LinuxQuestions.org

LinuxQuestions.org (/questions/)
-   Linux - Containers (https://www.linuxquestions.org/questions/linux-containers-122/)
-   -   How to keep the containers network device when I delete the hosts (https://www.linuxquestions.org/questions/linux-containers-122/how-to-keep-the-containers-network-device-when-i-delete-the-hosts-4175717026/)

paulspedding 09-22-2022 10:34 AM
How to keep the containers network device when I delete the hosts
 
So I got 2 identical machines in the cloud one of them has a public IP fine, without showing in the host. The other machine just will not do the same as the other instance.

So for the problematic machine every restart I do to it automatically creates a Veth on the host. Where as the working machine doesn’t. I have literally copy and pasted there profiles so they are the same except IP’s and yet they both behave differently.

Here is the working machines host and container.

Code:
root@routin:~# ip a
1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN group default qlen 1000
    link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00
    inet 127.0.0.1/8 scope host lo
      valid_lft forever preferred_lft forever
    inet6 ::1/128 scope host
      valid_lft forever preferred_lft forever
27: eth0@if2: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc noqueue state UP group default qlen 1000
    link/ether 00:26:3e:ec:b4:9c brd ff:ff:ff:ff:ff:ff link-netnsid 0
    inet 10.1.1.153/32 scope global eth0
      valid_lft forever preferred_lft forever
    inet6 fe80::216:3eff:feef:b49c/64 scope link
      valid_lft forever preferred_lft forever
root@routin:~# ip r
default via 10.1.1.1 dev eth0 proto static
10.1.1.0/24 dev eth0 proto kernel scope link src 10.1.1.153
root@routin:~# exitpaul@ubuntu:~$
paul@ubuntu:~$ ip a
1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN group default qlen 1000
    link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00
    inet 127.0.0.1/8 scope host lo
      valid_lft forever preferred_lft forever
    inet6 ::1/128 scope host
      valid_lft forever preferred_lft forever
2: eth0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc fq_codel state UP group default qlen 1000
    link/ether 02:00:25:f2:00:5b brd ff:ff:ff:ff:ff:ff
    altname enp0s3
    altname ens3
    inet 10.1.1.211/24 brd 10.1.1.255 scope global eth0
      valid_lft forever preferred_lft forever
    inet6 fe80::15ff:fef2:5e/64 scope link
      valid_lft forever preferred_lft forever
paul@ubuntu:~$ ip r
default via 10.1.1.1 dev eth0 proto static
10.1.1.0/24 dev eth0 proto kernel scope link src 10.1.1.211
paul@ubuntu:~$
And here is the broken machine:

Code:
root@jam:~# ip a
1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN group default qlen 1000
    link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00
    inet 127.0.0.1/8 scope host lo
      valid_lft forever preferred_lft forever
    inet6 ::1/128 scope host
      valid_lft forever preferred_lft forever
root@jam:~# ip r
root@jam:~# exit
exit
22/09/22  @ 16:27 @ ~
paul@demo ——›  ip a
1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN group default qlen 1000
    link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00
    inet 127.0.0.1/8 scope host lo
      valid_lft forever preferred_lft forever
    inet6 ::1/128 scope host
      valid_lft forever preferred_lft forever
2: eth0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc fq_codel state UP group default qlen 1000
    link/ether 02:00:1f:46:00:a9 brd ff:ff:ff:ff:ff:ff
    altname enp0s3
    altname ens3
    inet 10.1.1.236/24 brd 10.1.1.255 scope global eth0
      valid_lft forever preferred_lft forever
    inet 10.1.1.252/24 brd 10.1.1.255 scope global secondary eth0
      valid_lft forever preferred_lft forever
    inet 10.1.1.192/24 brd 10.1.1.255 scope global secondary eth0
      valid_lft forever preferred_lft forever
    inet 10.1.1.247/24 brd 10.1.1.255 scope global secondary eth0
      valid_lft forever preferred_lft forever
    inet 10.1.1.130/24 brd 10.1.1.255 scope global secondary eth0
      valid_lft forever preferred_lft forever
    inet 10.1.1.52/24 brd 10.1.1.255 scope global secondary eth0
      valid_lft forever preferred_lft forever
    inet6 fe30::6bff:fe2a:39/64 scope link
      valid_lft forever preferred_lft forever
22/09/22  @ 16:27 @ ~
paul@demo ——›  ip r
default via 10.1.1.1 dev eth0 proto static
10.1.1.0/24 dev eth0 proto kernel scope link src 10.1.1.236
22/09/22  @ 16:27 @ ~
paul@demo ——›
How come I can delete a interface off the host for one machine and it’s internal interface stays yet on the other it gets deleted?

paulspedding 09-23-2022 09:16 AM
I found the answer.

I was using a Macvlan in a hidden config.

sundialsvcs 09-26-2022 10:10 PM
"Glad you found it, Paul ..." Now, for the benefit of "the next poor schleb," can you please provide details ... "start to finish." What exactly is your configuration, how did you find the problem, and what exactly was wrong.

In this way, "the next poor schleb" might find a complete solution in just one thread. Instead of a :banghead: account of someone who had "found it" but didn't say exactly what "it" was.

(Begin at the beginning: "Identical machines in the cloud ..." Well, there are lots of ways to do that. And, so on.)

paulspedding 09-27-2022 04:06 AM
Quote:
Originally Posted by sundialsvcs (Post 6382792)
"Glad you found it, Paul ..." Now, for the benefit of "the next poor schleb," can you please provide details ... "start to finish." What exactly is your configuration, how did you find the problem, and what exactly was wrong.

In this way, "the next poor schleb" might find a complete solution in just one thread. Instead of a :banghead: account of someone who had "found it" but didn't say exactly what "it" was.

(Begin at the beginning: "Identical machines in the cloud ..." Well, there are lots of ways to do that. And, so on.)
Good point.


So with Linux Containers its kind of annoying when you first start as there are 3 possible configs. You have Configs, Profiles, Networks and Projects. These can all have there own type of network settings in some can override some are unique. In the Profile I was using the 'nic' type with the typemode of 'routed'.

But in the Config I was using macvlan. To see this I typed.
Code:
sudo lxc config device show jam
To which it then showed.
Code:
eth0:
  nictype: macvlan
  parent: eth0
  type: nic
Yet on the profile of this container you can see it using a Routed nictype and not a Macvlan.

You can see by typing.
Code:
sudo lxc profile show %PROFILENAMEHERE%
And the results would look like
Code:
devices:
  eth0:
    ipv4.address: 10.1.1.24
    nictype: routed
    parent: eth0
    type: nic
So it's kind of odd learning curve how you can apply multiple networking types in different ways using LXC/LXD so be carefull.


All times are GMT -5. The time now is 05:43 AM.