Networking Archives • Page 13 of 17 • Linux Tips

How To: Transfer Files Over SSH with SCP

In this article, you will learn how to transfer files over SSH by using ‘scp’. If you need to transfer files over a network, this is one way to do so. There are numerous other options to transfer files over a network, and this is just one of them. Feel free to leave a comment sharing other ways you use to transfer files over SSH.

We have had quite a number of articles about SSH. It stands for “Secure Shell” and you can click to find a collection of SSH articles. If you’re entirely new to SSH, you’d probably enjoy learning how to install/enable SSH. As mentioned, there have been a bunch of articles covering this subject.

Today, we’ll be using ‘scp’, which defines itself as:

OpenSSH secure file copy

It shares quite a bit of similarity with the ‘cp’ command, a command we oddly haven’t covered yet. That’s not too important. We’ll eventually get to that article and maybe remember to link back to this one. What’s important to know is that it shares some similar features and syntax.

So, today we’re going to use ‘scp’ to transfer files over SSH, over a network. It’s a pretty good skill to know, as copying files is probably the 2nd most frequently used feature over SSH, with device management being first. At least it’d be 2nd most for me, I’m not sure what you all do with it.

Transfer Files Over SSH with scp:

This article requires an open terminal, like many other articles on this site. If you don’t know how to open the terminal, you can do so with your keyboard – just press CTRL + ALT + T and your default terminal should open.

You’re also going to need a device you can SSH into, and I can’t really help you with that. You can always install OpenSSH on your desktop and practice by using SSH over localhost.

Now, with your terminal open, it’s time to transfer a file. The very most basic command to copy a file to the remote server would be:

scp <file> user@server:/path/to/folder/

1	scp <file> user@server:/path/to/folder/

You can also copy multiple files to the remote server:

scp <file_1> <file_1> user@server:/path/to/folder/

1	scp <file_1> <file_1> user@server:/path/to/folder/

You can copy all files from the directory you’re in (on the local device) easily enough. It’s just our beloved asterisk.

scp * user@server:/path/to/folder/

1	scp * user@server:/path/to/folder/

If you want to copy the files from the folder in AND the files from the folders in that directory (recursively), then the command just requires a -r flag.

scp -r * user@server:/path/to/folder/

1	scp -r * user@server:/path/to/folder/

You can actually go the other direction. If you want to transfer remote files to your local device, then you’d be looking at this command:

scp user@server:/path/to/file/<file>.txt <new_file>.txt

1	scp user@server:/path/to/file/<file>.txt <new_file>.txt

From that, you can use the above commands, extrapolate if you will, for ways to transfer the file to your local device from the remote device. It’s not terribly difficult, and ‘scp’ is something everyone should know if they’re working with remote computers on a regular basis.

By the way, setting up SSH on your computer isn’t a bad idea. Sometimes a malfunctioning computer can be accessed via SSH when nothing else will work. I have SSH installed (and secured) on pretty much every computer in the house.

Closure:

See? Now you know how to use ‘scp’ to copy files over a network via SSH. It’s a pretty awesome skill to have and one you may need until the end of time, or at least as long as Linux lasts as an operating system – which should be quite a long time. So, it’s in your best interests to learn how to use ‘scp’ with at least a basic level of skill.

Thanks for reading! If you want to help, or if the site has helped you, you can donate, register to help, write an article, or buy inexpensive hosting to start your own site. If you scroll down, you can sign up for the newsletter, vote for the article, and comment.

How To: Disable Password Login For SSH

In today’s article, I’m going to show you how to disable password login for SSH. Security is important and this can be a step in your securing SSH. After all, without password login enabled, you can’t have someone brute force your password!

And that right there nips in the bud why you’d want to make this change. You want to eliminate an attack vector, so you disable password login for SSH and you rely on SSH keys.

NOTE: I’ve had a number of SSH articles. Feel free to scroll through them, as they cover a number of SSH options. If you don’t know what SSH is, click the links in the above two paragraphs and we’ll be on the same page. In short, it’s a tool to remotely login to systems so that you can manage them without actually being at the computer physically. It’s a widely used tool.

It goes without saying that professional admins tend to do things like secure their SSH logins, and disabling password login for SSH is a good step to take. It’s also a step we laymen can use, ’cause it’s really a very simple operation. So, that’s what this article will be about. It’s an article that tells you how to …

Disable Password Login For SSH:

Now, you’re going to need to do some file editing. For that, we’re going to use our beloved nano. So, start with this command:

sudo nano /etc/ssh/sshd_config

1	sudo nano /etc/ssh/sshd_config

That will open the file with your nano editor, where you’ll look for a line that looks like this:

#PasswordAuthentication yes

1	#PasswordAuthentication yes

And you’re going to change the ‘yes’ to no, and remove the # that means ‘ignore this line’, so that it can take effect. When you’re done, the line should look like:

PasswordAuthentication no

1	PasswordAuthentication no

Now, you’ll want to save it. In nano, to save a file, you press CTRL + X, then Y, and then ENTER.

With that done, you need to restart the service. That’s easy enough, you just run the following command:

sudo systemctl restart sshd

1	sudo systemctl restart sshd

That will restart the service and use the new settings. So, be sure you have your ducks in a row, ’cause once you restart that service you’re not gonna be able to login with a password again. You’re not going to get a redo, ’cause it restarts the service – meaning it logs you out. So, you should probably be logged in using SSH certificates before making this change. That way, you know it works.

Closure:

Huh… I am not, in fact, out of ideas for articles. I’m pretty sure I’m going to run out of ideas eventually, but there seems to always be something new to impart. I did notice that I failed to search well and we have two articles about time zones and how to find out which you’re using. Ah well… Nobody else seemed to notice, so I’ve got that going for me.

How To: Make ‘curl’ Ignore Certificate Errors

In today’s article, we’re going to learn how to make ‘curl’ ignore certificate errors. If you do a lot of ‘curl’ing, this is something you’ll want to know. It’s not a dreadfully difficult task to ignore certificate errors, just a couple of options, but we might as well learn them both today.

We have previously covered the curl command, though the article only touched the surface – covering the basics that a regular Linux user might want to know. If you’re unfamiliar with curl, it’s a tool that’s used to transfer data to or from a server. It defines itself as a tool that you use to ‘transfer a URL’ and it’s an expansive application, with myriad options only a true guru would need or want to know.

What we haven’t really covered much is SSL and certificates. Briefly, SSL stands for “Secure Sockets Layer” and means that there’s a secure connection between you and the site. The certificate contains information like the URL and IP address – and is the confirmation used in the secure socket layer. Meaning, the certificate matches the site and this confirmation is what lets you use SSL without any warnings. Any break in the chain should throw an error up on your screen about a broken or missing certificate.

But, what if you still need that information? What if that data is essential? If the certificate is broken then curl will throw an error and not complete the transfer. It’s for this reason that you’ll want to learn how to …

Make ‘curl’ Ignore Certificate Errors:

Obviously, curl is an application used in the terminal, so this article requires an open terminal. If you don’t know how to open the terminal, you can do so with your keyboard – just press CTRL + ALT + T and your default terminal should open.

These days, everything is expected to have a security certificate and SSL. Even this site has one, as you can tell by the https:// in the URL. Some folks want them for everything on the web, but I’d contend not every site really needs to have one – especially sites that aren’t interactive and don’t collect personal information. But, I have one and would have one regardless – simply because we do exchange some personal information (like email addresses) and I want folks to know we take security seriously.

Moving on…

The syntax is simple and, again, we’re only tackling part of the curl application. It’s simply too large a program, with too many variables, to cover it all in just one article. You basically have two choices:

curl -k <external-server-address>

1	curl -k <external-server-address>

And the other option is:

curl --insecure <external-server-address>

1	curl --insecure <external-server-address>

Either of those will let you make curl ignore certificate errors, allowing you to fetch whatever it is you were after. I suppose you should be careful with this, always verifying what you fetch is what you were actually after. Be extra careful to ensure the address is the one intended, of course. Just practice some careful scrutiny and you’re likely to be just fine.

Closure:

Yup. Another article. This one will help you use curl and to ignore certificate errors. It’s especially useful if you use curl a great deal. If not, stick it in the back of your memory banks and recall it when you do end up needing it. You never know when a tool like this will come in handy.

Monitor TCP/UDP In Real Time

If you’re concerned about network connections to your device, you may want to monitor TCP/UDP in real time. It’s a pretty handy way to get a handle on what connections are being made by your devices. This article could be pretty complicated, but I’ll try to make sense of it all.

By the way, if you ever want to be sure you know something, try explaining it to someone who knows nothing about it. It can be pretty humbling. Leave a comment letting me know how I did.

Anyhow, moving on…

Network connections happen on various ports. Think of them as though your computer is a country and you have various ports that let traffic in and let traffic out.

There are different kinds of traffic, just like there are different kinds of ships. You have leisure, goods, military ships, etc… You have cruise ships, tankers, cargo yachts, kayaks, aircraft carriers, etc. Traffic is often bidirectional, meaning going in and leaving on the same port. In the case of TCP and UDP, traffic is bidirectional so one port will accommodate both.

So…

What Are TCP and UDP Anyhow?

In the case of TCP, you may be familiar with the expression TCP/IP, and it stands for Transmission Control Protocol. TCP requires a server/client relationship and should only be used on certain ports. Those ports are things like FTP, SSH, SMTP, Time, TELNET, etc… So, if you know you don’t have an FTP server running, you really shouldn’t be seeing traffic on ports 20 or 21.

UDP, on the other hand, stands for “User Datagram Protocol”. Unlike UDP, there’s no client/server relationship required. It just spews traffic out as needed/ordered. UDP connections do things like check a DNS server to find the IP address for a domain name. It then waits for the server to respond. There’s less latency and it’s more useful for ‘real time’ processes. Voice and video may be sent using UDP. Again, if you’re seeing unrecognized traffic, there may be a problem.

If you want, you can get a general idea (you don’t really need to memorize this) of what belongs where, with regards of TCP/UDP traffic. Wikipedia has an excellent list of TCP/UDP port numbers.

So, with that information at hand, you can see why you might want to …

Monitor TCP/UDP In Real Time:

Once you have your terminal open, you can use ‘netstat’ and examine this in detail. If you just want to check once, you can run this command:

sudo netstat -tulpn

1	sudo netstat -tulpn

To monitor it in real time, you’ll just use the ‘watch’ command. That’d look like this:

sudo watch netstat -tulpn

1	sudo watch netstat -tulpn

You can use the -n flag to change the refresh interval, as it will default to every two seconds. To make it every five seconds, the command would look like this:

sudo watch -n 5 netstat -tulpn

1	sudo watch -n 5 netstat -tulpn

You can change that number, but you can go no lower than once ever 0.1 seconds. I’m not sure if there’s an upper threshold, but it might have one.

Either way, you can monitor your TCP/IP connections in real time. It’s not really all that difficult. You may see some results that alarm you, but odds are you’re just new to checking the output. Before getting alarmed and making drastic changes to your computer, research to see if the connection is actually just normal traffic.

Closure:

And there you have it, you have another article to read. This one is about how you can monitor TCP/UDP in real time, a useful tool if you’re concerned with your network’s traffic. Malicious activities will likely require network ingress and egress, so this can help your security assessments. Strange connections don’t necessarily mean there’s a problem – but they do mean you might want to look to see what’s causing the connections.

Change Your DNS Servers To Google’s In Lubuntu

This article will show you how to change your DNS servers from your default servers to the DNS servers run by Google, specifically while you’re using Lubuntu. If you’re not using Lubuntu, the process is likely fairly similar. Either way, it should be a nice and fun article, and we’ll even do it in the GUI instead of the terminal.

So, what is DNS? DNS stands for ‘Domain Name System’. As you know by now, machines are identified by their IP address. It’d suck to have to remember numbers instead of names. It’s also possible to route multiple domain names to the same IP address. So, we have domain names and use those domain names to resolve to IP addresses.

If you’re like most people, right now the DNS servers you’re using have come from your ISP – the folks who provide your internet service. This means that they can see which sites you visit, based on the requests you make to the DNS service.

Some folks don’t like this and prefer to find another DNS provider. (There’s also Secure DNS which this article will not be touching on.) One of those companies that provides free DNS servers is Google. Like them or not, their DNS servers are robust and consistently updated, often making domain propagation quicker for you.

This article is for Lubuntu, as stated above, but you may very well be able to follow the same exact steps with your distro of choice. And, now that you have a general idea of what’s going on, let’s learn how to…

Change Your DNS Servers:

To get started with changing your DNS servers, you need to find your networking icon in your system tray. It’ll be down on the right, not far from the clock. Once you have found it, right click on it so that it brings up the menu to let you “Edit Connections”. It will look something like this:

change network settings — Of course, your version won’t have the nifty arrow.

You’ll want to click the gear icon. That’s why I put the arrow there!

That will open another window. This window will have tabs you need to worry about – or a tab you need to know about. You probably shouldn’t need both. The tabs you’re interested in will look like this:

changing the network connections — You should need one of those, probably not both of those…

Now, you should only need to edit one of those. If you’re still using IPv4 then you use that tab. If you’re using IPv6 then you’ll obviously want to use the appropriate tab. For example, the IPv4 would look like this:

screen to edit dns servers — This would be the tab you’re looking for, pretty much…

Now, where that arrow is is where you want to enter the new DNS server information. You separate them with a comma, though you can use a comma and a space – there will need to be a comma.

For Google’s IPv4 addresses, your choices for 8.8.8.8 and 8.8.4.4.

For Google’s IPv6 addresses, your choices are 2001:4860:4860::8888 and 2001:4860:4860::8844.

Note: The ifconfig or ip addr will help you tell if you have IPv4 or IPv6.

When you’re done, be sure to click the save button to ensure your new settings take effect. Remember the screen and changes, should things go pear shaped. You can undo this easily enough.

This will, of course, work with any set of DNS servers out there. You can use it with other servers if you aren’t a fan of Google. This can serve as a general guideline for other servers, should you wish.

Closure:

Yup… There it is. You have another article. This time, it tells you how to change your DNS servers if you use Lubuntu. Again, it’ll work for other distros, but I’m only including pictures/vouching for it with Lubuntu.