10

I rebooted my server, and an odd issue just came out. I am running on ArchLinux, the clients are Ubuntu, Android and Mac.

The problem is that accessing the internet via the clients is slow, about 2ko/s and slowly stop.

But downloading something from the server to the client directly is made at full speed. And, obviously, internet from the server is at his full speed (40mo/s).

I don't know what happened from the reboot, but this issue is here on all clients, and is only related to the traffic that openvpn forward to internet.

EDIT: Tried with tcp, did not solve. EDIT: Tested various fragment/mtu settings, no changes.

Here are all my confs:

╭─<root@Alduin>-</etc/openvpn>-<1:45:07>-◇
╰─➤ cat Alduin.conf ccd/Thunderaan
local 212.83.129.104
port 1194
proto udp
dev tun
ca keys/ca.crt
cert keys/Alduin.crt
key keys/Alduin.key
dh keys/dh1024.pem
server 10.8.0.0 255.255.255.0
ifconfig-pool-persist ipp.txt
push "dhcp-option DNS 10.8.0.1"
client-to-client
keepalive 5 60
ping-timer-rem
comp-lzo
persist-key
persist-tun
status openvpn-status.log
verb 3
client-config-dir ccd
topology subnet

ccd from here +++++++++++++++


ifconfig-push 10.8.0.2 255.255.255.0
push "redirect-gateway def1"

Client conf:

client
dev tun
proto udp
remote 212.83.129.104 1194
resolv-retry infinite
nobind
persist-key
persist-tun
ca ca.crt
cert name.crt
key name.key
ns-cert-type server
comp-lzo
verb 3

and some output that might help you:

╭─<cubox@Alduin>-<~>-<1:49:43>-◇
╰─➤ ip addr
1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN
    link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00
    inet 127.0.0.1/8 scope host lo
       valid_lft forever preferred_lft forever
    inet6 ::1/128 scope host
       valid_lft forever preferred_lft forever
2: eno1: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc mq state UP qlen 1000
    link/ether b8:ac:6f:94:e2:4e brd ff:ff:ff:ff:ff:ff
    inet 88.190.15.135/24 scope global eno1
       valid_lft forever preferred_lft forever
    inet 212.83.129.104/32 scope global eno1
       valid_lft forever preferred_lft forever
    inet6 2001:bc8:300a:dead::b12d/64 scope global
       valid_lft forever preferred_lft forever
    inet6 2a01:e0b:1000:15:baac:6fff:fe94:e24e/64 scope global dynamic
       valid_lft 2592000sec preferred_lft 604800sec
    inet6 fe80::baac:6fff:fe94:e24e/64 scope link
       valid_lft forever preferred_lft forever
3: eno2: <BROADCAST,MULTICAST> mtu 1500 qdisc noop state DOWN qlen 1000
    link/ether b8:ac:6f:94:e2:4f brd ff:ff:ff:ff:ff:ff
6: tun0: <POINTOPOINT,MULTICAST,NOARP,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast state UNKNOWN qlen 100
    link/none
    inet 10.8.0.1/24 brd 10.8.0.255 scope global tun0
       valid_lft forever preferred_lft forever
╭─<cubox@Alduin>-<~>-<1:49:47>-◇
╰─➤ route
Kernel IP routing table
Destination     Gateway         Genmask         Flags Metric Ref    Use Iface
default         88-190-15-1.rev 0.0.0.0         UG    0      0        0 eno1
10.8.0.0        *               255.255.255.0   U     0      0        0 tun0
88.190.15.0     *               255.255.255.0   U     0      0        0 eno1
╭─<cubox@Alduin>-<~>-<1:49:51>-◇
╰─➤ route -6
Kernel IPv6 routing table
Destination                    Next Hop                   Flag Met Ref Use If
::1/128                        ::                         U    256 0     0 lo
2001:bc8:300a:dead::/64        ::                         U    256 0     0 eno1
2a01:e0b:1000:15::/64          ::                         UAe  256 0     0 eno1
fe80::/64                      ::                         U    256 0     0 eno1
::/0                           fe80::225:45ff:fef6:947f   UGDAe 1024 2     0 eno1
::/0                           ::                         !n   -1  1  1891 lo
::1/128                        ::                         Un   0   2  5227 lo
2001:bc8:300a:dead::/128       ::                         Un   0   1     0 lo
2001:bc8:300a:dead::b12d/128   ::                         Un   0   1   131 lo
2a01:e0b:1000:15::/128         ::                         Un   0   1     0 lo
2a01:e0b:1000:15:baac:6fff:fe94:e24e/128 ::                         Un   0   3 29356 lo
fe80::/128                     ::                         Un   0   1     0 lo
fe80::baac:6fff:fe94:e24e/128  ::                         Un   0   1   311 lo
ff00::/8                       ::                         U    256 0     0 eno1
::/0                           ::                         !n   -1  1  1891 lo



-A POSTROUTING -s 10.8.0.0/24 -o eno1 -j MASQUERADE # The iptables rule

The iptables rule here is the only that is active on the server.

╰─➤ tc qd
qdisc mq 0: dev eno1 root
qdisc pfifo_fast 0: dev tun0 root refcnt 2 bands 3 priomap  1 2 2 2 1 2 0 0 1 1 1 1 1 1 1 1

EDIT: Here is a log from the Archlinux client connecting.

Oct  2 16:54:17 Groat ovpn-openvpn[9216]: OpenVPN 2.2.1 x86_64-linux-gnu [SSL] [LZO2] [EPOLL] [PKCS11] [eurephia] [MH] [PF_INET6] [IPv6 payload 20110424-2 (2.2RC2)] built on Feb 13 2013
Oct  2 16:54:17 Groat ovpn-openvpn[9216]: NOTE: the current --script-security setting may allow this configuration to call user-defined scripts
Oct  2 16:54:17 Groat ovpn-openvpn[9216]: LZO compression initialized
Oct  2 16:54:17 Groat ovpn-openvpn[9216]: Control Channel MTU parms [ L:1542 D:138 EF:38 EB:0 ET:0 EL:0 ]
Oct  2 16:54:17 Groat ovpn-openvpn[9216]: Socket Buffers: R=[212992->131072] S=[212992->131072]
Oct  2 16:54:17 Groat ovpn-openvpn[9216]: Data Channel MTU parms [ L:1542 D:1450 EF:42 EB:135 ET:0 EL:0 AF:3/1 ]
Oct  2 16:54:17 Groat ovpn-openvpn[9216]: Local Options hash (VER=V4): '41690919'
Oct  2 16:54:17 Groat ovpn-openvpn[9216]: Expected Remote Options hash (VER=V4): '530fdded'
Oct  2 16:54:17 Groat ovpn-openvpn[9217]: UDPv4 link local: [undef]
Oct  2 16:54:17 Groat ovpn-openvpn[9217]: UDPv4 link remote: [AF_INET]212.83.129.104:1194
Oct  2 16:54:17 Groat ovpn-openvpn[9217]: TLS: Initial packet from [AF_INET]212.83.129.104:1194, sid=edfcb034 3452d72c
Oct  2 16:54:17 Groat ovpn-openvpn[9217]: VERIFY OK: depth=1, /C=FR/ST=FR/L=Paris/O=Dragonborn/CN=Dragonborn_CA/[email protected]
Oct  2 16:54:17 Groat ovpn-openvpn[9217]: VERIFY OK: nsCertType=SERVER
Oct  2 16:54:17 Groat ovpn-openvpn[9217]: VERIFY OK: depth=0, /C=FR/ST=FR/L=Paris/O=Dragonborn/CN=Dragonborn/[email protected]
Oct  2 16:54:17 Groat ovpn-openvpn[9217]: Data Channel Encrypt: Cipher 'BF-CBC' initialized with 128 bit key
Oct  2 16:54:17 Groat ovpn-openvpn[9217]: Data Channel Encrypt: Using 160 bit message hash 'SHA1' for HMAC authentication
Oct  2 16:54:17 Groat ovpn-openvpn[9217]: Data Channel Decrypt: Cipher 'BF-CBC' initialized with 128 bit key
Oct  2 16:54:17 Groat ovpn-openvpn[9217]: Data Channel Decrypt: Using 160 bit message hash 'SHA1' for HMAC authentication
Oct  2 16:54:17 Groat ovpn-openvpn[9217]: Control Channel: TLSv1, cipher TLSv1/SSLv3 DHE-RSA-AES256-SHA, 1024 bit RSA
Oct  2 16:54:17 Groat ovpn-openvpn[9217]: [Dragonborn] Peer Connection Initiated with [AF_INET]212.83.129.104:1194
Oct  2 16:54:20 Groat ovpn-openvpn[9217]: SENT CONTROL [Dragonborn]: 'PUSH_REQUEST' (status=1)
Oct  2 16:54:20 Groat ovpn-openvpn[9217]: PUSH: Received control message: 'PUSH_REPLY,dhcp-option DNS 10.8.0.1,route 212.83.129.0 255.255.255.0,route-gateway 10.8.0.1,topology subnet,ping 5,ping-restart 60,redirect-gateway def1,ifconfig 10.8.0.3 255.255.255.0'
Oct  2 16:54:20 Groat ovpn-openvpn[9217]: OPTIONS IMPORT: timers and/or timeouts modified
Oct  2 16:54:20 Groat ovpn-openvpn[9217]: OPTIONS IMPORT: --ifconfig/up options modified
Oct  2 16:54:20 Groat ovpn-openvpn[9217]: OPTIONS IMPORT: route options modified
Oct  2 16:54:20 Groat ovpn-openvpn[9217]: OPTIONS IMPORT: route-related options modified
Oct  2 16:54:20 Groat ovpn-openvpn[9217]: OPTIONS IMPORT: --ip-win32 and/or --dhcp-option options modified
Oct  2 16:54:20 Groat ovpn-openvpn[9217]: ROUTE default_gateway=192.168.1.254
Oct  2 16:54:20 Groat ovpn-openvpn[9217]: TUN/TAP device tun0 opened
Oct  2 16:54:20 Groat ovpn-openvpn[9217]: TUN/TAP TX queue length set to 100
Oct  2 16:54:20 Groat ovpn-openvpn[9217]: do_ifconfig, tt->ipv6=0, tt->did_ifconfig_ipv6_setup=0
Oct  2 16:54:20 Groat ovpn-openvpn[9217]: /sbin/ifconfig tun0 10.8.0.3 netmask 255.255.255.0 mtu 1500 broadcast 10.8.0.255
Oct  2 16:54:20 Groat ovpn-openvpn[9217]: /sbin/route add -net 212.83.129.104 netmask 255.255.255.255 gw 192.168.1.254
Oct  2 16:54:20 Groat ovpn-openvpn[9217]: /sbin/route add -net 0.0.0.0 netmask 128.0.0.0 gw 10.8.0.1
Oct  2 16:54:20 Groat ovpn-openvpn[9217]: /sbin/route add -net 128.0.0.0 netmask 128.0.0.0 gw 10.8.0.1
Oct  2 16:54:20 Groat ovpn-openvpn[9217]: /sbin/route add -net 212.83.129.0 netmask 255.255.255.0 gw 10.8.0.1
Oct  2 16:54:20 Groat ovpn-openvpn[9217]: Initialization Sequence Completed

EDIT: Here is a tcpdump of the server downloading directly a file: http://sprunge.us/aaJX Here is the client downloading this ressource: http://sprunge.us/WUCC and here is a normal client from another openvpn (working) server: http://www4.slashusr.com/57552.tcpdump

EDIT: As asked in comments, here are raw tcpdump captures. The tun0 capture from the server failed, I don't know why. Server showing outside here, client showing tun0 here, client showing outside here and server downloading directly the file here.

EDIT: The server is running an i3, which is not used at anytime (not even during openvpn use). Same for the client, i7 totaly idle.

EDIT: The issue is still here. Please, help :(

11
  • I assume you have have looked at some captures with wireshark/tcpdump? The answer almost certainly can be found in a capture, if you capture in the right place.
    – Zoredache
    Sep 30, 2013 at 5:00
  • I have a tcpdump from the eno1 interface on a download from the client and one from the server (of the same file). And one from a working openvpn client too. I'll edit the question.
    – Cubox
    Sep 30, 2013 at 12:13
  • Can you add cpu information from the client and server while the traffic is being transferred? Sep 30, 2013 at 15:17
  • In your tcpdump, I don't see slow traffic (it might be too short tho). Every client gets the same ip-address 10.8.0.2? Can you omit that and rather push a route to your network 212.83.129.0?
    – ott--
    Oct 1, 2013 at 16:15
  • Each client have his own ccd with his own ip address. I don't understand what do you mean by a route to the network.
    – Cubox
    Oct 1, 2013 at 19:01

4 Answers 4

1

I am not sure if it is same cause but I think it worth to adjust about tun-mtu and mssfix as mentioned in openvpn-on-android-tcp-retransmissions-after-openvpn-server-reboot

EDIT: I found this one might be helpful too [RESOLVED] Inacceptable openVPN performance Changing a kernel parameter: net.inet.ip.fastforwarding = 1 (add in /etc/sysctl.conf on the your linux server)

1
  • Thanks for the answer. Changing the tun-mtu and mssfix option did not helped. The fastworwarding setting don't exist in Linux. Only BSD kernels.
    – Cubox
    Oct 21, 2013 at 11:59
0

Is the vpn server also the gateway server ? try removing the push-gateway, you clients need to only have an additional route.

3
  • Where do you see a push-gateway option?
    – Cubox
    Oct 1, 2013 at 16:02
  • At CCD options there is redirect gateway. You need yo check if there is a static route on clients to theirs real GW.
    – MealstroM
    Oct 7, 2013 at 9:24
  • There is. Clients are able to talk to anything on the internet, they just do it slowly.
    – Cubox
    Oct 8, 2013 at 22:33
0

Your postrouting iptables rule looks weird, try this

-A POSTROUTING -s 10.8.0.0/24 -o eno1 -j MASQUERADE

instead of the SNAT, and delete one of the IPs on eno1, if you dont need them both. Two public IP addresses on what looks to be a ethernet-interface just looks weird. Why that setup?

My guess is your openvpn-server is looping and dropping packets back and forth, leading to this problem.

1
  • I have the masquerade rule now, it don't resolved the issue. I have two is on my server, with one of them being the failover and the other the main one. For having a server with four ips on the eth0 interface (on another Archlinux server) for over a year, I can tell you that multiples ips are not the issue here...
    – Cubox
    Oct 4, 2013 at 12:06
0

Do you run your own dns server internally? I had issues with my network when I was running a powerdns setup for internal dns but didn't have a reverse zone configured correctly. Wireshark provided the answers in that case.

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .