PROBLEM: Memory problem 2.4.9-2.4.17

Marcin Prejsnar (alex@pronet.pl)
Thu, 24 Jan 2002 15:40:33 +0100


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

[1.] SUMMARY
Error packets received on interface ppp cause free memory decrease.

[2.] FULL DESCRIPTION
I found connection between number of error packets received from interface
ppp and amount of free memory in the system. While number of error packets
increase, parameter "skbuff_head_cache" (from /proc/slabinfo) increase too.
Every 2 errors appears - "shbuff_head_cache" rise 1. This parameter never
decrease, even after interface restart, and unload all possible modules.
Below some reports, from three points in time. In each case: first line -
just after system start, second - after several hours, third just - before
system crash (out of memory).

#ifconfig ppp0 | grep "RX packets"
RX packets:57 errors:0 dropped:0 overruns:0 frame:0
RX packets:714315 errors:9884 dropped:0 overruns:0 frame:0
RX packets:1529320 errors:46683 dropped:0 overruns:0 frame:0

#grep skbuff /pros/slabinfo
skbuff_head_cache 173 180 192 9 9 1
skbuff_head_cache 5373 5380 192 269 269 1
skbuff_head_cache 23245 23260 192 1163 1163 1

#vmstat
r b w swpd free buff cache si so bi bo in cs us sy id
0 0 0 0 33684 6812 11604 0 0 16 8 135 13 1 2 98
0 0 0 108 1892 21772 13916 0 0 0 1 300 6 0 3 97
0 0 0 4132 1964 456 2904 2 4 14 8 427 20 0 3 97

#free
total used free shared buffers cached
Mem: 62536 27100 35436 0 6740 10584
- -/+ buffers/cache: 9776 52760
Mem: 62536 60644 1892 0 21772 13916
- -/+ buffers/cache: 24956 37580
Mem: 62536 60564 1972 0 452 2900
- -/+ buffers/cache: 57212 5324

I can send you more detailed report from my system (including all slabinfo,
meminfo, and output from "vmstat" and "free" every one minute)

[3.] KEYWORDS
ppp interface, error packets, skbuff_head_cache, memory leak, linux-2.4.x,

[4.] KERNEL VERSION
Tested kernels: (All tested kernels have the same problem)
2.4.9
2.4.10
2.4.12
2.4.13
2.4.14
2.4.16
2.4.17
2.4.13-ac8

[5.] Not aplicapable ... I think...

[6.] I have replicate this problem on other hardware too. Two machines
connected with broken null-modem cable, and ping apropriate ppp interface
from one to another:
#ping -l 30000000 -s 3000 192.168.101.5

[7.] ENVIRONMEMT
[7.1] SOFTWARE
Router-firewall on RedHat linux 7.1 (with updates).

#ver_linux:
PPP 2.4.1
Linux C Library 2.2.4
Dynamic linker (ldd) 2.2.4
Procps 2.0.7
Net-tools 1.57
Modules Loaded ppp_deflate bsd_comp ppp_async serial ppp_generic slhc
isa-pnp af_packet 3c59x ipt_LOG ipt_limit iptable_filter ip_nat_ftp
iptable_nat ip_tables ip_conntrack_ftp ip_conntrack usb-uhci usbcore

[7.2] Processor info:
processor : 0
vendor_id : GenuineIntel
cpu family : 6
model : 6
model name : Celeron (Mendocino)
stepping : 5
cpu MHz : 501.143
cache size : 128 KB
fdiv_bug : no
hlt_bug : no
f00f_bug : no
coma_bug : no
fpu : yes
fpu_exception : yes
cpuid level : 2
wp : yes
flags : fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca
cmov pat pse36 mmx fxsr
bogomips : 999.42

[7.3] MODULE INFORMATION
ppp_deflate 42368 2 (autoclean)
bsd_comp 4704 0 (autoclean)
ppp_async 6768 1 (autoclean)
serial 56144 1 (autoclean)
ppp_generic 16448 3 (autoclean) [ppp_deflate bsd_comp ppp_async]
slhc 5312 0 (autoclean) [ppp_generic]
isa-pnp 30224 0 (autoclean) [serial]
af_packet 12608 0 (autoclean)
3c59x 26048 3 (autoclean)
ipt_LOG 3920 19 (autoclean)
ipt_limit 1392 17 (autoclean)
iptable_filter 2160 0 (autoclean) (unused)
ip_nat_ftp 4208 0 (unused)
iptable_nat 23376 1 [ip_nat_ftp]
ip_tables 14016 6 [ipt_LOG ipt_limit iptable_filter
iptable_nat]
ip_conntrack_ftp 4208 0 [ip_nat_ftp]
ip_conntrack 22992 2 [ip_nat_ftp iptable_nat ip_conntrack_ftp]
usb-uhci 21776 0 (unused)
usbcore 52000 1 [usb-uhci]

[7.4] Loaded driver and hardware information

#cat /proc/iomem
00000000-0009ffff : System RAM
000a0000-000bffff : Video RAM area
000c0000-000c7fff : Video ROM
000cc000-000cc7ff : Extension ROM
000cd000-000cd7ff : Extension ROM
000ce000-000ce7ff : Extension ROM
000f0000-000fffff : System ROM
00100000-03feffff : System RAM
00100000-001fc367 : Kernel code
001fc368-0024453f : Kernel data
03ff0000-03ff2fff : ACPI Non-volatile Storage
03ff3000-03ffffff : ACPI Tables
e0000000-e7ffffff : PCI Bus #01
e0000000-e3ffffff : S3 Inc. 86c368 [Trio 3D/2X]
e8000000-e9ffffff : Intel Corporation 440BX/ZX - 82443BX/ZX Host bridge
ed000000-ed00007f : 3Com Corporation 3c905C-TX [Fast Etherlink]
ed001000-ed00107f : 3Com Corporation 3c905C-TX [Fast Etherlink] (#3)
ed002000-ed00207f : 3Com Corporation 3c905C-TX [Fast Etherlink] (#2)
ed003000-ed003fff : PCI device 14d2:a003 (Titan Electronics Inc)
ed004000-ed004fff : PCI device 14d2:a003 (Titan Electronics Inc)
ed005000-ed005fff : PCI device 14d2:ffff (Titan Electronics Inc)
ed006000-ed006fff : PCI device 14d2:ffff (Titan Electronics Inc)
ffff0000-ffffffff : reserved

[8.] If you need more information please contact me: alex@pronet.pl.

[9.] I am astonished, nobody found this before...
- --
_____________________________________________
| Marcin Prejsnar <alex@pronet.pl>
| PGP key: hkp://wwwkeys.gpg.net/0x96b760e5

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.0.6 (GNU/Linux)
Comment: For info see http://www.gnupg.org

iD8DBQE8UBzphW+S5Ja3YOURAud8AJ4zkkQoxB1MVNUbhUpkZaHsYmPMcwCfd75+
f+nY+qZspVPd+EKKXui7WJY=
=qqVJ
-----END PGP SIGNATURE-----
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/