 1997
>Last-Modified:
>Originator: Alexander Litvin
>Organization:
Lucky Net Ltd., ISP, Ukraine
>Release: FreeBSD 2.2.5-STABLE
>Environment:
Output of dmesg:
-----------------------------------------------------------------------
Copyright (c) 1992-1997 FreeBSD Inc.
Copyright (c) 1982, 1986, 1989, 1991, 1993
The Regents of the University of California. All rights reserved.
FreeBSD 2.2.5-STABLE #0: Sat Nov 15 17:47:19 EET 1997
archer@zebra.carrier.kiev.ua:/usr/local/squid/src/sys/compile/zebra.s
CPU: Pentium Pro (267.27-MHz 686-class CPU)
Origin = "GenuineIntel" Id = 0x633 Stepping=3
Features=0x80f9ff<FPU,VME,DE,PSE,TSC,MSR,PAE,MCE,CX8,<b11>,MTRR,PGE,MCA,CMOV>
real memory = 134217728 (131072K bytes)
avail memory = 129630208 (126592K bytes)
Probing for devices on PCI bus 0:
chip0 <generic PCI bridge (vendor=8086 device=7180 subclass=0)> rev 3 on pci0:0
chip1 <generic PCI bridge (vendor=8086 device=7181 subclass=4)> rev 3 on pci0:1
chip2 <generic PCI bridge (vendor=8086 device=7110 subclass=1)> rev 1 on pci0:4:0
pci0:4:1: Intel Corporation, device=0x7111, class=storage (ide) [no driver assigned]
pci0:4:2: Intel Corporation, device=0x7112, class=0x0c, subclass=0x03 int d irq 11 [no driver assigned]
chip3 <generic PCI bridge (vendor=8086 device=7113 subclass=128)> rev 1 on pci0:4:3
fxp0 <Intel EtherExpress Pro 10/100B Ethernet> rev 1 int a irq 12 on pci0:10
fxp0: Ethernet address 00:a0:c9:1c:a3:3f
vga0 <VGA-compatible display device> rev 227 int a irq 14 on pci0:11
ahc0 <Adaptec 2940 Ultra SCSI host adapter> rev 0 int a irq 15 on pci0:12
ahc0: aic7880 Wide Channel, SCSI Id=7, 16 SCBs
(ahc0:0:0): "IBM DCAS-34330W S61A" type 0 fixed SCSI 2
sd0(ahc0:0:0): Direct-Access 4134MB (8467200 512 byte sectors)
sd0(ahc0:0:0): with 8205 cyls, 6 heads, and an average 171 sectors/track
(ahc0:1:0): "IBM DCAS-34330W S61A" type 0 fixed SCSI 2
sd1(ahc0:1:0): Direct-Access 4134MB (8467200 512 byte sectors)
sd1(ahc0:1:0): with 8205 cyls, 6 heads, and an average 171 sectors/track
(ahc0:2:0): "IBM DCAS-34330W S61A" type 0 fixed SCSI 2
sd2(ahc0:2:0): Direct-Access 4134MB (8467200 512 byte sectors)
sd2(ahc0:2:0): with 8205 cyls, 6 heads, and an average 171 sectors/track
Probing for devices on PCI bus 1:
Probing for devices on the ISA bus:
sc0 at 0x60-0x6f irq 1 on motherboard
sc0: VGA color <16 virtual consoles, flags=0x0>
ed0 not found at 0x280
ed1 not found at 0x300
ed2 not found at 0x320
sio0 not found at 0x3f8
fdc0: direction bit not set
fdc0: cmd 3 failed at out byte 1 of 3
fdc0 not found at 0x3f0
npx0 flags 0x7 on motherboard
npx0: INT 16 interface
ccd0-3: Concatenated disk drivers
-----------------------------------------------------------------------
Kernel config file:
-----------------------------------------------------------------------
machine "i386"
ident zerba
maxusers 256
options "MAXDSIZ=(256*1024*1024)"
options "DFLDSIZ=(256*1024*1024)"
options FAILSAFE
options INCLUDE_CONFIG_FILE # Include this file in kernel
config kernel root on sd0 dumps on sd0
cpu "I686_CPU" # aka Pentium Pro(tm)
#new math emulator
options "COMPAT_43"
options SYSVSHM
options SYSVSEM
options SYSVMSG
options "MD5"
options USERCONFIG #boot -c editor
options VISUAL_USERCONFIG #visual boot -c editor
options INET #Internet communications protocols
pseudo-device ether #Generic Ethernet
pseudo-device loop #Network loopback device
# dropped packets
options FFS #Fast filesystem
options PROCFS #Process filesystem
options NSWAPDEV=4
options QUOTA #enable disk quotas
controller scbus0 #base SCSI code
device sd0 #SCSI disks
options SCSI_REPORT_GEOMETRY
pseudo-device pty 32 #Pseudo ttys - can go as high as 256
pseudo-device speaker #Play IBM BASIC-style noises out your speaker
pseudo-device log #Kernel syslog interface (/dev/klog)
pseudo-device gzip #Exec gzipped a.out's
pseudo-device vn #Vnode driver (turns a file into a device)
pseudo-device snp 3 #Snoop device - to look at pty/vty/etc..
pseudo-device ccd 4 #Concatenated disk driver
controller isa0
#options "AUTO_EOI_1"
options "MAXMEM=(128*1024)"
#options BROKEN_KEYBOARD_RESET
device sc0 at isa? port "IO_KBD" tty irq 1 vector scintr
options MAXCONS=16 # number of virtual consoles
device npx0 at isa? port "IO_NPX" iosiz 0x0 flags 0x7 irq 13 vector npxintr
controller fdc0 at isa? port "IO_FD1" bio irq 6 drq 2 vector fdintr
disk fd0 at fdc0 drive 0
device sio0 at isa? port "IO_COM1" tty irq 4 vector siointr
#DDB, if available.
device ed0 at isa? port 0x280 net irq 5 iomem 0xd8000 vector edintr
device ed1 at isa? port 0x300 net irq 5 iomem 0xd8000 vector edintr
device ed2 at isa? port 0x320 net irq 5 iomem 0xd8000 vector edintr
controller pci0
controller ahc0
device fxp0
device vx0
---------------------------------------------------------------------------
>Description:
Busy proxy server crashes constantly at random periouds, with
the same simpthoms. Output of "(kgdb)bt":
--------------------
IdlePTD 1c3000
current pcb at 1ac9d0
panic: page fault
#0 boot (howto=260) at ../../kern/kern_shutdown.c:266
266 dumppcb.pcb_cr3 = rcr3();
(kgdb) bt
#0 boot (howto=260) at ../../kern/kern_shutdown.c:266
#1 0xf01116e2 in panic (fmt=0xf017b84f "page fault")
at ../../kern/kern_shutdown.c:390
#2 0xf017c3b6 in trap_fatal (frame=0xf01a0dc4) at ../../i386/i386/trap.c:742
#3 0xf017bea4 in trap_pfault (frame=0xf01a0dc4, usermode=0)
at ../../i386/i386/trap.c:653
#4 0xf017bb7f in trap (frame={tf_es = -266731504, tf_ds = -267255792,
tf_edi = 0, tf_esi = -161442560, tf_ebp = -266727920,
tf_isp = -266727956, tf_ebx = -158123136, tf_edx = 0,
tf_ecx = -162092544, tf_eax = -158123264, tf_trapno = 12, tf_err = 0,
tf_eip = -267186362, tf_cs = 8, tf_eflags = 66118, tf_esp = 0,
tf_ss = -158123136}) at ../../i386/i386/trap.c:311
#5 0xf0130f46 in vget (vp=0xf6933b80, lockflag=1) at ../../kern/vfs_subr.c:817
#6 0xf015cd3c in ffs_sync (mp=0xf6573600, waitfor=2, cred=0xf3c17880,
p=0xf01b7c94) at ../../ufs/ffs/ffs_vfsops.c:819
#7 0xf0132407 in sync (p=0xf01b7c94, uap=0x0, retval=0x0)
at ../../kern/vfs_syscalls.c:360
#8 0xf01112ed in boot (howto=256) at ../../kern/kern_shutdown.c:199
#9 0xf01116e2 in panic (fmt=0xf017b84f "page fault")
at ../../kern/kern_shutdown.c:390
#10 0xf017c3b6 in trap_fatal (frame=0xf01a0f48) at ../../i386/i386/trap.c:742
#11 0xf017bea4 in trap_pfault (frame=0xf01a0f48, usermode=0)
at ../../i386/i386/trap.c:653
#12 0xf017bb7f in trap (frame={tf_es = -154664944, tf_ds = -267059184,
tf_edi = -1073741824, tf_esi = -1073741824, tf_ebp = -266727540,
tf_isp = -266727568, tf_ebx = -265166040, tf_edx = 172511825,
tf_ecx = -159965440, tf_eax = 87502848, tf_trapno = 12, tf_err = 0,
tf_eip = -267083713, tf_cs = 8, tf_eflags = 66054, tf_esp = -266715032,
tf_ss = -266714668}) at ../../i386/i386/trap.c:311
#13 0xf014a03f in tcp_fasttimo () at ../../netinet/tcp_timer.c:111
#14 0xf0124677 in pffasttimo (arg=0x0) at ../../kern/uipc_domain.c:234
#15 0xf01090fc in softclock () at ../../kern/kern_clock.c:715
----------
Last three lines are always the same.
The only source of network load is squid-1.NOVM.17 and named (caching only).
Squid gets more then 50'000 requests per hour at peeks.
>How-To-Repeat:
I was able to force it to crash by loading proxy with a requests
from "webjamma" started at 7 machines at the local network, each of them
making up to 200 concurrent requests.
>Fix:
>Audit-Trail:
>Unformatted:
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199711171527.RAA01984>