Нестабільныя праблемы з "Деловая сеть, SML-Night"

[booxter]

Сітуацыя такая: ёсць роўтэр Asus WL-500g Premium (Linux, найноўшая прашыўка "ад Алега"). Ёсць правайдэр - "Деловая сеть", тарыф "SML-Night". Роўтэр (192.168.1.5) спалучаецца праз ADSL-мадэм (192.168.1.254) з PPTP VPN-серверам (172.31.3.11), з наступным канфігурацыйным файлам, аўтаматычна згенераваным праз Web-інтэрфейс (user і password змененыя):

[booxter@my root]$ cat /tmp/ppp/options.wan0
noauth refuse-eap
user 'user'
password 'password'
connect true
sync pty '/usr/sbin/pptp --idle-wait 0 172.31.3.11 --nolaunchpppd --nobuffer --sync'
lock
nomppe-stateful mtu 1400
idle 1800 demand
maxfail 0
usepeerdns
persist
ipcp-accept-remote ipcp-accept-local noipdefault
ktune
default-asyncmap nopcomp noaccomp
novj nobsdcomp nodeflate
lcp-echo-interval 10
lcp-echo-failure 6
unit 0

Глядзім табліцу маршрутаў:
[booxter@my root]$ route -n
Kernel IP routing table
Destination Gateway Genmask Flags Metric Ref Use Iface
172.31.3.11 192.168.1.254 255.255.255.255 UGH 1 0 0 vlan1
10.0.0.0 0.0.0.0 255.255.255.0 U 0 0 0 br0
192.168.1.0 0.0.0.0 255.255.255.0 U 0 0 0 vlan1
127.0.0.0 0.0.0.0 255.0.0.0 U 0 0 0 lo
0.0.0.0 212.98.160.243 0.0.0.0 UG 0 0 0 ppp0
0.0.0.0 192.168.1.254 0.0.0.0 UG 1 0 0 vlan1

Пінгуем gmail.com (ці іншы сайт):
[booxter@my root]$ ping gmail.com
PING gmail.com (64.233.171.83): 56 data bytes
64 bytes from 64.233.171.83: icmp_seq=0 ttl=239 time=160.0 ms
64 bytes from 64.233.171.83: icmp_seq=1 ttl=239 time=204.1 ms

--- gmail.com ping statistics ---
2 packets transmitted, 2 packets received, 0% packet loss
round-trip min/avg/max = 160.0/182.0/204.1 ms

Спрабуем загрузіць нешта праз http (наступныя спробы ідуць адна за адной, амаль без перапынку):
[booxter@my root]$ wget http://linux.by
Connecting to linux.by[87.252.226.1]:80
index.html 100% |********************************************************************************************************| 61056 00:00 ETA
[booxter@my root]$ wget http://linux.by
Connecting to linux.by[87.252.226.1]:80
index.html 100% |********************************************************************************************************| 61054 00:00 ETA
[booxter@my root]$ wget http://linux.by
Connecting to linux.by[87.252.226.1]:80
wget: Unable to connect to remote host (87.252.226.1): No route to host
[booxter@my root]$ wget http://linux.by
Connecting to linux.by[87.252.226.1]:80
wget: Unable to connect to remote host (87.252.226.1): No route to host
[booxter@my root]$ wget http://linux.by
Connecting to linux.by[87.252.226.1]:80
index.html 100% |********************************************************************************************************| 61054 00:00 ETA

Як бачым, http спрацоўвае, але не заўжды... Не заўжды з першага разу, не заўжды яно адразу перастае працаваць, не заўжды зноў пачынае працаваць. Прытым пінг заўжды працуе.

Вось syslog.log:
[booxter@my root]$ cat /tmp/syslog.log
Jan 1 02:00:03 syslogd started: BusyBox v1.1.3
Jan 1 02:00:03 kernel: klogd started: BusyBox v1.1.3 (2008.02.25-18:27+0000)
Jan 1 02:00:03 kernel: CPU revision is: 00029006
Jan 1 02:00:03 kernel: Primary instruction cache 16kb, linesize 16 bytes (2 ways)
Jan 1 02:00:03 kernel: Primary data cache 16kb, linesize 16 bytes (2 ways)
Jan 1 02:00:03 kernel: Linux version 2.4.20 (root@localhost) (gcc version 3.2.3 with Broadcom modifications) #91 Thu Mar 13 12:10:57 MSK 2008
Jan 1 02:00:03 kernel: Setting the PFC to its default value
Jan 1 02:00:03 kernel: Determined physical RAM map:
Jan 1 02:00:03 kernel: memory: 02000000 @ 00000000 (usable)
Jan 1 02:00:03 kernel: On node 0 totalpages: 8192
Jan 1 02:00:03 kernel: zone(0): 8192 pages.
Jan 1 02:00:03 kernel: zone(1): 0 pages.
Jan 1 02:00:03 kernel: zone(2): 0 pages.
Jan 1 02:00:03 kernel: Kernel command line: root=/dev/mtdblock2 noinitrd init=/linuxrc console=ttyS0,115200
Jan 1 02:00:03 kernel: CPU: BCM4704 rev 9 pkg 0 at 264 MHz
Jan 1 02:00:03 kernel: Calibrating delay loop... 263.78 BogoMIPS
Jan 1 02:00:03 kernel: Memory: 30100k/32768k available (1854k kernel code, 2668k reserved, 136k data, 72k init, 0k highmem)
Jan 1 02:00:03 kernel: Dentry cache hash table entries: 4096 (order: 3, 32768 bytes)
Jan 1 02:00:03 kernel: Inode cache hash table entries: 2048 (order: 2, 16384 bytes)
Jan 1 02:00:03 kernel: Mount-cache hash table entries: 512 (order: 0, 4096 bytes)
Jan 1 02:00:03 kernel: Buffer-cache hash table entries: 1024 (order: 0, 4096 bytes)
Jan 1 02:00:03 kernel: Page-cache hash table entries: 8192 (order: 3, 32768 bytes)
Jan 1 02:00:03 kernel: Checking for 'wait' instruction... unavailable.
Jan 1 02:00:03 kernel: POSIX conformance testing by UNIFIX
Jan 1 02:00:03 kernel: PCI: Initializing host
Jan 1 02:00:03 kernel: PCI: Fixing up bus 0
Jan 1 02:00:03 kernel: PCI: Fixing up bridge
Jan 1 02:00:03 kernel: PCI: Fixing up bus 1
Jan 1 02:00:03 kernel: Linux NET4.0 for Linux 2.4
Jan 1 02:00:03 kernel: Based upon Swansea University Computer Society NET3.039
Jan 1 02:00:03 kernel: Initializing RT netlink socket
Jan 1 02:00:03 kernel: Starting kswapd
Jan 1 02:00:03 kernel: Journalled Block Device driver loaded
Jan 1 02:00:03 kernel: devfs: v1.12c (20020818) Richard Gooch (rgooch@atnf.csiro.au)
Jan 1 02:00:03 kernel: devfs: boot_options: 0x1
Jan 1 02:00:03 kernel: NTFS driver v1.1.22 [Flags: R/O]
Jan 1 02:00:03 kernel: pty: 256 Unix98 ptys configured
Jan 1 02:00:03 kernel: Serial driver version 5.05c (2001-07-08) with MANY_PORTS SHARE_IRQ SERIAL_PCI enabled
Jan 1 02:00:03 kernel: ttyS00 at 0xb8000300 (irq = 3) is a 16550A
Jan 1 02:00:03 kernel: ttyS01 at 0xb8000400 (irq = 3) is a 16550A
Jan 1 02:00:03 kernel: HDLC line discipline: version $Revision$, maxframe=4096
Jan 1 02:00:03 kernel: N_HDLC line discipline registered.
Jan 1 02:00:03 kernel: loop: loaded (max 8 devices)
Jan 1 02:00:03 kernel: PPP generic driver version 2.4.2
Jan 1 02:00:03 kernel: PPP Deflate Compression module registered
Jan 1 02:00:03 kernel: PPP BSD Compression module registered
Jan 1 02:00:03 kernel: MPPE/MPPC encryption/compression module registered
Jan 1 02:00:03 kernel: PPPoL2TP kernel driver, V0.13 (oleg@cs.msu.su)
Jan 1 02:00:03 kernel: Amd/Fujitsu Extended Query Table v1.3 at 0x0040
Jan 1 02:00:03 kernel: Flash Id: Vendor: 0x0001 Device: 0x007e
Jan 1 02:00:03 kernel: number of CFI chips: 1
Jan 1 02:00:03 kernel: Flash device: 0x800000 at 0x1c000000
Jan 1 02:00:03 kernel: Physically mapped flash: squashfs filesystem found at block 909
Jan 1 02:00:03 kernel: Creating 5 MTD partitions on "Physically mapped flash":
Jan 1 02:00:03 kernel: 0x00000000-0x00040000 : "boot"
Jan 1 02:00:03 kernel: 0x00040000-0x007f0000 : "linux"
Jan 1 02:00:03 kernel: 0x000e35c0-0x007f0000 : "rootfs"
Jan 1 02:00:03 kernel: 0x007f0000-0x00800000 : "nvram"
Jan 1 02:00:03 kernel: 0x003e0000-0x007f0000 : "flashfs"
Jan 1 02:00:03 kernel: sflash: found no supported devices
Jan 1 02:00:03 kernel: NET4: Linux TCP/IP 1.0 for NET4.0
Jan 1 02:00:03 kernel: IP Protocols: ICMP, UDP, TCP, IGMP
Jan 1 02:00:03 kernel: IP: routing cache hash table of 512 buckets, 4Kbytes
Jan 1 02:00:03 kernel: TCP: Hash tables configured (established 2048 bind 4096)
Jan 1 02:00:03 kernel: Linux IP multicast router 0.06 plus PIM-SM
Jan 1 02:00:03 kernel: ip_conntrack version 2.1 (256 buckets, 2048 max) - 352 bytes per conntrack
Jan 1 02:00:03 kernel: ip_conntrack_pptp version 1.9 loaded
Jan 1 02:00:03 kernel: ip_nat_pptp version 1.5 loaded
Jan 1 02:00:03 kernel: ip_tables: (C) 2000-2002 Netfilter core team
Jan 1 02:00:03 kernel: ipt_time loading
Jan 1 02:00:03 kernel: NET4: Unix domain sockets 1.0/SMP for Linux NET4.0.
Jan 1 02:00:03 kernel: IPv6 v0.8 for NET4.0
Jan 1 02:00:03 kernel: IPv6 over IPv4 tunneling driver
Jan 1 02:00:03 kernel: NET4: Ethernet Bridge 008 for NET4.0
Jan 1 02:00:03 kernel: 802.1Q VLAN Support v1.7 Ben Greear <greearb@candelatech.com>
Jan 1 02:00:03 kernel: All bugs added by David S. Miller <davem@redhat.com>
Jan 1 02:00:03 kernel: FAT: bogus logical sector size 25600
Jan 1 02:00:03 kernel: FAT: bogus logical sector size 25600
Jan 1 02:00:03 kernel: NTFS: Unable to set blocksize 512.
Jan 1 02:00:03 kernel: VFS: Mounted root (squashfs filesystem) readonly.
Jan 1 02:00:03 kernel: Mounted devfs on /dev
Jan 1 02:00:03 kernel: Freeing unused kernel memory: 72k freed
Jan 1 02:00:03 kernel: Algorithmics/MIPS FPU Emulator v1.5
Jan 1 02:00:03 kernel: eth0: Broadcom BCM47xx 10/100 Mbps Ethernet Controller 4.108.9.0
Jan 1 02:00:03 kernel: PCI: Enabling device 01:02.0 (0004 -> 0006)
Jan 1 02:00:03 kernel: 11 Wireless Controller 4.108.9.0
Jan 1 02:00:04 kernel: device eth0 entered promiscuous mode
Jan 1 02:00:04 kernel: device vlan0 entered promiscuous mode
Jan 1 02:00:04 kernel: device eth1 entered promiscuous mode
Jan 1 02:00:04 kernel: br0: port 2(eth1) entering listening state
Jan 1 02:00:04 kernel: br0: port 1(vlan0) entering listening state
Jan 1 02:00:04 kernel: br0: port 2(eth1) entering learning state
Jan 1 02:00:04 kernel: br0: port 2(eth1) entering forwarding state
Jan 1 02:00:04 kernel: br0: topology change detected, propagating
Jan 1 02:00:04 kernel: br0: port 1(vlan0) entering learning state
Jan 1 02:00:04 kernel: br0: port 1(vlan0) entering forwarding state
Jan 1 02:00:04 kernel: br0: topology change detected, propagating
Jan 1 02:00:04 dnsmasq[74]: started, version 2.22 cachesize 150
Jan 1 02:00:04 dnsmasq[74]: DHCP, IP range 10.0.0.2 -- 10.0.0.254, lease time 24h
Jan 1 02:00:04 dnsmasq[74]: DHCP, /tmp/dnsmasq.log will be written every 8h
Jan 1 02:00:04 dnsmasq[74]: read /etc/hosts - 5 addresses
Jan 1 02:00:04 dnsmasq[74]: reading /tmp/resolv.conf
Jan 1 02:00:04 kernel: usb.c: registered new driver usbdevfs
Jan 1 02:00:04 kernel: usb.c: registered new driver hub
Jan 1 02:00:04 kernel: usb-uhci.c: $Revision: 1.275 $ time 13:26:21 Feb 24 2008
Jan 1 02:00:04 kernel: usb-uhci.c: High bandwidth mode enabled
Jan 1 02:00:04 kernel: PCI: Enabling device 01:03.0 (0000 -> 0001)
Jan 1 02:00:04 kernel: ECHI PCI device 30381106 found.
Jan 1 02:00:04 kernel: UCHI reg 0x41 = 10
Jan 1 02:00:04 kernel: UCHI reg 0x41 changed to = 0
Jan 1 02:00:04 kernel: usb-uhci.c: USB UHCI at I/O 0x100, IRQ 12
Jan 1 02:00:04 kernel: usb-uhci.c: Detected 2 ports
Jan 1 02:00:04 kernel: usb.c: new USB bus registered, assigned bus number 1
Jan 1 02:00:04 kernel: hub.c: USB hub found
Jan 1 02:00:04 kernel: hub.c: 2 ports detected
Jan 1 02:00:04 kernel: PCI: Enabling device 01:03.1 (0000 -> 0001)
Jan 1 02:00:04 kernel: ECHI PCI device 30381106 found.
Jan 1 02:00:04 kernel: UCHI reg 0x41 = 10
Jan 1 02:00:04 kernel: UCHI reg 0x41 changed to = 0
Jan 1 02:00:04 kernel: usb-uhci.c: USB UHCI at I/O 0x120, IRQ 12
Jan 1 02:00:04 kernel: usb-uhci.c: Detected 2 ports
Jan 1 02:00:04 kernel: usb.c: new USB bus registered, assigned bus number 2
Jan 1 02:00:04 kernel: hub.c: USB hub found
Jan 1 02:00:04 kernel: hub.c: 2 ports detected
Jan 1 02:00:04 kernel: usb-uhci.c: v1.275:USB Universal Host Controller Interface driver
Jan 1 02:00:04 kernel: PCI: Enabling device 01:03.2 (0000 -> 0002)
Jan 1 02:00:04 kernel: ehci_hcd 01:03.2: PCI device 1106:3104
Jan 1 02:00:04 kernel: ehci_hcd 01:03.2: irq 12, pci mem c00f6000
Jan 1 02:00:04 kernel: usb.c: new USB bus registered, assigned bus number 3
Jan 1 02:00:04 kernel: ECHI PCI device 31041106 found.
Jan 1 02:00:04 kernel: ECHI reg 0x49 = 80010f20
Jan 1 02:00:04 kernel: ECHI reg 0x49 changed to = 80010f00
Jan 1 02:00:04 kernel: ECHI reg 0x4b = 80010f09
Jan 1 02:00:04 kernel: ECHI reg 0x4b changed to = 80010f29
Jan 1 02:00:04 kernel: PCI: 01:03.2 PCI cache line size set incorrectly (0 bytes) by BIOS/FW, correcting to 32
Jan 1 02:00:04 kernel: ehci_hcd 01:03.2: USB 2.0 enabled, EHCI 1.00, driver 2003-Dec-29/2.4
Jan 1 02:00:04 kernel: hub.c: USB hub found
Jan 1 02:00:04 kernel: hub.c: 4 ports detected
Jan 1 02:00:04 kernel: usb.c: registered new driver usblp
Jan 1 02:00:04 kernel: printer.c: v0.13: USB Printer Device Class driver
Jan 1 02:00:04 kernel: Linux video capture interface: v1.00
Jan 1 02:00:05 kernel: SCSI subsystem driver Revision: 1.00
Jan 1 02:00:05 kernel: Initializing USB Mass Storage driver...
Jan 1 02:00:05 kernel: usb.c: registered new driver usb-storage
Jan 1 02:00:05 kernel: USB Mass Storage support registered.
Jan 1 02:00:05 kernel: vlan1: Setting MAC address to 00 1a 92 bc c9 22.
Jan 1 02:00:05 kernel: VLAN (vlan1): Underlying device (eth0) has same MAC, not checking promiscious mode.
Jan 1 02:00:06 pppd[93]: pppd 2.4.2 started by booxter, uid 0
Jan 1 02:00:06 pppd[93]: Using interface ppp0
Jan 1 02:00:06 pppd[93]: local IP address 10.64.64.64
Jan 1 02:00:06 pppd[93]: remote IP address 10.112.112.112
Jan 1 02:00:08 pppd[93]: Starting link
Jan 1 02:00:08 pppd[93]: Serial connection established.
Jan 1 02:00:08 pppd[93]: Connect: ppp0 <--> /dev/pts/1
Jan 1 02:00:11 login[103]: root login on `pts/0'
Jan 1 02:00:12 pppd[93]: Local IP address changed to 192.168.252.104
Jan 1 02:00:12 pppd[93]: Remote IP address changed to 212.98.160.243
Jan 1 02:00:12 dnsmasq[74]: read /etc/hosts - 5 addresses
Jan 1 02:00:12 dnsmasq[74]: reading /tmp/resolv.conf
Jan 1 02:00:12 dnsmasq[74]: using nameserver 212.98.160.65#53
Jan 1 02:00:12 dnsmasq[74]: using nameserver 212.98.160.50#53
Jan 1 02:00:12 PPTP: connect to ISP
Aug 23 18:13:02 ntp client: Synchronizing time with time.nist.gov ...

Вось resolv.conf (правайдэрскія DNS-ы):
[booxter@my root]$ cat /etc/resolv.conf
nameserver 212.98.160.50
nameserver 212.98.160.65

Такія ж сімптомы назіраю і на кліенцкіх машынах, спалучаных з інтэрнэтам праз роўтэр. У чым можа быць праблема?

[Llama]

Ёсць меркаванне што гэты персанальны гаўнароўцер не прыстасаваны для таго каб прапускаць пра сябе шмат злучэнняў і яго хапае каб абслугоўваць адзін-два кампутары, а не сетку ў кампаніі.
Магчыма проста тухне пад connection tracking. Паспрабуй з адным асобным кампом. Паназірай за колькасцю строк у /proc/net/ip_conntrack.
Я з такой праблемай не сутыкаўся, і гэта толькі здагадкі.

[kif0rt]

У меня тоже SML-Night и сейчас он сильно глючит. Год или около того работало всё без нареканий. Пару дней назад началась такая байда, что соединение устанавливается, а интернет не пингуется. Звонил им в тех. поддержку. Поначалу тупо перезагружали свой сервер и после этого связь у меня восстанавливалась, но как только я сделаю дисконнект - всё повторяется по новой. Вчера паренёк посмекалистей спросил IP vpn-сервера через который я подключаюсь. Оказалось, что я пользовался старым глючным серваком (172.31.3.2) и нужно было перейти на новый (172.31.3.11). Поэтому, если кто-то ещё пользуется не 11-м сервером, то звоните в тех. поддержку и просите, чтобы вас прописали на новый сервер. НО, проблема осталась и с новыми настройками.

P.S. Извиняюсь, что не совсем по сабжу.