You are viewing a plain text version of this content. The canonical link for it is here.
Posted to users@cloudstack.apache.org by Dan Dong <do...@gmail.com> on 2014/11/10 22:12:57 UTC
SSVM could not access Management Server.
Hi, All,
When debugging why my ISOs could not be registered, I found when I logged
into SSVM, I could not even ping the Management Server, although they are
both in the 192.168.0.0/24 network, so of course could not ping outside
world. Here are my simple network settings of my cloud(1Management Server +
1 KVM hypervisor):
1. Management server have 2 NICs:
em2 pointing outside with 10.0.0.100/24
em1 pointing inside with 192.168.0.100/24 (also serves as DNS
and Gateway of the cloud)
2. One KVM hypervisor which has 1 NIC: em1 with 192.168.0.101/24
3. VMs created on KVM hypervisor will sit on the same network of
192.168.0.0/24
The weird thing is that I can access the internet from the KVM hypervisor
as NAT is enabled on the Management Server, but for the SSVM(IP of eth2:
192.168.0.89) running on it, it could not even see the Management
Server(192.168.0.100 on em1). Should one manually re-configure routing
tables on the SSVM to solve this problem or it is caused by the initial
network design of the cloud? Thanks!
Cheers,
Dan
Re: SSVM could not access Management Server.
Posted by Dan Dong <do...@gmail.com>.
Thanks Rajani, I checked that the agent is running on SSVM:
root@s-1-VM:~# service cloud status
CloudStack cloud service (type=secstorage) is running: process id: 3887
When I run
*SSVM health check, it outputs:First DNS server is 192.168.0.100PING
192.168.0.100 (192.168.0.100): 56 data bytes64 bytes from 192.168.0.22
<http://192.168.0.22>: Destination Host UnreachableVr HL TOS Len ID Flg
off TTL Pro cks Src Dst Data 4 5 00 5400 4859 0 0040 40 01
f05f 192.168.0.22 192.168.0.100 64 bytes from 192.168.0.22
<http://192.168.0.22>: Destination Host UnreachableVr HL TOS Len ID Flg
off TTL Pro cks Src Dst Data 4 5 00 5400 4959 0 0040 40 01
ef5f 192.168.0.22 192.168.0.100 --- 192.168.0.100 ping statistics ---2
packets transmitted, 0 packets received, 100% packet lossWARNING: cannot
ping DNS serverroute followsKernel IP routing tableDestination
Gateway Genmask Flags Metric Ref Use
Iface169.254.0.0 0.0.0.0 255.255.0.0 U 0 0
0 eth0192.168.0.0 0.0.0.0 255.255.255.0 U 0 0
0 eth1192.168.0.0 0.0.0.0 255.255.255.0 U 0 0
0 eth2192.168.0.0 0.0.0.0 255.255.255.0 U 0 0
0 eth3================================================*
*Actually although DNS server is set to *
*192.168.0.100( as it is the internal NIC of the Management Server), there
are nothing on it. Just /etc/hosts mappings of hostnames and IPs.*
*Could this DNS IP settings a problem? Or I should set it to 8.8.8.8, but
this VMs are just private and not visible to the outside, confused here.*
*cheers,Dan*
2014-11-10 22:14 GMT-06:00 Rajani Karuturi <ra...@apache.org>:
> Did you check if the agent is running on SSVM?
> This wiki might help.
>
> https://cwiki.apache.org/confluence/display/CLOUDSTACK/SSVM,+templates,+Secondary+storage+troubleshooting
>
> ~Rajani
>
> On Tue, Nov 11, 2014 at 5:04 AM, Dan Dong <do...@gmail.com> wrote:
>
> > Some more info, on SSVM I can see the following routes, there is route
> > there, so why SSVM(eth2: 192.168.0.134) could not contact Management
> > Server(192.168.0.100)? Can someone help here? Thanks
> > root@s-1-VM:~# ip route
> > default via 192.168.0.100 dev eth2
> > 169.254.0.0/16 dev eth0 proto kernel scope link src 169.254.1.47
> > 172.20.10.0/24 via 192.168.0.100 dev eth1
> > 172.20.10.30 via 192.168.0.100 dev eth1
> > 192.168.0.0/24 dev eth1 proto kernel scope link src 192.168.0.22
> > 192.168.0.0/24 dev eth2 proto kernel scope link src 192.168.0.134
> > 192.168.0.0/24 dev eth3 proto kernel scope link src 192.168.0.25
> >
> >
> > 2014-11-10 15:12 GMT-06:00 Dan Dong <do...@gmail.com>:
> >
> > > Hi, All,
> > > When debugging why my ISOs could not be registered, I found when I
> > > logged into SSVM, I could not even ping the Management Server, although
> > > they are both in the 192.168.0.0/24 network, so of course could not
> ping
> > > outside world. Here are my simple network settings of my
> > cloud(1Management
> > > Server + 1 KVM hypervisor):
> > >
> > > 1. Management server have 2 NICs:
> > > em2 pointing outside with 10.0.0.100/24
> > > em1 pointing inside with 192.168.0.100/24 (also serves as
> DNS
> > > and Gateway of the cloud)
> > > 2. One KVM hypervisor which has 1 NIC: em1 with 192.168.0.101/24
> > > 3. VMs created on KVM hypervisor will sit on the same network of
> > > 192.168.0.0/24
> > >
> > > The weird thing is that I can access the internet from the KVM
> hypervisor
> > > as NAT is enabled on the Management Server, but for the SSVM(IP of
> eth2:
> > > 192.168.0.89) running on it, it could not even see the Management
> > > Server(192.168.0.100 on em1). Should one manually re-configure routing
> > > tables on the SSVM to solve this problem or it is caused by the initial
> > > network design of the cloud? Thanks!
> > >
> > > Cheers,
> > > Dan
> > >
> > >
> >
>
Re: SSVM could not access Management Server.
Posted by Rajani Karuturi <ra...@apache.org>.
Did you check if the agent is running on SSVM?
This wiki might help.
https://cwiki.apache.org/confluence/display/CLOUDSTACK/SSVM,+templates,+Secondary+storage+troubleshooting
~Rajani
On Tue, Nov 11, 2014 at 5:04 AM, Dan Dong <do...@gmail.com> wrote:
> Some more info, on SSVM I can see the following routes, there is route
> there, so why SSVM(eth2: 192.168.0.134) could not contact Management
> Server(192.168.0.100)? Can someone help here? Thanks
> root@s-1-VM:~# ip route
> default via 192.168.0.100 dev eth2
> 169.254.0.0/16 dev eth0 proto kernel scope link src 169.254.1.47
> 172.20.10.0/24 via 192.168.0.100 dev eth1
> 172.20.10.30 via 192.168.0.100 dev eth1
> 192.168.0.0/24 dev eth1 proto kernel scope link src 192.168.0.22
> 192.168.0.0/24 dev eth2 proto kernel scope link src 192.168.0.134
> 192.168.0.0/24 dev eth3 proto kernel scope link src 192.168.0.25
>
>
> 2014-11-10 15:12 GMT-06:00 Dan Dong <do...@gmail.com>:
>
> > Hi, All,
> > When debugging why my ISOs could not be registered, I found when I
> > logged into SSVM, I could not even ping the Management Server, although
> > they are both in the 192.168.0.0/24 network, so of course could not ping
> > outside world. Here are my simple network settings of my
> cloud(1Management
> > Server + 1 KVM hypervisor):
> >
> > 1. Management server have 2 NICs:
> > em2 pointing outside with 10.0.0.100/24
> > em1 pointing inside with 192.168.0.100/24 (also serves as DNS
> > and Gateway of the cloud)
> > 2. One KVM hypervisor which has 1 NIC: em1 with 192.168.0.101/24
> > 3. VMs created on KVM hypervisor will sit on the same network of
> > 192.168.0.0/24
> >
> > The weird thing is that I can access the internet from the KVM hypervisor
> > as NAT is enabled on the Management Server, but for the SSVM(IP of eth2:
> > 192.168.0.89) running on it, it could not even see the Management
> > Server(192.168.0.100 on em1). Should one manually re-configure routing
> > tables on the SSVM to solve this problem or it is caused by the initial
> > network design of the cloud? Thanks!
> >
> > Cheers,
> > Dan
> >
> >
>
Re: SSVM could not access Management Server.
Posted by dongdan39 <do...@gmail.com>.
Maybe the problem is in the NIC configurations on the KVM Hypervisor?
auto em1
iface em1 inet static
address 192.168.0.101
netmask 255.255.255.0
gateway 192.168.0.100
dns-nameservers 192.168.0.100
auto cloudbr0
.......
auto cloudbr1
.......
Note that the default GW is pointed to the internal NIC of the Management server, so why the KVM host could ping the Management server and access the outside internet, but the SSVM running in this KVM could not? Any hints ?
Dan
> 在 Nov 10, 2014，5:34 PM，Dan Dong <do...@gmail.com> 写道：
>
> Some more info, on SSVM I can see the following routes, there is route there, so why SSVM(eth2: 192.168.0.134) could not contact Management Server(192.168.0.100)? Can someone help here? Thanks
> root@s-1-VM:~# ip route
> default via 192.168.0.100 dev eth2
> 169.254.0.0/16 dev eth0 proto kernel scope link src 169.254.1.47
> 172.20.10.0/24 via 192.168.0.100 dev eth1
> 172.20.10.30 via 192.168.0.100 dev eth1
> 192.168.0.0/24 dev eth1 proto kernel scope link src 192.168.0.22
> 192.168.0.0/24 dev eth2 proto kernel scope link src 192.168.0.134
> 192.168.0.0/24 dev eth3 proto kernel scope link src 192.168.0.25
>
>
> 2014-11-10 15:12 GMT-06:00 Dan Dong <do...@gmail.com>:
>> Hi, All,
>> When debugging why my ISOs could not be registered, I found when I logged into SSVM, I could not even ping the Management Server, although they are both in the 192.168.0.0/24 network, so of course could not ping outside world. Here are my simple network settings of my cloud(1Management Server + 1 KVM hypervisor):
>>
>> 1. Management server have 2 NICs:
>> em2 pointing outside with 10.0.0.100/24
>> em1 pointing inside with 192.168.0.100/24 (also serves as DNS and Gateway of the cloud)
>> 2. One KVM hypervisor which has 1 NIC: em1 with 192.168.0.101/24
>> 3. VMs created on KVM hypervisor will sit on the same network of 192.168.0.0/24
>>
>> The weird thing is that I can access the internet from the KVM hypervisor as NAT is enabled on the Management Server, but for the SSVM(IP of eth2: 192.168.0.89) running on it, it could not even see the Management Server(192.168.0.100 on em1). Should one manually re-configure routing tables on the SSVM to solve this problem or it is caused by the initial network design of the cloud? Thanks!
>>
>> Cheers,
>> Dan
>>
>
Re: SSVM could not access Management Server.
Posted by Dan Dong <do...@gmail.com>.
Some more info, on SSVM I can see the following routes, there is route
there, so why SSVM(eth2: 192.168.0.134) could not contact Management
Server(192.168.0.100)? Can someone help here? Thanks
root@s-1-VM:~# ip route
default via 192.168.0.100 dev eth2
169.254.0.0/16 dev eth0 proto kernel scope link src 169.254.1.47
172.20.10.0/24 via 192.168.0.100 dev eth1
172.20.10.30 via 192.168.0.100 dev eth1
192.168.0.0/24 dev eth1 proto kernel scope link src 192.168.0.22
192.168.0.0/24 dev eth2 proto kernel scope link src 192.168.0.134
192.168.0.0/24 dev eth3 proto kernel scope link src 192.168.0.25
2014-11-10 15:12 GMT-06:00 Dan Dong <do...@gmail.com>:
> Hi, All,
> When debugging why my ISOs could not be registered, I found when I
> logged into SSVM, I could not even ping the Management Server, although
> they are both in the 192.168.0.0/24 network, so of course could not ping
> outside world. Here are my simple network settings of my cloud(1Management
> Server + 1 KVM hypervisor):
>
> 1. Management server have 2 NICs:
> em2 pointing outside with 10.0.0.100/24
> em1 pointing inside with 192.168.0.100/24 (also serves as DNS
> and Gateway of the cloud)
> 2. One KVM hypervisor which has 1 NIC: em1 with 192.168.0.101/24
> 3. VMs created on KVM hypervisor will sit on the same network of
> 192.168.0.0/24
>
> The weird thing is that I can access the internet from the KVM hypervisor
> as NAT is enabled on the Management Server, but for the SSVM(IP of eth2:
> 192.168.0.89) running on it, it could not even see the Management
> Server(192.168.0.100 on em1). Should one manually re-configure routing
> tables on the SSVM to solve this problem or it is caused by the initial
> network design of the cloud? Thanks!
>
> Cheers,
> Dan
>
>