How to set the interface name in kickstart.cfg file? We know if we install CentOS system in a Server, there maybe eth0, ens0, ensxxx, eno1 and so on. I have questions, 1. Why does it generate different interface names? why not all eth0? 2. Is it possible to configure the interface name in the kickstart.cfg?

I currently have a linux system with only one physical ethernet port. This physical port has the eth0 network interface attached to it. I want to create another ethernet interface which also makes use of this single physical port. How could I achieve this using the /etc/network/interfaces file? My ultimate goal is to be able to access two different networks using a single ethernet port. For example, if I have the following interfaces file:

auto lo
iface lo inet loopback

auto eth0
iface eth0 inet static
        address 192.168.1.11
        netmask 255.255.255.0
        gateway 192.168.1.1

auto eth1
iface eth1 inet static
        address 192.168.16.100
        netmask 255.255.255.0
        gateway 192.168.16.1

I would expect to be able to ping both networks, but so far I can only ping eth0 even if I am pinging eth1 from the right network (192.168.16.0/24). I am sure I am missing some sort of configuration since it would seem too easy to simply add a new entry in the interfaces file and hope that it magically works. I am aware that I can add multiple IPs to a single ethernet interface using the command ip addr add 192.168.16.100/24 dev eth0 but I would very much prefer to have two separate interfaces. Thanks!

System is Ubuntu 16.10. It has been updated quite some times, so it might have some remaining stuff from earlier versions in there (at least back till 14.04, probably much longer). The issue it the following: I have two network interfaces in the system (on mainboard). One was always named eth0. The second was not in use. Now I have a new situation in my network and want to use the second interface. I learned a lot about interface rename and the new rename rules. So the second interface choses the name enp5s0 or similar, can't remember, therefore I need to rename it. But here is where my trouble starts. eth0 was handled in /etc/udev/rules.d/70-persistent-net.rules which, I learned, is deprecated. I tried to add a second rule for "eth1" there (using its mac instead the one of eth0), but it did not work. So after reading a bit I got rid of the old file and maybe I misunderstood something, but I figured eth* names are discouraged today, so I've chosen to call them lan0 and lan1 (which describes what they will be used for). Then I added new files in /etc/systemd/network:

10-eth-lan0.link
10-eth-lan1.link

With this content:

[Match]
MACAddress=00:23:54:96:74:07/06

[Link]
Name=lan0/1

for the different MACAddresses and the different names. For eth0 this seems to work, for eth1 (or whatever reason) it does not. And I am quite puzzled why the heck it won't work. I tripple checked the mac. dmesg | grep eth shows me this:

[    6.233684] sky2 0000:08:00.0 eth0: addr 00:23:54:96:74:07
[    6.233975] sky2 0000:05:00.0 eth1: addr 00:23:54:96:74:06
[    6.661954] sky2 0000:08:00.0 lan0: renamed from eth0
[    8.513836] sky2 0000:05:00.0 rename3: renamed from eth1

I have absolutely no clue why it settled for "rename3" now... I experimented a bit with ip link set name after I saw the systemd link file not working. What puzzles me most is that it works for eth0 -> lan0 but not for eth1 -> lan1. What can I look into to find out why the rename is not working? Who is caching stuff here? The "ip link set name" did have some persistent effects (like the change to "rename3" from the enp5s0 or similar). [edit] I recognized one more thing now: My loopback interface gets renamed to lan1. Don't know why. But it happens on every boot now. Also I got rid of the *.link files described above and the renames still happen just the same.... ??? In /etc there is now just one mention of lan0 in /etc/network/interfaces and no mention of lan1. Somebody stores that stuff somewhere and I don't know who or where.

I've searched all related questions to this topic, to no avail. I've got an ethernet card on my laptop, but the connection will not work. I checked in the BIOS, and the card is enabled. I can't undestand why mncli list my ethernet connection as "unavailable"... Here are below the results of different command I used to investigate the issue. I used Linux Mint on a Dell Latitude e5520. Any advise highly appreciated. Best regards, MC

mc@mc-Latitude-E5520:~$ lspci
00:00.0 Host bridge: Intel Corporation 2nd Generation Core Processor Family DRAM Controller (rev 09)
00:02.0 VGA compatible controller: Intel Corporation 2nd Generation Core Processor Family Integrated Graphics Controller (rev 09)
[...]
09:00.2 Mass storage controller: O2 Micro, Inc. O2 Flash Memory Card (rev 05)
0a:00.0 Ethernet controller: Broadcom Inc. and subsidiaries NetXtreme BCM5761 Gigabit Ethernet PCIe (rev 10)

mc@mc-Latitude-E5520:~$  sudo lshw -c network -sanitize
  *-network                 
       description: Wireless interface
       product: Centrino Advanced-N 6205 [Taylor Peak]
       vendor: Intel Corporation
       physical id: 0
       bus info: pci@0000:02:00.0
       logical name: wlan0
       version: 34
       serial: [REMOVED]
       width: 64 bits
       clock: 33MHz
       capabilities: pm msi pciexpress bus_master cap_list ethernet physical wireless
       configuration: broadcast=yes driver=iwlwifi driverversion=6.8.0-58-generic firmware=18.168.6.1 6000g2a-6.ucode ip=[REMOVED] latency=0 link=yes multicast=yes wireless=IEEE 802.11
       resources: irq:27 memory:e5300000-e5301fff
  *-network
       description: Ethernet interface
       product: NetXtreme BCM5761 Gigabit Ethernet PCIe
       vendor: Broadcom Inc. and subsidiaries
       physical id: 0
       bus info: pci@0000:0a:00.0
       logical name: enp10s0
       version: 10
       serial: [REMOVED]
       capacity: 1Gbit/s
       width: 64 bits
       clock: 33MHz
       capabilities: pm vpd msi pciexpress bus_master cap_list ethernet physical tp 10bt 10bt-fd 100bt 100bt-fd 1000bt 1000bt-fd autonegotiation
       configuration: autonegotiation=on broadcast=yes driver=tg3 driverversion=6.8.0-58-generic firmware=5761-v3.78 latency=0 link=no multicast=yes port=twisted pair
       resources: irq:30 memory:e4110000-e411ffff memory:e4100000-e410ffff

mc@mc-Latitude-E5520:~$ ip a
1: lo:  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 noprefixroute 
       valid_lft forever preferred_lft forever
2: enp10s0:  mtu 1500 qdisc mq state DOWN group default qlen 1000
    link/ether d0:67:e5:34:4a:16 brd ff:ff:ff:ff:ff:ff
3: wlan0:  mtu 1500 qdisc noqueue state UP group default qlen 1000
    link/ether a0:88:b4:b2:b8:c0 brd ff:ff:ff:ff:ff:ff
    inet 192.168.1.40/24 brd 192.168.1.255 scope global dynamic noprefixroute wlan0
       valid_lft 42747sec preferred_lft 42747sec
    inet6 2a01:e0a:b55:b420:2cf2:f8bb:5484:1f9e/64 scope global temporary dynamic 
       valid_lft 86140sec preferred_lft 85723sec
    inet6 2a01:e0a:b55:b420:1f08:4448:2f2f:c436/64 scope global dynamic mngtmpaddr noprefixroute 
       valid_lft 86140sec preferred_lft 86140sec
    inet6 fe80::b80:ca0c:1458:40d/64 scope link noprefixroute 
       valid_lft forever preferred_lft forever
mc@mc-Latitude-E5520:~$ ip r
default via 192.168.1.254 dev wlan0 proto dhcp src 192.168.1.40 metric 600 
192.168.1.0/24 dev wlan0 proto kernel scope link src 192.168.1.40 metric 600 

mc@mc-Latitude-E5520:~$ sudo  nmcli device
DEVICE   TYPE      STATE                   CONNECTION       
wlan0    wifi      connected               Freebox-5C7A7D 1 
lo       loopback  connected (externally)  lo               
enp10s0  ethernet  unavailable             --               
mc@mc-Latitude-E5520:~$ nmcli con show
NAME                UUID                                  TYPE      DEVICE 
Freebox-5C7A7D 1    7f9fb305-48f8-4547-bf6b-965830a9d664  wifi      wlan0  
lo                  1ad9e920-362a-459e-831d-02e79abba496  loopback  lo     
Freebox-5C7A7D      e78dddb9-7d34-4233-94bb-430b67d424c0  wifi      --     
GLORIOUS EVOLUTION  8cbbc124-2599-4247-b1cc-0fbdd2dd131e  wifi      --     
Maia's Galaxy A12   1ad8128c-afd8-4030-9f60-79a20d719ca6  wifi      --     
Wired connection 1  6bcb00fe-a7d6-3778-bcbf-85def4429cb4  ethernet  --     
mc@mc-Latitude-E5520:~$ sudo nmcli -f all connection 
NAME                UUID                                  TYPE      TIMESTAMP   TIMESTAMP-REAL                    AUTOCONNECT  AUTOCONNECT-PRIORITY  READONLY  DBUS-PATH                                   ACTIVE  DEVICE  STATE      ACTIVE>
Freebox-5C7A7D 1    7f9fb305-48f8-4547-bf6b-965830a9d664  wifi      1746012961  Wed 30 Apr 2025 01:36:01 PM CEST  yes          0                     no        /org/freedesktop/NetworkManager/Settings/4  yes     wlan0   activated  /org/f>
lo                  1ad9e920-362a-459e-831d-02e79abba496  loopback  1746012957  Wed 30 Apr 2025 01:35:57 PM CEST  no           0                     no        /org/freedesktop/NetworkManager/Settings/1  yes     lo      activated  /org/f>
Freebox-5C7A7D      e78dddb9-7d34-4233-94bb-430b67d424c0  wifi      1746010526  Wed 30 Apr 2025 12:55:26 PM CEST  yes          0                     no        /org/freedesktop/NetworkManager/Settings/6  no      --      --         --    >
GLORIOUS EVOLUTION  8cbbc124-2599-4247-b1cc-0fbdd2dd131e  wifi      1735586099  Mon 30 Dec 2024 08:14:59 PM CET   yes          0                     no        /org/freedesktop/NetworkManager/Settings/5  no      --      --         --    >
Maia's Galaxy A12   1ad8128c-afd8-4030-9f60-79a20d719ca6  wifi      1735661656  Tue 31 Dec 2024 05:14:16 PM CET   yes          0                     no        /org/freedesktop/NetworkManager/Settings/2  no      --      --         --    >
Wired connection 1  6bcb00fe-a7d6-3778-bcbf-85def4429cb4  ethernet  0           never                             yes          1                     no        /org/freedesktop/NetworkManager/Settings/3  no      --      --         --    >

I am new to the Linux kernel. Recently, I read some articles about Linux "network interface", understanding that it is a logical abstraction layer of the underlying physical adapter (NIC), which defines a set of functions (to be implemented by the driver, e.g. hard_start_xmit) to be used by the Linux kernel to talk to the NIC. In Linux, it is represented by the structure net_device. For each "network interface", there are also some other fields, such as the MAC address of the tied physical adapter, IP address/default gateway/DNS, etc. **Q1:** Is it possible that multiple "network interface" (net_device) are tied to a single physical adapter and all are in the UP state? E.g. different IP addresses and connected to different subnets? **Q2:** If Q1's answer is yes, when the network stack sends an outgoing packet to the physical adapter, how does it decide which "network interface" to use (e.g. call hard_start_xmit)? Is it based on the route table with the target IP address? **Q3:** When an incoming frame arrives at the physical adapter driver, as multiple network interfaces are tied to it, does the driver need to identify which "network interface" should handle the frame (i.e. put data into its queue/buffer)? If yes, how? I am trying to understand more details of the data send/receive process in the Linux kernel for the multiple NIC or multiple "network interface" cases.

I need to create a second separate virtual interface on Linux. I choose dummy. And do these steps: $ cat /etc/modules-load.d/dummy.conf # Load dummy.ko at boot dummy $ cat /etc/sysconfig/network-scripts/ifcfg-ethdummy1 NAME=ethdummy1 DEVICE=ethdummy1 MACADDR=00:22:22:ff:ff:ff IPADDR=10.10.10.1 NETMASK=255.255.255.0 ONBOOT=yes TYPE=Ethernet NM_CONTROLLED=no Seems like everything works. But on the net I see that people do these lines also: $ cat /etc/modprobe.d/dummy.conf install dummy /sbin/modprobe --ignore-install dummy; /sbin/ip link set name ethdummy1 dev dummy0 What's the point of this line ? Thanks.

Moving from X to Wayland caused all my mouse pointers to double in size. Fixing the system pointers was easy enough, but Blender uses its own custom pointer(s). Is there a way to make Blender use the system pointers instead of its own? Failing that, is there a way to make it less gigantic? (If it helps, GIMP seems to use the same pointer, albeit at a reasonable size... so maybe it's a GTK thing? But Inkscape uses the system pointer...) Curiously, when I first start Blender, the pointer is a reasonable size, but looks "blurry" like it's been subjected to raster resizing (left-most). Once I move it enough to trigger a pointer change, however, it turns enormous (second to left). Second-to-right is how the pointer is configured in System Settings, and what I get in most other programs. Right-most is how it looks in GIMP; also the wrong pointer (and the *same* wrong pointer), but at least the size is reasonable.

I understood the very basic concept of how to use /etc/network/interfaces, but all I find online are examples, example after example, which I can copy-paste from. What I miss is an explanation of the syntax, an explanation of the meaning of the commands and which order the commands require. I want to understand, because most of the time copy-paste is not enough, because I'm not working on a fresh machine, so I can't just overwrite existing configurations because it would break a lot of stuff. man interfaces was not very helpful since it is written very complicated. Example questions I have: what does inet in an iface line mean exactly (I could not even find it in the manpage), what does manual in an iface line mean exactly (many examples use it, but according to manpage it needs an extra config file then, which the examples don't present), when do I use or need them? When not? When I create a bridge, what exactly happens to the interfaces?

I am running Debian 9.13. I tried to check what is the difference between auto eth1 and allow-hotplug eth1 in /etc/network/interfaces. I have eth1 networking interface connected via USB. I tried rebooting, running systemctl restart networking and plugging/unplugging and it seems that main difference between allow-hotplug and auto is that if interface is marked as auto, command systemctl restart networking fails when eth1 is not connected. This leads to the conclusion that allow-hotplug is in fact preferable in all cases maybe except situation where I know that interface won't go away (lo, built-in interfaces). Is it correct? Is there any other difference?

I've got a task to write a script that displays information about which monitors are connected and which interface is used for it. The result of a script execution should look something like this:

ASUS VA24EHF                HDMI
ASUS VA24EHF                DisplayPort
Built-in monitor            eDP

The main problem is that xrandr contains information about interface but there's no information about physical device (can't understand which monitor is used here). Utils like hwinfo provide exact information about device but doesn't have one about interface it is connected via.

shell
$ xrandr | grep connected
eDP-1 connected 1368x768+1920+0 (normal left inverted right x axis y axis) 344mm x 194mm
HDMI-1 connected primary 1920x1080+0+0 (normal left inverted right x axis y axis) 527mm x 296mm
DP-1 disconnected (normal left inverted right x axis y axis)
DP-2 disconnected (normal left inverted right x axis y axis)
DP-3 disconnected (normal left inverted right x axis y axis)
HDMI-2 disconnected (normal left inverted right x axis y axis)

$ hwinfo --monitor
34: None 00.0: 10002 LCD Monitor                                
  [Created at monitor.125]
  Unique ID: rdCR.BJ+fX804+aD
  Parent ID: _Znp.DoEy7b1LcY9
  Hardware Class: monitor
  Model: "VA24EHF"
  Vendor: AUS 
  Device: eisa 0x2445 "VA24EHF"
  Serial ID: "R6LMTF187096"
  Resolution: 720x400@70Hz
  Resolution: 640x480@60Hz
  Resolution: 640x480@67Hz
  Resolution: 640x480@72Hz
  Resolution: 640x480@75Hz
  Resolution: 800x600@56Hz
  Resolution: 800x600@60Hz
  Resolution: 800x600@72Hz
  Resolution: 800x600@75Hz
  Resolution: 832x624@75Hz
  Resolution: 1024x768@60Hz
  Resolution: 1024x768@70Hz
  Resolution: 1024x768@75Hz
  Resolution: 1280x1024@75Hz
  Resolution: 1280x720@60Hz
  Resolution: 1280x960@60Hz
  Resolution: 1280x1024@60Hz
  Resolution: 1920x1080@60Hz
  Size: 527x296 mm
  Year of Manufacture: 2022
  Week of Manufacture: 51
  Detailed Timings #0:
     Resolution: 1920x1080
     Horizontal: 1920 2008 2052 2200 (+88 +132 +280) +hsync
       Vertical: 1080 1084 1089 1125 (+4 +9 +45) +vsync
    Frequencies: 148.50 MHz, 67.50 kHz, 60.00 Hz
  Driver Info #0:
    Max. Resolution: 1920x1080
    Vert. Sync Range: 48-100 Hz
    Hor. Sync Range: 30-120 kHz
    Bandwidth: 148 MHz
  Config Status: cfg=new, avail=yes, need=no, active=unknown
  Attached to: #28 (VGA compatible controller)

35: None 01.0: 10002 LCD Monitor
  [Created at monitor.125]
  Unique ID: wkFv.BKbJi+0HUrB
  Parent ID: _Znp.DoEy7b1LcY9
  Hardware Class: monitor
  Model: "BOE CQ LCD Monitor"
  Vendor: BOE "BOE CQ"
  Device: eisa 0x0747 
  Resolution: 1920x1080@60Hz
  Size: 344x194 mm
  Year of Manufacture: 2017
  Week of Manufacture: 18
  Detailed Timings #0:
     Resolution: 1920x1080
     Horizontal: 1920 1968 2000 2190 (+48 +80 +270) -hsync
       Vertical: 1080 1083 1089 1120 (+3 +9 +40) +vsync
    Frequencies: 117.73 MHz, 53.76 kHz, 48.00 Hz
  Year of Manufacture: 2017
  Week of Manufacture: 18
  Detailed Timings #1:
     Resolution: 1920x1080
     Horizontal: 1920 1968 2000 2230 (+48 +80 +310) -hsync
       Vertical: 1080 1083 1089 1140 (+3 +9 +60) +vsync
    Frequencies: 152.60 MHz, 68.43 kHz, 60.03 Hz
  Config Status: cfg=new, avail=yes, need=no, active=unknown
  Attached to: #28 (VGA compatible controller)

Could you please explain how can I get exact information which interface some physical device (monitor) are connected via?

I could test using IPERF linux Utility. But I want to specifically send/receive data on particular interface. Is there any way to do that? I tried using iperf -c server_ip -B {IP_address of my Mellanox Device}. Still I am getting speed of 9XX M/s where if the traffic was routed through Mellanox IFace it would be above 16 Gb/s. Note: I am not setting ips manually. The DHCP server assigns me automatically. Thanks

I wish you could help me with this matter which is WiFi hard blocked in my Linux machine and can't get the WiFi up. My device is an HP Pavilion G6 and WiFi card is Ralink 5390; the default of the manufacturer. The story begin when I have dual booted my Windows and ParrotOS distro my WiFi was working perfectly fine in both OS's I don’t know why but someday my WiFi decided to stop in every Linux thing! When you type the rfkill list all command in the terminal you will see response like this

0: phy0: Wireless LAN
        Soft blocked: no
        Hard blocked: yes
1: hci0: Bluetooth
        Soft blocked: no
        Hard blocked: no

Here what i did. I have read every forum in the internet and tried every solution possible for it. - Like turning off the battery and AC cable. - Tried to boot every Linux distro with usb bootable iso (Ubuntu, Debian, arch Linux, core Linux, fedora, LinuxMint and many many more ). - Tried to removing the pin 20 in my wireless card. - Tried to BlackList WMI and many more things. - Tried to restore the bios settings, it doesn't really have that much of a setting. - Tried to removing RF-kill and reinstall it (I don't know what I'm doing anymore) - Tried to turn off the Fast boot setting in Windows. And nothing has worked! Note that the WiFi is working fine in the Windows system and no issues at all. I really need it and can't afford a new WiFi adapter it's too much expensive in my city.

I want to map each hardware NIC to a specific linux network interface (ethX) persistently, as currently when an ethernet cable is reconnected, it picks an interface seemingly at random. This is trivial with either systemd or udev , but BusyBox does not seem to have a built-in way to do this. In /sys/class/net, the symlinks from ethX -> ../../devices/platform/soc/XXX.bus/XXX.ethernet/net/ethX never change. I have tried using nameif, but this only maps MAC -> interface, and the MAC address is hardcoded in u-boot for each ethX, and is not representative of the actual NIC's MAC. **EDIT:** This question will remain unanswered for now, because while I still haven't found a reliable way to map NIC -> interface in this embedded BusyBox environment, in my case, the hardware itself is also misbehaving in a way that cannot be solved with software alone.

I've configured an interface 'nonlocal' with type dummy ip link add nonlocal type dummy on that interface are some IPv6 addresses configured. Now an apache can start with some ip based vhosts on above-said ipv6 addresses. The "real" addresses get configured on the phyiscal interface if a second machine crashes (failover). Everything works a expected. But I would like to know the exact characteristics of such a dummy interface (on RHEL6 & 7). Is there a homepage or something similar which explains that type?

# Looking for a simple audio interface for Unix For fun, I'm learning x86 assembly language by porting an old game. The trickiest part is finding a way to play simple sounds (square waves, mostly). The game currently expects to be able to write directly to the speaker. How can I most easily play audio in current Unix systems? ## Criteria To be more precise, this is what I'm looking for (most important criteria first): 1. Can play a waveform. (8-bit, 8000Hz is fine). 1. Simple enough to call from assembly language. 1. Works on a modern Debian GNU/Linux system without needing root for more than simple package installation. 1. Users will not have to install (m)any libraries or other dependencies. 1. Portable across as many UNIX systems as possible. (Particularly {Net,Free,Open}BSD. I expect Solaris, MacOS, and WSL may have their own ways of doing things.) ### Preferences The code I'm porting is 32-bit x86 assembly language and I'd rather not convert it to 64-bit at this time. Also, the code does not need to link with libc and it would be nice to keep it that way, if possible. ____ #### Research Here are some of the avenues I have researched: - /dev/audio This is what I had hoped would work: I'd just open [/dev/audio](https://man.netbsd.org/audio.4) as a file and write bytes to it. Unfortunately, it appears that the Linux kernel no longer comes with that device by default, so folks would have to install a kernel module (snd-pcm-oss) as root. (Also, although this is a lesser issue, Linux uses the old SunOS semantics and does not allow multiplexing). - Direct speaker access (/dev/tty0, /dev/input/by-path/platform-pcspkr-event-spkr) This is what the game was original written to use. However, writing directly to a PC's builtin speaker requires root to configure and a kernel module (pcspkr) under Linux. Also, it does not work on laptops and other machines that don't even have a PC speaker attached. - ALSA Linux does come with a sound system that is available without a kernel module called the Advanced Linux Sound Architecture (ALSA). Other Unix systems can [emulate it](https://ftp.netbsd.org/pub/pkgsrc/current/pkgsrc/audio/alsa-lib/index.html) , although I'm not sure how well. It requires a lot more setup and although it appears it could be controlled directly through ioctl, in general it needs an external library (alsa-lib). The interface is ugly in pure assembly, but can be done in C. On Linux, ALSA is inherently dynamically linked and also uses dlopen to loads up even more libraries (pulse, pipewire) at runtime depending upon the configuration. This is a problem because the assembly code I'm working on is 32-bit, but most 64-bit boxes won't have 32-bit versions of libalsa, pulse, and pipewire installed. - Pipe to a fork/exec'd program A syscall to pipe(2) could be used to create file descriptors so the game can just send raw data to a program, such as sox's [play](https://manpages.debian.org/bookworm/sox/play.1.en.html) command, executed in a fork. While this has the nice property of pushing the problem of portability onto some other program, I'd rather avoid a runtime dependency on an external program.

When using the terminal tool ip, there is a number of flags for every interface. Example: eth0: mtu 1500 qdisc noqueue What is the meaning of M-DOWN? What command to be used to make it up or down?

By default ifconfig will show me all available interfaces , but what if I just want to display active ones? Like, en0 only in below. en0: flags=8863 mtu 1500 ether 14:10:9f:e0:eb:c9 inet6 fe80::1610:9fff:fee0:ebc9%en0 prefixlen 64 scopeid 0x4 inet X.X.X.X netmask 0xffffff00 broadcast 101.6.69.255 nd6 options=1 media: autoselect **status: active** en3: flags=8963 mtu 1500 options=60 ether 32:00:14:e7:4f:80 media: autoselect **status: inactive** Notice ifconfig en0 will not satisfy, en0 is not always the active one ;) I'm running Mac OS X.

I tried to change the network mask using Linux ip command and got confused. Google search doesn't clarify it either. If someone could explain? Say I have interface **enp0s31f6** on my Linux machine

rtuser@rtuser:~$ ip addr show dev enp0s31f6
2: enp0s31f6:  mtu 1500 qdisc fq_codel state DOWN group default qlen 1000
    link/ether 98:fa:9b:c7:37:a2 brd ff:ff:ff:ff:ff:ff

I want to assign ip address to the interface:

rtuser@rtuser:~$ sudo ip addr add 192.168.20.1/24 dev enp0s31f6
rtuser@rtuser:~$ ip addr show dev enp0s31f6
2: enp0s31f6:  mtu 1500 qdisc fq_codel state DOWN group default qlen 1000
    link/ether 98:fa:9b:c7:37:a2 brd ff:ff:ff:ff:ff:ff
    inet 192.168.20.1/24 scope global enp0s31f6
       valid_lft forever preferred_lft forever

Now I realized that my network mask is not correct, I want to change

rtuser@rtuser:~$ sudo ip addr change 192.168.20.1/25 dev enp0s31f6
rtuser@rtuser:~$ ip addr show dev enp0s31f6
2: enp0s31f6:  mtu 1500 qdisc fq_codel state DOWN group default qlen 1000
    link/ether 98:fa:9b:c7:37:a2 brd ff:ff:ff:ff:ff:ff
    inet 192.168.20.1/24 scope global enp0s31f6
       valid_lft forever preferred_lft forever
    inet 192.168.20.1/25 scope global enp0s31f6
       valid_lft forever preferred_lft forever

I got couple same ip addresses on interface. Well, maybe change doesn't work lets try to replace it

rtuser@rtuser:~$ sudo ip addr replace 192.168.20.1/26 dev enp0s31f6
rtuser@rtuser:~$ ip addr show dev enp0s31f6
2: enp0s31f6:  mtu 1500 qdisc fq_codel state DOWN group default qlen 1000
    link/ether 98:fa:9b:c7:37:a2 brd ff:ff:ff:ff:ff:ff
    inet 192.168.20.1/24 scope global enp0s31f6
       valid_lft forever preferred_lft forever
    inet 192.168.20.1/25 scope global enp0s31f6
       valid_lft forever preferred_lft forever
    inet 192.168.20.1/26 scope global enp0s31f6
       valid_lft forever preferred_lft forever

I would expect change/replace ip command would change properties of ip address, but it seems they all add/change/replace keep adding the same ip over and over again. Why is that?

Need to add a static route with nmcli to a device interface. I would like to mimic a command like this with ip route (traffic sent to the interface NOT a next-hop ip)

ip route add 4.4.0.0/30 dev ens4

I already know how to use to add a next hop interface but I cant find any documentation on how to send a route to a specific interface How can I accomplish this?

For the sake of public record, I'm asking here at SE rather than on the standardization mailing list, so that it'd be more accessible to people. With practically every headers that specify functions (or function-like interfaces), there's the following: > The following shall be declared as functions and may also be defined as macros. Function prototypes shall be provided. It's troubling me because it has the potential implication that some **functions** that I use may not be assignable to function pointers as they *may* have been defined as **macros**. However, the second part ("prototypes shall be provided") seems to suggest, that, if those said **functions** are named differently from what the standard says, then that different name must have a **identical prototype** (or at least one compatible in every enumerable aspect) to that specified in the standard. Is my interpretation correct? If not, what's the real interpretation? Can I assign the listed interfaces below that paragraph to function pointers?