Take the 2-minute tour ×
Server Fault is a question and answer site for system and network administrators. It's 100% free, no registration required.

I have a very weird situation on my Ubuntu 12.04 LTS Server. I can not access (ping) my gateway, although I believe my config is ok - I attach the outputs. Any hints where to look? (I changed the beginning of the IP to something different, just obfuscation) ping 5.9.10.129

PING 5.9.10.129 (5.9.10.129) 56(84) bytes of data.
From 5.9.10.129 (5.9.10.129) icmp_seq=2 Destination Host Unreachable
From 5.9.10.129 (5.9.10.129) icmp_seq=3 Destination Host Unreachable
From 5.9.10.129 (5.9.10.129) icmp_seq=4 Destination Host Unreachable

uname -r

3.2.0-29-generic

ifconfig eth0

eth0      Link encap:Ethernet  HWaddr 3c:97:0e:0e:54:d7  
      inet addr:5.9.10.142  Bcast:5.9.10.159  Mask:255.255.255.224
      inet6 addr: fe80::8e70:5aff:feda:c4ac/64 Scope:Link
      UP BROADCAST RUNNING MULTICAST  MTU:1500  Metric:1
      RX packets:1216 errors:0 dropped:0 overruns:0 frame:0
      TX packets:490 errors:0 dropped:0 overruns:0 carrier:0
      collisions:0 txqueuelen:1000 
      RX bytes:107470 (107.4 KB)  TX bytes:34344 (34.3 KB)
      Interrupt:17 Memory:d2500000-d2520000

ip route

default via 5.9.10.129 dev eth0  metric 100
5.9.10.128/27 via 5.9.10.129 dev eth0 
5.9.10.128/27 dev eth0  proto kernel  scope link  src 5.9.10.142

route -n

Kernel IP routing table
Destination     Gateway         Genmask         Flags Metric Ref    Use Iface
0.0.0.0         5.9.10.129      0.0.0.0         UG    1000      0        0 eth0
5.9.10.128      5.9.10.129      255.255.255.224 UG    0         0        0 eth0
5.9.10.128      0.0.0.0         255.255.255.224 U     0         0        0 eth0

iptables -L

Chain INPUT (policy ACCEPT)
target     prot opt source               destination       


Chain FORWARD (policy ACCEPT)
target     prot opt source               destination         

Chain OUTPUT (policy ACCEPT) target     prot opt source               destination 

UPD: Eric, this is how routing information looks on a working server:

0.0.0.0         78.47.198.49    0.0.0.0         UG    100    0        0 eth0
78.47.198.48    78.47.198.49    255.255.255.240 UG    0      0        0 eth0
78.47.198.48    0.0.0.0         255.255.255.240 U     0      0        0 eth0

As I understand it, Hetzner tries to ensure security by this, so I can not take over an IP by changing my MAC. But this is another server, which has another netmask (255.255.255.240)

UPD2: BatchyX, on the working server:

78.47.198.49 dev eth0  src 78.47.198.60 
    cache 

on the broken:

5.9.10.129 dev eth0  src 5.9.10.142
    cache 
share|improve this question
1  
255.255.255.224 is a very strange netmask –  Christopher Perrin Aug 18 '12 at 13:43
1  
why is 224 a strange mask? –  Lucas Kauffman Aug 18 '12 at 13:45
3  
255.255.255.224 isn't a strange mask at all, it corresponds to a /27. –  BatchyX Aug 18 '12 at 15:17

2 Answers 2

up vote 1 down vote accepted

I'm affraid there is some problem with Hetzner IP class 5.9.x.x. We experience the same problems with our servers in this IP class. Some people are unable to connect and the traceroute show 'no route to host' at the beginning of traceroute (as if the local ISP would denied routing):

ping 5.9.200.x

Destination Host Unreachable

EDIT: found in http://www.webhostingtalk.com/archive/index.php/t-1155588.html :

Quote: "there are some ISP around (even an italian residential one) that appear to be filtering 5.0.0.0/8 ... it's not an hetzner issue! it's an issue of your home ISP!"

share|improve this answer

The 2nd route seems to be an error : "5.9.10.129/27 via 5.9.10.129 dev eth0" You should not need any route as your server is in the 5.9.10.128/27 netwoork.

share|improve this answer
    
Sorry, this was my mistake, it should read 5.9.10.128. I just checked everything again and it should be the right output now. I checked on my other server, this kind of route is on all servers (but different IP) –  leviathanus Aug 18 '12 at 14:29
    
Why would you pass all trafic to local subnet thru gateway? –  coincoin Aug 18 '12 at 14:47
    
Because this server is hosted, and this is the default route Hetzner suggests. Otherwise you do not have access to other machines. –  leviathanus Aug 18 '12 at 14:59
    
No, Route #3 should be enough to access other machines on the same subnet, unless there are some really crappy and unknown reasons. –  coincoin Aug 18 '12 at 15:04
    
Dear Eric, it seems "crappy and unknown" as I have the "same" settings on my other (working) server. I updated the post so you may see the other (working conf). –  leviathanus Aug 18 '12 at 15:08

Your Answer

 
discard

By posting your answer, you agree to the privacy policy and terms of service.

Not the answer you're looking for? Browse other questions tagged or ask your own question.