Return to BSD News archive
Path: euryale.cc.adfa.oz.au!newshost.anu.edu.au!harbinger.cc.monash.edu.au!news.mira.net.au!yarrina.connect.com.au!news.mel.connect.com.au!munnari.OZ.AU!spool.mu.edu!howland.reston.ans.net!newsfeed.internetmci.com!in1.uu.net!news.mindspring.com!not-for-mail
From: esa@kaamos.mindspring.com (Esa Ahola)
Newsgroups: comp.unix.bsd.freebsd.misc
Subject: 2.1 news server crashes
Date: 20 Jan 1996 12:55:14 -0500
Organization: MindSpring Enterprises, Inc.
Lines: 92
Message-ID: <4draa2$ppl@kaamos.mindspring.com>
References: <DLCLL6.K7J@snews.zippo.com>
NNTP-Posting-Host: kaamos.mindspring.com
In article <DLCLL6.K7J@snews.zippo.com>, <kachun@pathlink.com> wrote:
> We are running a P100/128M/FreeBSD 2.1/News Server, which has a
> very heavy load - 300+ connections constantly. About once a week,
> it panics with kmem_map full.
We are seeing the same thing, also on a dedicated news server
serving 100-200 newsreaders at a time. The crashes don't seem to
correlate with periods of highest load.
Our configuration is:
ASUS P100 P55TP4XE (hope I got the alphabet soup right)
128MB RAM
Two ASUS/ncr53c810 PCI SCSI controllers
4 Barracuda disks (2x2G, 2x4G) in well-cooled external cabinets
ZNYX ZX34X DC21140 Ethernet
The crash dump shows:
#0 boot (howto=256) at ../../i386/i386/machdep.c:892
#1 0xf0113493 in panic (fmt=0xf018eb85 "kmem_malloc: kmem_map too small")
at ../../kern/subr_prf.c:124
#2 0xf018ec78 in kmem_malloc (map=0xf026f06c, size=4096, waitflag=0)
at ../../vm/vm_kern.c:313
#3 0xf010c21b in malloc (size=256, type=41, flags=0)
at ../../kern/kern_malloc.c:138
#4 0xf0109c32 in fork1 (p1=0xf222f900, isvfork=0, retval=0xefbfff8c)
at ../../kern/kern_fork.c:117
#5 0xf0109b80 in fork (p=0xf222f900, uap=0xefbfff94, retval=0xefbfff8c)
at ../../kern/kern_fork.c:66
#6 0xf01a35b3 in syscall (frame={tf_es = 39, tf_ds = 39, tf_edi = 18,
tf_esi = 18, tf_ebp = -272639332, tf_isp = -272629788, tf_ebx = 2,
tf_edx = 33, tf_ecx = 10, tf_eax = 2, tf_trapno = 518, tf_err = 518,
tf_eip = 134568821, tf_cs = 31, tf_eflags = 518, tf_esp = -272639356,
tf_ss = 39}) at ../../i386/i386/trap.c:878
#7 0xf0198e0b in Xsyscall ()
#8 0xfa15 in ?? ()
#9 0xfbe1 in ?? ()
#10 0x8bbb in ?? ()
#11 0xb498 in ?? ()
#12 0x10d3 in ?? ()
The kernel configuration differs from the 2.1 stock generic kernel as follows:
11,12c11,12
< ident GENERIC
< maxusers 10
---
> ident BRICKBAT
> maxusers 254
22c22
< options "SCSI_DELAY=15" #Be pessimistic about Joe SCSI device
---
> options "SCSI_DELAY=5" #Be pessimistic about Joe SCSI device
24a25,30
> options COMCONSOLE #prefer serial console to video console
> options KTRACE #kernel tracing
> options "CHILD_MAX=256"
> options "OPEN_MAX=512"
> options "MAXMEM=131072"
> options "NMBCLUSTERS=2048"
30c36
< config kernel root on wd0
---
> config kernel root on sd0
40,46c46,52
< controller wdc0 at isa? port "IO_WD1" bio irq 14 vector wdintr
< disk wd0 at wdc0 drive 0
< disk wd1 at wdc0 drive 1
<
< controller wdc1 at isa? port "IO_WD2" bio irq 15 vector wdintr
< disk wd2 at wdc1 drive 0
< disk wd3 at wdc1 drive 1
---
> # controller wdc0 at isa? port "IO_WD1" bio irq 14 vector wdintr
> # disk wd0 at wdc0 drive 0
> # disk wd1 at wdc0 drive 1
>
> # controller wdc1 at isa? port "IO_WD2" bio irq 15 vector wdintr
> # disk wd2 at wdc1 drive 0
> # disk wd3 at wdc1 drive 1
122c128
< pseudo-device pty 16
---
> pseudo-device pty 64
123a130
> pseudo-device bpfilter 4
--
Esa Ahola esa@mindspring.com
MindSpring Enterprises, Inc. 404.815.0770