Static interface: No TCP on reboot
Shadow Hawkins on Sunday, 21 November 2004 16:53:17
Hi,
I have configured a Sixxs tunnel and subnet on my Debian box. I configured the internal interface of my server with a static IP address, which the clients on my network use as gateway. This works fine.
The problem I have is with some services I want to run for my internal network (like an SSH daemon and a DNS cacher). When I reboot my server the internal interface does not accept TCP connections. Connecting to the interface through UDP, or forwarding traffic through it to the Sixxs network works fine.
When I do a /etc/init.d/networking restart the problem goes away.
Of course the first thing I thought of was a firewall problem, but even if I do an iptables -F and set all default policies on ACCEPT the problem stays.
My /etc/network/interfaces:
auto lo
iface lo inet loopback
auto eth0
iface eth0 inet dhcp
up /sbin/wondershaper eth0 2375 360
down /sbin/wondershaper remove eth0
auto sixxs
iface sixxs inet6 v4tunnel
address 2001:838:300:1ea::2
netmask 64
endpoint 213.197.27.252
up ip link set dev sixxs mtu 1280
up ip tunnel change sixxs ttl 64
up ip route add 2000::/3 via 2001:838:300:1ea::1
auto eth1
iface eth1 inet static
address 192.168.0.1
network 192.168.0.0
netmask 255.255.255.0
broadcast 192.168.0.255
up ip address add 2001:838:3f7:0::1 dev eth1
up ip route add 2001:838:3f7:0::/64 dev eth1
auto eth2
iface eth2 inet static
address 192.168.1.1
network 192.168.1.0
netmask 255.255.255.0
broadcast 192.168.1.255
Route -A inet6:
Destination Next Hop Flags Metric Ref Use Iface
::1/128 :: U 0 0 0 lo
2001:838:300:1ea::/128 :: U 0 0 0 lo
2001:838:300:1ea::2/128 :: U 0 0 0 lo
2001:838:300:1ea::/64 :: U 256 1 0 sixxs
2001:838:3f7::1/128 :: U 0 38 0 lo
2001:838:3f7:0:202:44ff:fe18:a516/128 2001:838:3f7:0:202:44ff:fe18:a516 UC 0 6 0 eth1
2001:838:3f7::/64 :: U 1024 0 0 eth1
2000::/3 2001:838:300:1ea::1 UG 1024 501 0 sixxs
fe80::/128 :: U 0 0 0 lo
fe80::5038:5b64/128 :: U 0 0 0 lo
fe80::208:74ff:fe03:b4a/128 :: U 0 0 0 lo
fe80::250:fcff:fe0c:d134/128 :: U 0 16 0 lo
fe80::7200:ff:fe00:201d/128 :: U 0 0 0 lo
fe80::/64 :: U 256 0 0 eth0
fe80::/64 :: U 256 0 0 sixxs
fe80::/64 :: U 256 0 0 eth1
fe80::/64 :: U 256 0 0 eth2
ff02::1/128 ff02::1 UC 0 1 1 eth1
ff02::1:ff00:1/128 ff02::1:ff00:1 UC 0 1 0 eth1
ff00::/8 :: U 256 0 0 eth0
ff00::/8 :: U 256 0 0 sixxs
ff00::/8 :: U 256 0 0 eth1
ff00::/8 :: U 256 0 0 eth2
Does someone have an idea what may be causing this problem?
Static interface: No TCP on reboot
Shadow Hawkins on Sunday, 21 November 2004 18:29:45
Hi,
I'm not quite sure if the static eth-interface is your problem...
I had similar problems (box not reachable via ssh after remote reboot although sshd etc. running) when setting the sixxs iface to 'auto' in /etc/network/interfaces (running debian testing, kernel 2.6.8).
As far as I can see, it's unnecessary to do so, because 'ifup -a' is called to early (before establishing the ppp link in my case), so bringing up the iface will fail.
But this is no problem since 'aiccu' will bring the tunnel iface up once the ipv4 link is up.
But perhaps the failing 'ifup -a' causes the problems with tcp either directly or indirectly? After removing the 'auto' line the problems vanished, but I'm not sure if the prob was related to something else...
What's the ifconfig output after reboot when tcp doesn't work? All interfaces up or something missing?
HTH,
Wolfgang
Static interface: No TCP on reboot
Shadow Hawkins on Sunday, 21 November 2004 19:36:08
I have a static tunnel (so I don't use AICCU), so if I remove the 'auto' line the tunnel doesn't come up at all. I tried removing it and doing an ifup sixxs manually after a reboot though, but this didn't solve the problem.
Also, I don't think the problem is in the sixxs interface. SSH'ing to the IP of that interface works fine. It's just the statically configured interface that's giving problems.
The ifconfig output looks completely fine. All interfaces come up and are assigned the right IP addresses:
eth0 Link encap:Ethernet HWaddr 00:08:74:03:0B:4A
inet addr:80.56.91.100 Bcast:255.255.255.255 Mask:255.255.255.0
inet6 addr: fe80::208:74ff:fe03:b4a/64 Scope:Link
UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1
RX packets:887 errors:0 dropped:0 overruns:0 frame:0
TX packets:147 errors:0 dropped:0 overruns:0 carrier:0
collisions:0 txqueuelen:1000
RX bytes:66112 (64.5 KiB) TX bytes:14473 (14.1 KiB)
Interrupt:5 Base address:0xe400
eth1 Link encap:Ethernet HWaddr 00:50:FC:0C:D1:34
inet addr:192.168.0.1 Bcast:192.168.0.255 Mask:255.255.255.0
inet6 addr: fe80::250:fcff:fe0c:d134/64 Scope:Link
inet6 addr: 2001:838:3f7::1/128 Scope:Global
UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1
RX packets:124 errors:0 dropped:0 overruns:0 frame:0
TX packets:166 errors:0 dropped:0 overruns:0 carrier:0
collisions:0 txqueuelen:1000
RX bytes:12566 (12.2 KiB) TX bytes:19839 (19.3 KiB)
Interrupt:9 Base address:0xe800
eth2 Link encap:Ethernet HWaddr 70:00:00:00:20:1D
inet addr:192.168.1.1 Bcast:192.168.1.255 Mask:255.255.255.0
inet6 addr: fe80::7200:ff:fe00:201d/64 Scope:Link
UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1
RX packets:0 errors:0 dropped:0 overruns:0 frame:0
TX packets:3 errors:0 dropped:0 overruns:0 carrier:0
collisions:0 txqueuelen:1000
RX bytes:0 (0.0 b) TX bytes:278 (278.0 b)
Interrupt:10 Base address:0xec00
lo Link encap:Local Loopback
inet addr:127.0.0.1 Mask:255.0.0.0
inet6 addr: ::1/128 Scope:Host
UP LOOPBACK RUNNING MTU:16436 Metric:1
RX packets:79 errors:0 dropped:0 overruns:0 frame:0
TX packets:79 errors:0 dropped:0 overruns:0 carrier:0
collisions:0 txqueuelen:0
RX bytes:6636 (6.4 KiB) TX bytes:6636 (6.4 KiB)
sixxs Link encap:IPv6-in-IPv4
inet6 addr: fe80::5038:5b64/64 Scope:Link
inet6 addr: 2001:838:300:1ea::2/64 Scope:Global
UP POINTOPOINT RUNNING NOARP MTU:1280 Metric:1
RX packets:12 errors:0 dropped:0 overruns:0 frame:0
TX packets:11 errors:0 dropped:0 overruns:0 carrier:0
collisions:0 txqueuelen:0
RX bytes:1220 (1.1 KiB) TX bytes:1364 (1.3 KiB)
Static interface: No TCP on reboot
Shadow Hawkins on Sunday, 21 November 2004 20:17:00
Sorry, didn't recognize your tunnel is also static...
Can you test if the tcp connections are "rejected" or "dropped", i.e. is there a timeout or simply nothing?
And is the behavior the same for ipv4 and ipv6? The interface has both address types assigned, so might be that the software defaults to a protocol if both are available. If default == ipv6, do you use ip6tables?
After all, I don't believe this will solve your problem (sorry), since restart of 'networking' doesn't really do much... kernel options are set like with 'start' and then there's nothing more than ifdown -a ... ifup -a, so this doesn't look like a real problem solution. Perhaps someone else here has a better idea?
So long,
Wolfgang
Static interface: No TCP on reboot
Jeroen Massar on Monday, 22 November 2004 09:28:56 eth1 Link encap:Ethernet HWaddr 00:50:FC:0Cbiggrin1:34 inet addr:192.168.0.1 Bcast:192.168.0.255 Mask:255.255.255.0 inet6 addr: fe80::250:fcff:fe0c:d134/64 Scope:Link inet6 addr: 2001:838:3f7::1/128 Scope:Global
Shouldn't this /128 be a /64 ? :)
You do have:
up ip address add 2001:838:3f7:0::1 dev eth1 up ip route add 2001:838:3f7:0::/64 dev eth1
Change the first one to /64 and remove the second line.
Static interface: No TCP on reboot
Shadow Hawkins on Tuesday, 23 November 2004 10:11:54 You do have: up ip address add 2001:838:3f7:0::1 dev eth1 up ip route add 2001:838:3f7:0::/64 dev eth1 Change the first one to /64 and remove the second line.
I knew it had to be something simple like that. Thanks!
Kind of weird that UDP still worked though, while TCP didn't.
Posting is only allowed when you are logged in. |