LXC FTP not working inside container

[mariuscotan]

Hello,

i have a problem after i have upgraded my sistem to Proxmox 4.X version. I have converted my containers to LXC but since then, i can't use ftp anymore.

My production container is a centos 5.10 x64 (i know, this is not officialy suppoorted) and i have also tried with a new machine with Debian 8.2 x64, officialy downloaded from the proxmox dashboard.

pveversion
pve-manager/4.0-57/cc7c2b53 (running kernel: 4.2.3-2-pve)

I've tried pureftpd, proftpd, vsftpd but none is working (centos and debian)
I've read in a forum that ipv6 can cause problems so i have deactivated ipv6 in my vms

vi /etc/sysctl.conf
net.ipv6.conf.all.disable_ipv6=1

I have another server with proxmox 2.x and openvz and i have there the same centos 5.10 vm and that is working without any problems.

The only message that i get in filezilla or winscp is:

In centos 5.10

Error:    Could not connect to server
Status:    Waiting to retry...
Status:    Resolving address of ar....ro
Status:    Connecting to 185.........:21...
Status:    Connection established, waiting for welcome message...
Error:    Could not connect to server

In debian 8.2

Status:    Resolving address of www2.....
Status:    Connecting to 185.1.......:21...
Status:    Connection established, waiting for welcome message...
Status:    Initializing TLS...
Status:    Verifying certificate...
Status:    TLS connection established.
Status:    Connected
Status:    Retrieving directory listing...
Command:    PWD
Response:    257 "/" is your current location
Command:    TYPE I
Response:    200 TYPE is now 8-bit binary
Command:    PORT 192,168,0,10,254,143
Response:    500 I won't open a connection to 192.168.0.10 (only to 178.115.129.162)
Command:    PASV
Response:    227 Entering Passive Mode (185,110,24,147,120,129)
Command:    MLSD
Error:    Connection timed out after 20 seconds of inactivity
Error:    Failed to retrieve directory listing

Do you have an ideea how to solve this?

Br,
Marius

 

[Rokixz]

Hello,
I am not sure if it is related to your problem, but after the last upgrade today I noticed a problem with a hostname - it put short hostname instead a full and some services refuse to start, after reboot hostname still gets shortened so services must be restarted hand to hand. Is there a way to put a normal hostname instead a shortened? What I mean server.hostname.com gets renamed to server only....

 

[dietmar]

it put short hostname instead a full and some services refuse to start, after reboot hostname still gets shortened so services must be restarted hand to hand.

/etc/hostname should not contain a FQDN. Please report a bug to the service maintainer if a service requires that. Instead, put an alias into /etc/hosts (see "man hostname")

 

[Rokixz]

/etc/hostname should not contain a FQDN. Please report a bug to the service maintainer if a service requires that. Instead, put an alias into /etc/hosts (see "man hostname")

Thank you for an update. Yes, it was set previously in /etc/hosts

Whenever I am updating /etc/hostname and /etc/hosts - the fqdn and hostname still remains the short hostname... Maybe there is still a problem with that? I am using Debian distros - Wheezy and Jessie and the situation is same. I have restarted the main node after upgrade.

 

[dietmar]

Whenever I am updating /etc/hostname and /etc/hosts

You need to set that in the container config (DNS tab on GUI)

 

[Rokixz]

You need to set that in the container config (DNS tab on GUI)

Dear dietmar,

I am very sorry for possible off topic and repeating myself. I have already stated that everything was fine before today upgrades, I had no issues since proxmox 3, I believe there might be an issue in Proxmox itself rather than my containers, as a proof I have created testing machine (Debian 8):



Commands from the virtual machine

root@test:~# hostname
test
root@test:~# hostname -f
test
root@test:~# hostname -s
test
root@test:~# cat /etc/hostname
test
root@test:~# cat /etc/hosts
127.0.0.1 localhost
::1 localhost ip6-localhost ip6-loopback
ff02::1 ip6-allnodes
ff02::2 ip6-allrouters
127.0.1.1 test

As you see no fqdn is supplied after installation, LXC config looks normal:

root@node1:~# cat /etc/pve/lxc/104.conf
arch: amd64
cpulimit: 1
cpuunits: 1024
hostname: test.machine.org
memory: 512
net0: bridge=vmbr1,hwaddr=36:65:35:63:62:64,ip=dhcp,name=eth0,type=veth
ostype: debian
rootfs: ct:vm-104-disk-1,size=8G
swap: 512


I think it would be faster to resolve an issue if you could try to set up the container, if you wish I can deploy your ssh key for investigation. Thank you for your help.

 

[asmar]

Same problem on Proxmox4. Under DNS GUI I have the full hostname but after every reboot it keeps getting the short as described above.

 

[asmar]

I'm pasting below my config + output 9just changed my IP + hostname for the example):

root@proxmox2 ~ # cat /etc/pve/lxc/101.conf
arch: amd64
cpulimit: 6
cpuunits: 1024
hostname: email1.mydomain.com
memory: 8192
nameserver: 208.67.222.222 8.8.8.8 8.8.4.4
net0: bridge=vmbr0,gw=192.168.1.10,hwaddr=32:63:32:30:38:36,ip=192.168.1.10/32,name=eth0,type=veth
ostype: debian
rootfs: local:101/vm-101-disk-1.raw,size=60G
searchdomain: google.de
swap: 8192



root@email1:~# hostname -f
email1

root@email1:~# cat /etc/hosts
127.0.0.1 localhost
::1 localhost ip6-localhost ip6-loopback
ff02::1 ip6-allnodes
ff02::2 ip6-allrouters
192.168.1.10 email1 email1.mydomain.com

 

[asmar]

/etc/hostname should not contain a FQDN. Please report a bug to the service maintainer if a service requires that. Instead, put an alias into /etc/hosts (see "man hostname")

Even if you put an alias after a reboot it change it back to the short name.