Bug 16090

Summary:	No buffer space available
Product:	Base System	Reporter:	Charlie Root <root>
Component:	kern	Assignee:	Matthew N. Dodd <mdodd>
Status:	Closed FIXED
Severity:	Affects Only Me
Priority:	Normal
Version:	3.4-STABLE
Hardware:	Any
OS:	Any

Description Charlie Root 2000-01-12 23:40:01 UTC

an 12 16:18:36 genesis routed[14626]: Send bcast sendto(ep1, 10.0.0.255.520): No buffe
r space available
Jan 12 16:20:37 genesis last message repeated 4 times
Jan 12 16:25:37 genesis last message repeated 10 times
Jan 12 16:25:48 genesis routed[14626]: sendto(ep1, 224.0.0.1): No buffer space availabl
e
Jan 12 16:26:07 genesis routed[14626]: Send bcast sendto(ep1, 10.0.0.255.520): No buffe
r space available
Jan 12 16:26:37 genesis routed[14626]: Send bcast sendto(ep1, 10.0.0.255.520): No buffe
r space available
Jan 12 16:28:37 genesis last message repeated 4 times
Jan 12 16:33:37 genesis last message repeated 10 times
Jan 12 16:34:03 genesis shutdown: shutdown by root:
Jan 12 16:34:06 genesis syslogd: exiting on signal 15

It stops responding and routing at that time. The intranet is all NT Server (4 machines). One hosts about 60 websites, and is responsible for 75% of the network load.

This was probably existing before, though the load, and the fact that the machine is "make worlded", new kern, and rebooted every 30 days, so it just wasn't long enough to present the problem.
Since the machine has presented no problems for the last year, I seldom look hard. Will do more next time, and give further details then..

Fix:

reboot was used, though there are better. I will look closer next time. This time around, had about 5 people phoning, and another 7 ICQing so extreme measures were taken in a hurry.

Shutdown to single user has no effect. Will look deaper next time.
How-To-Repeat:
recurs about 1.5 days after a reboot. This started recently, when a large number of web sites were moved to the lan. ep1 is the intranet NIC.

This is similiar to PR #15164, #12394, and maybe #9357 (all rl driver).

Comment 1 Vallo Kallaste 2000-01-13 05:57:32 UTC

On Wed, Jan 12, 2000 at 06:38:26PM -0500, Charlie Root <root@genesis.setjmp.net> wrote:

> >Number:         16090
> >Category:       kern
> >Synopsis:       No buffer space available
> >Confidential:   no
> >Severity:       serious
> >Priority:       medium
> >Responsible:    freebsd-bugs
> >State:          open
> >Quarter:        
> >Keywords:       
> >Date-Required:
> >Class:          sw-bug
> >Submitter-Id:   current-users
> >Arrival-Date:   Wed Jan 12 15:40:01 PST 2000
> >Closed-Date:
> >Last-Modified:
> >Originator:     Charlie &
> >Release:        FreeBSD 3.4-STABLE i386
> >Organization:
> setjmp Software
> 		181 Genesee Street
> 		Suite 504
> 		Utica, NY 13501
> >Environment:
> 
> 	FreeBSD genesis.setjmp.net 3.4-STABLE FreeBSD 3.4-STABLE #0: Thu Jan  6 11:21:01 EST 2000     root@genesis.setjmp.net:/usr/src/sys/compile/GENESIS  i386
> 
> 	This box is a gateway between internet (10/Base) and lan (same). It uses 2 3com 3c509 isa cards for this. It is routing a number of websites. And getting considerable load now (running on a Pentium 60Mhz, with 104M ram). natd is used as well, static mapping for 4 ips.
> 
> 	The cards:
> 
> Jan 12 17:01:10 genesis /kernel: ep0 at 0x300-0x30f irq 10 on isa
> Jan 12 17:01:10 genesis /kernel: ep0: utp[*UTP*] address 00:60:97:93:05:1c
> Jan 12 17:01:10 genesis /kernel: ep1 at 0x200-0x20f irq 11 on isa
> Jan 12 17:01:10 genesis /kernel: ep1: aui/utp[*UTP*] address 00:20:af:1b:9f:fe
> 
> >Description:
> 
> an 12 16:18:36 genesis routed[14626]: Send bcast sendto(ep1, 10.0.0.255.520): No buffe
> r space available
> Jan 12 16:20:37 genesis last message repeated 4 times
> Jan 12 16:25:37 genesis last message repeated 10 times
> Jan 12 16:25:48 genesis routed[14626]: sendto(ep1, 224.0.0.1): No buffer space availabl
> e
> Jan 12 16:26:07 genesis routed[14626]: Send bcast sendto(ep1, 10.0.0.255.520): No buffe
> r space available
> Jan 12 16:26:37 genesis routed[14626]: Send bcast sendto(ep1, 10.0.0.255.520): No buffe
> r space available
> Jan 12 16:28:37 genesis last message repeated 4 times
> Jan 12 16:33:37 genesis last message repeated 10 times
> Jan 12 16:34:03 genesis shutdown: shutdown by root:
> Jan 12 16:34:06 genesis syslogd: exiting on signal 15
> 
> 	It stops responding and routing at that time. The intranet is all NT Server (4 machines). One hosts about 60 websites, and is responsible for 75% of the network load.
> 
> 	This was probably existing before, though the load, and the fact that the machine is "make worlded", new kern, and rebooted every 30 days, so it just wasn't long enough to present the problem. 
> 	Since the machine has presented no problems for the last year, I seldom look hard. Will do more next time, and give further details then..

The same thing was discussed about a week ago in the local Estonian
newsgroup, exactly same symptoms. It must be bug, most likely in the ep
driver because changing card always helps. Same OS version as well.
-- 

Vallo Kallaste
vallo@matti.ee

Comment 2 Matthew N. Dodd freebsd_committer

2000-07-18 07:54:26 UTC

Responsible Changed
From-To: freebsd-bugs->mdodd

'ep' seems to be my problem.

Comment 3 Cy Schubert 2001-03-03 18:45:54 UTC

Just occurred on a 4.2-STABLE system CVSupped on March 2 @ 04:15 PST.

FreeBSD cwsys 4.2-STABLE FreeBSD 4.2-STABLE #0: Fri Mar  2 19:28:47 PST 
2001     root@cwsys:/opt/cvs-420s/src/sys/compile/CWSYS  i386

The card is a 3C509B connected via BNC.

At the time the problem occurred, I was performing a CVS checkout to an 
NFS mounted filesystem, an rsync via the rsync protocol outbound 
through this interface, and using X from an X terminal (a smaller 
FreeBSD system on the network running clients on this system).  As far 
as this network is concerned, heavy network load.

Bringing down and up the ep0 interface fixed the problem.


Regards,                         Phone:  (250)387-8437
Cy Schubert                        Fax:  (250)387-5766
Team Leader, Sun/Alpha Team   Internet:  Cy.Schubert@osg.gov.bc.ca
Open Systems Group, ITSD, ISTA
Province of BC

Comment 4 Asymmetric 2001-08-04 21:27:06 UTC

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

This has been happening quite a bit as well on my firewall box.  The
machine is a P-133 w/48MB of RAM.. here are some relevant output messages..

This sometimes happens as soon as two hours after a reboot, I haven't
correlated it with high network traffic.  The failure appears to be a
result of using the ep device, since the ed device has no problem
continuing to function.

This machine has a very stripped down kernel running on a laptop.  This
machine was last cvsupped and installed two weeks ago to RELENG_4. (Jul
14).  The ep device is as listed in dmesg, a 3Com 3C589D.  The ed device is
a Linksys.

I can and will provide any other statistics that are requested when this
next happens.. thanks.

[vishnu-root] /root # ping 10.0.0.1
PING 10.0.0.1 (10.0.0.1): 56 data bytes
ping: sendto: No buffer space available
ping: sendto: No buffer space available
^C
- - --- 10.0.0.1 ping statistics ---
2 packets transmitted, 0 packets received, 100% packet loss

[vishnu-root] /root # netstat -m
86/112/4096 mbufs in use (current/peak/max):
          86 mbufs allocated to data
34/44/1024 mbuf clusters in use (current/peak/max)
116 Kbytes allocated to network (3% of mb_map in use)
0 requests for memory denied
0 requests for memory delayed
0 calls to protocol drain routines

[vishnu-root] /root # ifconfig -a
lo0: flags=8049<UP,LOOPBACK,RUNNING,MULTICAST> mtu 16384
          inet6 ::1 prefixlen 128
          inet6 fe80::1%lo0 prefixlen 64 scopeid 0x1
          inet 127.0.0.1 netmask 0xff000000
ppp0: flags=8010<POINTOPOINT,MULTICAST> mtu 1500
sl0: flags=c010<POINTOPOINT,LINK2,MULTICAST> mtu 552
faith0: flags=8000<MULTICAST> mtu 1500
gif0: flags=8010<POINTOPOINT,MULTICAST> mtu 1280
gif1: flags=8010<POINTOPOINT,MULTICAST> mtu 1280
gif2: flags=8010<POINTOPOINT,MULTICAST> mtu 1280
gif3: flags=8010<POINTOPOINT,MULTICAST> mtu 1280
ep0: flags=8c43<UP,BROADCAST,RUNNING,OACTIVE,SIMPLEX,MULTICAST> mtu 1500
          inet 10.0.0.2 netmask 0xffffff00 broadcast 10.0.0.255
          inet6 fe80::260:97ff:fe93:cec3%ep0 prefixlen 64 scopeid 0x9
          ether 00:60:97:93:ce:c3
          media: Ethernet 10baseT/UTP

ed0: flags=8843<UP,BROADCAST,RUNNING,SIMPLEX,MULTICAST> mtu 1500
          inet 216.239.237.193 netmask 0xfffffff0 broadcast 216.239.237.207
          inet6 fe80::2e0:98ff:fe76:cf22%ed0 prefixlen 64 scopeid 0xa
          ether 00:e0:98:76:cf:22

[vishnu-root] /root # top

last pid:   371;  load averages:  0.07,  0.02,  0.01    up 0+02:45:55  15:50:29
18 processes:  1 running, 17 sleeping
CPU states:  0.0% user,  0.0% nice,  0.0% system,  0.0% interrupt,  100% idle
Mem: 4868K Active, 5616K Inact, 4780K Wired, 28K Cache, 6560K Buf, 29M Free
Swap: 176M Total, 176M Free

- - ---cut the rest off---

[vishnu-root] /var/log # dmesg
Copyright (c) 1992-2001 The FreeBSD Project.
Copyright (c) 1979, 1980, 1983, 1986, 1988, 1989, 1991, 1992, 1993, 1994
          The Regents of the University of California. All rights reserved.
FreeBSD 4.3-STABLE #14: Sat Jul 14 21:35:17 GMT 2001
      root@vishnu.rfnj.org:/usr/src/sys/compile/VISHNU
Timecounter "i8254"  frequency 1193182 Hz
CPU: Pentium/P54C (133.22-MHz 586-class CPU)
    Origin = "GenuineIntel"  Id = 0x52c  Stepping = 12
    Features=0x1bf<FPU,VME,DE,PSE,TSC,MSR,MCE,CX8>
real memory  = 50331648 (49152K bytes)
config> di sio0
No such device: sio0
Invalid command or syntax.  Type `?' for help.
config> en pcic0
config> po pcic0 0xfcfc
config> ir pcic0 0
config> iom pcic0 0xd0000
config> f pcic0 0
config> q
avail memory = 45805568 (44732K bytes)
Preloaded elf kernel "kernel" at 0xc0332000.
Preloaded userconfig_script "/boot/kernel.conf" at 0xc033209c.
Intel Pentium detected, installing workaround for F00F bug
md0: Malloc disk
npx0: <math processor> on motherboard
npx0: INT 16 interface
pcib0: <SiS 85c501> on motherboard
pci0: <PCI bus> on pcib0
isab0: <SiS 85c503 PCI-ISA bridge> at device 1.0 on pci0
isa0: <ISA bus> on isab0
atapci0: <Generic PCI ATA controller> port
0x3f4-0x3f7,0x374-0x377,0x1f4-0x1f7,0x174-0x177 irq 14 at device 1.1 on pci0
atapci0: Busmastering DMA not supported
pci0: <Trident model 9660 VGA-compatible display device> at 17.0
pcic-pci0: <Cirrus Logic PD6729/6730 PC-Card Controller> port 0xfcfc-0xfcff
at device 19.0 on pci0
pcib1: <SiS 85c601> on motherboard
pci1: <PCI bus> on pcib1
orm0: <Option ROM> at iomem 0xc0000-0xc9fff on isa0
fdc0: <NEC 72065B or clone> 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
ata0 at port 0x1f0-0x1f7,0x3f6 irq 14 on isa0
ata1 at port 0x170-0x177,0x376 irq 15 on isa0
atkbdc0: <Keyboard controller (i8042)> at port 0x60,0x64 on isa0
atkbd0: <AT Keyboard> flags 0x1 irq 1 on atkbdc0
kbd0 at atkbd0
vga0: <Generic ISA VGA> at port 0x3c0-0x3df iomem 0xa0000-0xbffff on isa0
sc0: <System console> at flags 0x100 on isa0
sc0: VGA <16 virtual consoles, flags=0x300>
pcic0: <Cirrus logic 672x> at port 0xfcfc iomem 0xd0000 drq 0 on isa0
pcic0: Polling mode
pccard0: <PC Card bus (classic)> on pcic0
pccard1: <PC Card bus (classic)> on pcic0
IP packet filtering initialized, divert enabled, rule-based forwarding
disabled, default to deny, unlimited logging
ad0: 1376MB <TOSHIBA MK1403MAV> [2796/16/63] at ata0-master BIOSPIO
acd0: CDROM <TOSHIBA CD-ROM XM-1502B> at ata1-master using BIOSPIO
Mounting root from ufs:/dev/ad0s1a
pccard: card inserted, slot 0
pccard: card inserted, slot 1
ep0: <3Com Etherlink III 3C589> at port 0x280-0x28f irq 3 slot 0 on pccard0
ep0: Ethernet address 00:60:97:93:ce:c3
ed0 at port 0x300-0x31f irq 4 slot 1 on pccard1
ed0: address 00:e0:98:76:cf:22, type NE2000 (16 bit)

-----BEGIN PGP SIGNATURE-----
Version: PGPfreeware 7.0.3 for non-commercial use <http://www.pgp.com>

iQA/AwUBO2xamo06eIXREedLEQKC+wCfchev64mXbPj3hPnBtNbsbEKn26MAoK/w
AROgoOAeoUYPVut7IvhxpmEL
=4aWx
-----END PGP SIGNATURE-----

Comment 5 Bruce M Simpson freebsd_committer

2006-09-23 21:33:25 UTC

State Changed
From-To: open->closed

Presumably fixed by mbuf allocator changes.