Restarting Network through SSH

Question

On server A, I want to issue the following commands to Server B through ssh.

service network stop
sleep 5
service network start

The problem is because I issued a network 'stop', then my current ssh connection is lost as well. Therefore I cannot execute the succeeding commands (sleep 5 and service network start). Note that I cannot use (service network restart).

Does anyone have a workaround / solution for this?

score 9 · Answer 1 · answered Jun 11 '11 at 13:32

If you are doing this interactively, why not start a screen session? It would look something like this:

screen

(scren shell starts)

service network restart

(SSH session disconnects, but the network restart continues in the screen session)

(Wait a few seconds)

(SSH back into the host once the restart finishes)

screen -r

(Reconnect to screen and check for errors)

IMHO, it's always scary to restart a network interface remotely. What happens when it doesn't come back up? Do you have a console or other means into the host if something bad happens?

score 5 · Answer 2 · answered Jun 07 '11 at 23:25

The exact commands available to do this vary based on Linux distribution. On option which is pretty standard is to schedule and "at" job for 5 seconds in the future to restart networking. Another one is to use the nohup command.

echo "sleep 5; /etc/init.d/networking start" | at now
nohup sh -c 'sleep 5; /etc/init.d/networking start' &

Other distributions have the daemon command to turn the resulting program into a daemon that is no longer associated with the shell.

score 3 · Answer 3 · answered Jun 07 '11 at 23:35

3

A very simple way to do this is by using the and operator:

service network stop && sleep 5 && service network start

answered Jun 07 '11 at 23:35

Jordan S. Jones

1,083

score 2 · Answer 4 · edited Jun 11 '11 at 14:23

2

Try this (maybe installing cron if needed):

$ at now+5min
at> service network stop
at> sleep 5
at> service network start
at> [control-D]

Then logout, wait 6 min and relogin

edited Jun 11 '11 at 14:23

Caleb

12,121
4
39
49

answered Jun 11 '11 at 13:23

RainerFromReading

21
1

score 2 · Answer 5 · answered Oct 23 '15 at 08:39

This works with modern Debian and Ubuntu, while all the other answers will not work.

screen
sudo ifdown --exclude=lo -a && sudo ifup --exclude=lo -a

Please keep in mind that it can take a little bit to get the interface back. In my case about ~15 seconds as I do have a bond.

score 1 · Answer 6 · answered Sep 06 '21 at 15:45

1

Generally, systemctl restart networking (Debian/Ubuntu), systemctl restart network (CentOS) should work, without a tmux/screen session.

answered Sep 06 '21 at 15:45

x-yuri

2,526

score 1 · Answer 7 · answered Jun 07 '11 at 23:24

1

Why not put it into a shell script and execute that via SSH?

answered Jun 07 '11 at 23:24

Christopher Armstrong

236

score 0 · Answer 8 · answered Jun 10 '11 at 03:17

0

It sounds like you want either screen or tmux. These will allow you to preserve your session through the loss of a network connection. They're really quite useful, almost all of my terminal sessions are through screen.

answered Jun 10 '11 at 03:17

Jeff Tang

151

score 0 · Answer 9 · answered Jul 23 '19 at 13:35

#!/bin/sh

# CentOS Linux release 7.4.1708 (Core) 

# 1. restart the network service
# 2. take the NIC [ens32] down
# 3. bring the NIC [ens32] up

systemctl restart network \
  && ifdown ens32 \
  && ifup ens32

#!/bin/sh

# Linux far-seer-01 4.9.0-8-amd64 #1 SMP Debian 4.9.144-3.1 (2019-02-19) x86_64 GNU/Linux

/etc/init.d/networking restart \
    && ifdown eth0 \
    && ifup eth0

e.g.

[root@localhost tmp]# ip addr show
1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN qlen 1
    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: ens32: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast state UP qlen 1000
    link/ether 00:0c:29:b3:74:da brd ff:ff:ff:ff:ff:ff
    inet 192.168.224.129/24 brd 192.168.224.255 scope global dynamic ens32
       valid_lft 1796sec preferred_lft 1796sec
    inet6 fe80::f06e:8b57:23fc:b25/64 scope link 
       valid_lft forever preferred_lft forever
[root@localhost tmp]# cat /etc/redhat-release 
CentOS Linux release 7.4.1708 (Core) 
[root@localhost tmp]# cat net.sh 
#!/bin/sh

# 1. restart the network service
# 2. take the NIC [ens32] down
# 3. bring the NIC [ens32] up

systemctl restart network \
  && ifdown ens32 \
  && ifup ens32
[root@localhost tmp]# sh net.sh 
Device 'ens32' successfully disconnected.
Connection successfully activated (D-Bus active path: /org/freedesktop/NetworkManager/ActiveConnection/17)

$ cat /etc/os-release 
PRETTY_NAME="Debian GNU/Linux 9 (stretch)"
NAME="Debian GNU/Linux"
VERSION_ID="9"
VERSION="9 (stretch)"
ID=debian
HOME_URL="https://www.debian.org/"
SUPPORT_URL="https://www.debian.org/support"
BUG_REPORT_URL="https://bugs.debian.org/"
$ cat net.sh 
#!/bin/sh

# Linux far-seer-01 4.9.0-8-amd64 #1 SMP Debian 4.9.144-3.1 (2019-02-19) x86_64 GNU/Linux

/etc/init.d/networking restart \
    &&ifdown eth0 \
    && ifup eth0
$ sh net.sh 
[ ok ] Restarting networking (via systemctl): networking.service.
ifdown: interface eth0 not configured
Internet Systems Consortium DHCP Client 4.3.5
Copyright 2004-2016 Internet Systems Consortium.
All rights reserved.
For info, please visit https://www.isc.org/software/dhcp/

Listening on LPF/eth0/00:0c:29:8a:67:72
Sending on   LPF/eth0/00:0c:29:8a:67:72
Sending on   Socket/fallback
DHCPDISCOVER on eth0 to 255.255.255.255 port 67 interval 5
DHCPREQUEST of 192.168.224.128 on eth0 to 255.255.255.255 port 67
DHCPOFFER of 192.168.224.128 from 192.168.224.254
DHCPACK of 192.168.224.128 from 192.168.224.254
bound to 192.168.224.128 -- renewal in 847 seconds.

Restarting Network through SSH

9 Answers9

Linked