cloudstack-users mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Ana Paula de Sousa <apso0...@gmail.com>
Subject Unable to ping System VMs
Date Thu, 17 Apr 2014 19:23:02 GMT
Hi,
I'm currently trying cloudstack on my college scientific project and I've
been struggling with a few things.
First of all, I have two machines with Ubuntu (where the firewall is unable
in both of them), one is acting as the hypervisor (with primary local
storage and KVM) and the other as the management server (with the secondary
storage). These machines are at a lab where they are connected physically
with each other and also with the internal network of the lab, that
provides them access to internet.

My network is like this:

- I have the 10.16.22... range to connect the hypervisor and management
server to the internet;
- I have the 192.168.22... range to connect the hypervisor with the
managament server.

On my hypervisor I have 4 bridges, cloud0 (created automatically by
cloudstack with IP 169.254.0.1), cloudbr1 (with  IP 10.16.22.100), cloudbr2
(with IP 192.168.22.70) and virbr0 (created automatically by kvm with IP
192.168.122.1).When I type brctl show it shows the following message:

bridge name    bridge id        STP enabled    interfaces
cloud0        8000.00e04c681730    no        eth0
                            vnet0
                            vnet3
                            vnet7
cloudbr1    8000.1c6f65d74a4b    no        eth2
cloudbr2    8000.5cd998b16f2d    no        eth1
                            vnet1
                            vnet2
                            vnet4
                            vnet5
                            vnet6
virbr0        8000.000000000000    yes

As we can see, cloud0 is linked to eth0, cloudbr1 is linked to eth2 and
cloudbr2 is linked to eth1.

On my Management Server I don't have any bridges, but I have 2 interfaces,
eth0 (with IP 192.168.22.71) and eth2 (with IP 10.16.22.101).

On both machines eth2 is the interface connecting them to the internet, and
they are linked physically through eth1 (on the hypervisor) and eth0 (on
the management server).

I created a basic zone with the following informations:

- IPv4 DNS1: 8.8.8.8, Internal DNS 1: 8.8.4.4, DefaultSharedNetworkOffering;
- Pod has the IP range of 192.168.22.100 to 192.168.22.150 and the gateway
192.168.22.1;
- The management traffic has the IP range of 192.168.22.2 to 192.168.22.20
and also the gateway 192.168.22.1;
- I'm also using the 192.168.22.70 of my hypervisor as the host IP and the
192.168.22.71 as the secondary storage (since the secondary storage is at
the management server);

I enabled the zone and it created 2 System VMs.

The VM Proxy has the following interfaces/IP Adresses:

Public IP Address    192.168.22.3
Private IP Address    192.168.22.116
Link Local IP Address    169.254.0.69

And the SSVM:

Public IP Address    192.168.22.2
Private IP Address    192.168.22.135
Link Local IP Address    169.254.0.55

I can ping all the interfaces of my hypervisor through the management
server (except the cloud0) and vice versa.

Both System VMs are as "Running", but I'm unable to ping them through my
hypervisor and I can't find why.

If there is any information missing here that could help to resolve this
problem I would gladly give to you.

Thanks.

-- 
Ana Paula de Sousa Oliveira
Graduando em Ciência da Computação
Universidade Federal de Goiás

Mime
  • Unnamed multipart/alternative (inline, None, 0 bytes)
View raw message