[ISSUE] Network get down

[didraro]

I have a big problem ... i have 3 vps : 2 Centos + 1 Windows.

Wit my actual network interfaces configuration now :


auto lo
iface lo inet loopback


auto eth0
iface eth0 inet static
address 176.223.223.114
netmask 255.255.255.252
gateway 176.223.223.113
post-up echo 1 > /proc/sys/net/ipv4/conf/eth0/proxy_arp


auto vmbr0
iface vmbr0 inet static
address 176.223.224.88
netmask 255.255.255.248
bridge_ports none
bridge_stp off
bridge_fd 0




Only the first virtual machine wit ip 176.223.224.90 can be ping and to have internet...

The 176.223.224.91 Windows , and the 176.223.224.92 Centos machine will not ping ...





I did understand that i have to change the network interface to this :

auto lo
iface lo inet loopback


auto eth0
iface eth0 inet manual


auto vmbr0
iface vmbr0 inet static
address 176.223.223.114
netmask 255.255.255.252
gateway 176.223.223.113
bridge_ports eth0
bridge_stp off
bridge_fd 0




All work for about 3 - 4 hours and after that none of the virtual machines will not have internet . I do not understand why . First time it does ping and work and after 3 -4 hours nothing worked ...


Can someone tell me or help me somehow ... please . I need help please all who can help me .

 

[thheo]

If the network config I have suggested for the host works then it has to be correct from the network topology point of view.
You have to debug what is going on when things stop working. Can't you get help from the local ISP/hosting provider?
They could tell you for example if they have the proper arp entry for the IP addresses from your /29 subnet on their router.
You could also use tcpdump to see if there is traffic and if it work in both directions.
You can also use brctl showmacs to see the mac address table properly populated
This doesn't seem to be a proxmox problem, but a network issue. You could even have faulty equipment in the network that
could cause such behaviour..

 

[didraro]

thheo i do not understand how , becouse wit your configuration was worked but after a couple of hours the internet get down from all my machines and even if i reboot the machine will not fix the problem .

To fix i have to nano interfaces and remove all and insert your config wich you gived me again and reboot and then will work again ....

What to do ?

 

[didraro]

The providers ISP they will tell that it is not a problem from them ... Really i am much unhappy ... How to make it work ... and why does it does like this ...

 

[didraro]

I did have done some test and after the internet get down from my machines i did enter in the proxmox server and try to ping the ip of machine and it does ping work ...

But when i try to ping the domain wich it is hosted on the that server i see other ip wich it is of company .

I really do not understand ...


Who can tell me ???


So inside my proxmox server i try to ping the virtual machine :

root@server:~# ping 176.223.224.90
PING 176.223.224.90 (176.223.224.90) 56(84) bytes of data.
64 bytes from 176.223.224.90: icmp_req=1 ttl=64 time=0.023 ms
64 bytes from 176.223.224.90: icmp_req=2 ttl=64 time=0.035 ms
64 bytes from 176.223.224.90: icmp_req=3 ttl=64 time=0.028 ms
64 bytes from 176.223.224.90: icmp_req=4 ttl=64 time=0.034 ms
64 bytes from 176.223.224.90: icmp_req=5 ttl=64 time=0.035 ms



But if i try to ping the domain consolight.ro wich it is hosted and pointed to the 176.223.224.90 .

root@server:~# ping consolight.ro
PING consolight.ro (176.223.208.8) 56(84) bytes of data.
64 bytes from 176.223.208.8: icmp_req=1 ttl=63 time=0.683 ms
64 bytes from 176.223.208.8: icmp_req=2 ttl=63 time=0.618 ms
64 bytes from 176.223.208.8: icmp_req=3 ttl=63 time=0.686 ms
64 bytes from 176.223.208.8: icmp_req=4 ttl=63 time=0.609 ms


I see this ip wich does not belong to me , and i think this is why my virtual machines get internet disconected ... inside them .

 

[thheo]

That cannot be the cause. What dns have you setup in /etc/resolv.conf on the proxmox server? The way I checked your domain
the IP address for consolight.ro is 224.90
Maybe there is just some DNS problem. This domain was moved from another hosting? which maybe it is routed by the same ISP and
they didn't clear the zone in their dns server

 

[didraro]

search consolight.ro
nameserver 89.36.29.240
nameserver 83.246.15.240


those are the nameservers wich i have ...

 

[didraro]

Yes it was been upgrated dns ... but why does machines vps get disconect ?

I do not understand really why .

 

[mir]

Your setup does not conform to DNS standards

; <<>> DiG 9.9.3-rpz2+rl.13214.22-P2-Debian-1:9.9.3.dfsg.P2-4 <<>> consolight.ro
;; global options: +cmd
;; Got answer:
;; ->>HEADER<<- opcode: QUERY, status: NOERROR, id: 13401
;; flags: qr rd ra; QUERY: 1, ANSWER: 1, AUTHORITY: 2, ADDITIONAL: 3


;; OPT PSEUDOSECTION:
; EDNS: version: 0, flags:; udp: 4096
;; QUESTION SECTION:
;consolight.ro.			IN	A


;; ANSWER SECTION:
consolight.ro.		14400	IN	A	176.223.224.90


;; AUTHORITY SECTION:
consolight.ro.		86400	IN	NS	ns2.consolight.ro.
consolight.ro.		86400	IN	NS	ns1.consolight.ro.


;; ADDITIONAL SECTION:
ns1.consolight.ro.	86400	IN	A	176.223.224.90
ns2.consolight.ro.	86400	IN	A	176.223.224.90


;; Query time: 115 msec
;; SERVER: 192.168.2.201#53(192.168.2.201)
;; WHEN: Sun Jan 26 14:43:20 CET 2014
;; MSG SIZE  rcvd: 126

{ns1,ns2}.consolight.ro points to the same A record!!!!

See for yourself: http://dnscheck.pingdom.com/?domain=consolight.ro

 

[didraro]

Yes becouse it is just one ip. so ns1 and ns2 will be pointed to same ip

 

[didraro]

Is there any problem or what ???

 

[thheo]

It doesn't explain the fact that you don't have network connectivity from your VMs.. but you should have 2 different IPs on different machines even.
Btw, what do you mean by the machines don't have internet after a while? Can you ping the gateway from the VM? maybe something else is going on?

 

[didraro]

I mean that virtual machines example for ip : 176.223.224.90 ( wich it is Openvz ) , 176.223.224.92 ( KVM - Windows ) does work to ping if i ping them i get reply if i ping from machines to other ip it get ping .

But after 4 -5 hours virtual machines will lose internet and if i enter from console proxmox to each machine and try to ping they not respond and not have internet on them ...


What i see it is that if i do not start the machine 176.223.224.90 and try to ping it it will give me the message :

C:\Users\RAzvan>ping 176.223.224.90


Pinging 176.223.224.90 with 32 bytes of data:
Reply from 176.223.223.114: TTL expired in transit.
Reply from 176.223.223.114: TTL expired in transit.
Reply from 176.223.223.114: TTL expired in transit.
Reply from 176.223.223.114: TTL expired in transit.


Ping statistics for 176.223.224.90:
Packets: Sent = 4, Received = 4, Lost = 0 (0% loss),


C:\Users\RAzvan>ping 176.223.224.90


Pinging 176.223.224.90 with 32 bytes of data:
Reply from 176.223.224.90: bytes=32 time=5ms TTL=54
Reply from 176.223.224.90: bytes=32 time=3ms TTL=54
Reply from 176.223.224.90: bytes=32 time=4ms TTL=54
Reply from 176.223.224.90: bytes=32 time=4ms TTL=54



i think the folowing configuration from the /etc/network/interfaces it is not corect since my machines get down ...


auto lo
iface lo inet loopback


auto eth0
iface eth0 inet manual


auto vmbr0
iface vmbr0 inet static
address 176.223.223.114
netmask 255.255.255.252
gateway 176.223.223.113
bridge_ports eth0
bridge_stp off
bridge_fd 0




Please can you help me ... Someone please .

 

[didraro]

I have 2 diferent gateways ip's maybe this is the problem ...

 

[mir]

And also there is no PTR for {ns1,ns2}.consolight.ro this could very well be a cause for missing routing.

 

[didraro]

No i think it is something from network interfaces becouse i have 2 gateways wich are diferent


1 gateway it is from main ip server

and other it is from the aditional subnet class ip

So ... Can someone tell me how to do ?

 

[thheo]

Something happens with your setup after those 4-5 hours that you mention. Check the logs from /var/log, check dmesg.. maybe is something obvious..