New on LowEndTalk? Please Register and read our Community Rules.
All new Registrations are manually reviewed and approved, so a short delay after registration may occur before your account becomes active.
All new Registrations are manually reviewed and approved, so a short delay after registration may occur before your account becomes active.
Weird error with IP block adding to SolusVM
a-super-random-user
Member
in Help
We've recently ordered a new /24 block from one of our data centers. And after adding new block, some weird error occurred with our SolusVM node (KVM slave).
When creating new VM's with some IP's (2x.1xx.xx4.3 and some) KVM container does get internet access configurations using DHCP.
But when using some IP's like 2x.1xx.xx4.220 those containers doesn't get internet access even if it gets correct configurations from DHCP. (Tested OS's Windows Server 2008/2012, CentOS 6.5, Ubuntu 14.04)
We've configured new IP block using SolusVM GUI and also configured br0:1 to use first usable IP of the new block as Gateway for the new block.
Help is much appreciated. Thank you.
This discussion has been closed.
Comments
nullroute, contact upstream.
First usable IP shouldn't be gateway. Your gateway should be 2x.1xx.xx4.1 (Assuming that you got 2x.1xx.xx4.0/24).
This IP space is routed to one of your IPs or the Datacentre provides routing layer ?
Wow. You must have some insider information, as there is not enough info to determine that from what OP said only.
/sarcamstart
There's only 15,000 IPs that could match the specifications provided by OP. I tracerouted all of them and found a nullroute-like thing on one of them.
/sarcasmend
Naw, but seriously, the behaviour he describes I'm willing to bet on it being a CC-style nullroute.
@Rallias OP never mentioned it's CC. It might be all sort of things ranging from misconfiguration, through lack of configuration to null route. Not enough data to determine null route.
Still willing to bet dollars to doughnuts it's a CC-style null-route.
Yes. We've used x.x.x.1 as our gateway. Sorry about that typo.
None of the IP's are being null-routed.
It's being routed to First usable IP of our original block.
It's not CC. It's from ubiquity. Their 32GB server range.
Entire /24 Block is routed to the server. When I traceroute it, I can see packets are hitting the server and bouncing back. So I think you owe some dollars to @Clouvider
Thank you for your replies. We've confirmed twice that none of the IP's are null-routed and our configurations should be correct as few of the IP's are working.
so is the IP block working or not?
Can you do: ip -4 route on the VPS on the new subnet and the host node ? You can PM the result for privacy.
It do work. But some IP's aren't working.
PM'd
that means there's error in your configuration. which dc is it?
I've also contacted SolusVM to check if my configurations are incorrect. But they've confirmed that those are the correct configurations.
DC is Ubiquity
does ubiquity require IP block routed to the node?
Is that a problem : )?
From the networking point of view the host node becomes a router. Similar configuration, but you need to have forwarding switched on and gw IP brought up for each subnet on the host node. Host node routing table will be used to route packets.
well setup will be a tad different..........
Can't be more right than that
Was able to resolve this issue. Thank you @Clouvider and @Sady for your help.
FYI,
I had to recreate all the network-interface files including eth0, br0 and br0:1 using the same configurations. I believe this must be a CentOS issue or something related software.
I recreated whole network interface configurations using same configurations. Then restart the network. In my case I had to reboot the node.
Thank you for all the members who have helped me here.