The EFF's 'Let's Encrypt' Plans Wildcard Certificates For Subdomains

Long-time Slashdot reader jawtheshark shares an announcement from the EFF's free, automated, and open TLS certificate authority at LetsEncrypt.org: Let's Encrypt will begin issuing [free] wildcard certificates in January of 2018... A wildcard certificate can secure any number of subdomains of a base domain (e.g. *.example.com). This allows administrators to use a single certificate and key pair for a domain and all of its subdomains, which can make HTTPS deployment significantly easier.
58% of web traffic is now encrypted, Let's Encrypt reports, crediting in part the 47 million domains they've secured since December of 2015. "Our hope is that offering wildcards will help to accelerate the Web's progress towards 100% HTTPS," explains their web page, noting that they're announcing the wild card certificates now in conjunction with a request for donations to support their work.

Frosty pi... oh no..

[gibbsjoh]

Damn, I thought I was going to get first post AND call out a dupe, but damnit I was beaten!

1 out of 2 ain't bad.

"Haven't we seen story this before?"

-JG

Re:

[Anonymous Coward]

Been using wildcard certs for over a decade. Why this is news?

If they want to improve things change it so a company only needs to buy for one top level domain (mycompany.com) and any depth of subdomains from that can use the cert (mycompany.com, test.mycompany.com, env1.qa.mycompany.com, etc...).
 

Re:

[tepples]

i think the news is that you won't have to spend beaucoup bucks per year for such a certificate.

Re: Frosty pi... oh no..

[dougdonovan]

100% https...3 times...Internet, Intranet & Extranet supported via VPN...gotta keep the vendors happy.

Comment removed

[account_deleted]

Comment removed based on user account deletion

Re: SSL-certificates used to mean more than encryp

[Anonymous Coward]

Let's call such certificates EV certificates

Do you have any idea what you're talking about?!

[Anonymous Coward]

Sorry, I have to ask, are you just playing dumb in some failed attempt to be "funny" or "sarcastic", or are you really just ignorant about how these sorts of digital certs actually work?

Are you really unaware of the differences between Domain Validated Certificates [wikipedia.org] and Extended Validation Certificates [wikipedia.org]? Are you unaware of how they're obtained? Are you unaware of how modern browsers indicate the use of such certificates to the browser's user?

I really hope you're just trying to joke around, but failed miserabl

Re:

[tepples]

AC #54769865 probably believes that a web browser ought to be showing the same sort of interstitial before a cleartext HTTP site or an HTTPS site using a domain-validated certificate that it shows before an HTTPS site using a self-signed certificate. This interstitial would make it clear that the user is visiting the website of an entity other than an established business.

Re:

[toonces33]

The majority of browser users will click past any warnings about certificates without thinking about it. So I think you are correct.

Re:

[Antique Geekmeister]

I'm afraid that to the average user, there is no difference. The little "green" label or "locked" icon continues to indicate that the certificate is valid and the user has little reason, and not many resources, to verify that they are dealing with a validated but fraudulent, SSL certificate. Even automated tools that mirror content, such as for git repositories or software repositories, can be fooled by such certificates.

Re:

[Antique Geekmeister]

> The little "green" label or "locked" icon continues to indicate that the certificate is valid

I need to revise this. Some browsers provide additional indicators that a certificate has "extended validation". But the ordinary user simply does not care nor will they notice.

Re: Do you have any idea what you're talking about

[Zero__Kelvin]

The certificate guarantees that if you were trying to connect to fraud.com that you in fact connected to fraud.com and there is no man in the middle. It works as intended, you simply don't know what it does. No cert guarantees that once you connect to TrumpUniversity.biz Donnie won't screw you deeply.

Typosquatting

[tepples]

The certificate guarantees that if you were trying to connect to fraud.com that you in fact connected to fraud.com

Then what makes it clear to Bank of America account holders that "bankofarnerica.com" (that's ARNERICA) isn't the site they're looking for?

Re: Typosquatting

[Zero__Kelvin]

Certainly not standard DSL certs. You seem to think LetsEncrypt is doing something different than everyone else here other than providing free when others charge. They aren't. They are issuing non-EV certs that are just like paid for non-EV certs. I'm afraid nothing will protect an idiot from their own idiocy.

Re: Typosquatting

[Zero__Kelvin]

Fucking phone!!! Clearly that is supposed to say SSL not DSL.

Re:

[tepples]

You seem to think LetsEncrypt is doing something different than everyone else here other than providing free when others charge.

I'm aware of what a DV SSL certificate does and does not do. Others aren't. Or they are but want browsers to display a more conspicuous indication of lack of organization validation for certificates that are only DV, such as an interstitial.

Re: Typosquatting

[Zero__Kelvin]

Right. That is why you asked a question that made it clear that you didn't know. Off you go now ...

Re:

[fnj]

Then what makes it clear to Bank of America account holders that "bankofarnerica.com" (that's ARNERICA) isn't the site they're looking for?

Their eyeballs looking at the URL in the address bar, and their brain interpreting the text, makes it clear. Unfortunately, as in your example, it can be pretty subtle, and if UTF-8 URLs catch on, it becomes downright impossible, as UTF-8 has multiple code points which render indistinguishable from each other in various fonts.

Re:

[hawkinspeter]

For me, NoScript provides decent protection against that. If I've visited the correct site previously, then I would've white-listed JavaScript for that domain (and possibly a couple of their related domains). Then, if I visit the fraudulent domain, no JavaScript would run and the chances are that the site would look very different.

SSL certs are not primarily for identifying that you've visited the correct domain (as in the one that you think you were connecting to), but are to prevent man-in-the-middle at

Re:

[mi]

and Extended Validation Certificates

This is great, I had no idea, these existed... Thanks.

... failed miserably

The word "miserably" is overused.

Re:

[phantomfive]

Since the 1990s, the world has realized that actually, we want to have all our web traffic encrypted. Why? Because with the advent of wifi, launching a MITM attack is too easy. There are plenty of good reasons to encrypt traffic, which is why everyone uses ssh, not telnet.

"Trust" turned out to be not a big of problem as everyone feared. Most of the time when I go to Amazon.com, it really is the real Amazon. In fact, it's never not been the real Amazon. However, it still is a real problem, and once traffic

Good stuff

[DaMattster]

I just don't see why it has to wait until January of 2018 to implement.

Re: Good stuff

[Zero__Kelvin]

Because they have to implement the support and test it. I'm sure they would love to sprinkle magic faerie dust and it would just work, but that isn't how technology works ... unless you are Agile. Agile is magic faerie dust.

Re: Good stuff

[Zero__Kelvin]

That was the most stupid and uninformed post I have seen in quite some time. Checkout (clone) the certbot code and look at it. You will start to see why you look like such a fool right now to anyone with a clue.

Free certificates...

[__aaclcg7560]

I pay $15 per month for VPS web hosting at DreamHost [dreamhost.com] and get "Let's Encrypt" certificates for free on my domains and subdomains. Other options included self-signed (free) and Comodo (paid) certificates.

Re:

[i.r.id10t]

Seems you like over paying. linode.com has a similar vps for $5/mo. $10/mo doubles that...

Re:

[account_deleted]

Comment removed based on user account deletion

Re:

[spire3661]

Amazon Lightsail, and Digital Ocean both offer $5/mo tiers as well. (20 GB SSD, 512 MB RAM, 1 TB transfer).

Re:

[qubezz]

You can also go away, advertising. Anybody can get Let'sEncrypt certificates for free for their domain, that's the whole point.

Re:

[fluffernutter]

Obviously these certs aren't as secure as other certs, and the purpose of the short expiry is to put a hard limit on any exposure to 90 days. I agree it's annoying and this is why I haven't made an attempt to use these certs yet. Although being able to create a wildcard cert is interesting indeed. At least I will only need to have one cert reissued every 90 days instead of five.

What the fuck are you talking about?!?!?!

[Anonymous Coward]

Although being able to create a wildcard cert is interesting indeed. At least I will only need to have one cert reissued every 90 days instead of five.

LOL! It's very clear that you have never actually used Let's Encrypt. It supports the subject alt name extension so that one cert can be used for multiple hosts.

Fuck, just look at Slashdot's cert, if you're browsing this site using HTTPS. The Let's Encrypt provided cert I'm seeing used here has a CN of slashdot.org, but it also supports these names:

apache.sla

Re:

[account_deleted]

Comment removed based on user account deletion

Re:

[fluffernutter]

Feel free to go through my comments over the past years and feel free to point out anywhere I claimed I was an expert in anything, so I'm not going to apologize nor do your criticisms really register with me.

Anyhow, the idiotic comments aside; that's interesting I will have to look into it. Perhaps this will help me with the $50 cert I currently have registered. I was under the mistaken assumption that wildcards were the only way to make it accept various hostnames. I assumed that it was limited in thi

Re:

[fluffernutter]

I was ready to spend $50 CDN on a cert. Couldn't find anything multiple-hostname for that.

Re:

[bill_mcgonigle]

At least I will only need to have one cert reissued every 90 days instead of five.

There are certainly some cluster-type cases where a wildcard will be handy, but in general people have used wildcard certs to make key management easier. Now that we have cron jobs/an API to do key management, I am more inclined to have multiple certs running all over the place, to isolate a break. CAA and DANE records integrated with Let's Encrypt will smooth over the potential downsides of everybody having tons of certs.

Re:

[fluffernutter]

Obviously you don't work in a corporate environment that is totally isolated from the internet. Both outgoing and incoming traffic not allowed unless absolutely necessary. Certainly no cron jobs allowed to pull things down when they like.

Re:

[FrankHaynes]

*I* have such systems deployed and would love to read your solution for this problem of isolated/insular nets that require Internet access for authentication.

Use sneakernet or an internal CA

[tepples]

You have two options on an air-gapped network:

A. Every two months, sneakernet CSRs to a machine that isn't air-gapped, run the ACME DNS challenge on that machine, and sneakernet the certificates back to the air-gapped network. The one thing you can't do if both the server and the client are air-gapped is OCSP.

B. Set up an internal certificate authority, and deploy its root certificate throughout the internal network. This may fail in Android 7, which distrusts user-installed root certificates unless each ap

Re:

[fuzzyfuzzyfungus]

Isn't an isolated network that you have exclusive control over pretty much an ideal case for using your own root?

CAs are a necessary evil when you expect to deal with 3rd parties, because they've managed to get themselves trusted by a variety of vendors and you haven't; but if it's all your stuff, you can set it to trust your root and call it a day.

Re: 90 day certificates

[guruevi]

In those cases any outside certificates are useless since you can't verify trust. You only need to have an Internal CA system for those sorts of setup.

Re:

[chispito]

Yes, where I have used them I have automated the renewal process, but still what the fuck is the point of wasting my time with that shit?

I'm trying to figure out how an automated process wastes your time. Can you explain?

Re: 90 day certificates

[Brockmire]

A fucking cron job is tedious? Seriously? It's copy and paste for fuck sakes. The only problem I have is reusing the certs in programs that you paste the cert and key instead of pointing to the files. One of these days I'll figure out the sed syntax and do that through automatic post processing script when renewing certs automatically. It'll take 5 minutes to google the sed syntax and test it out. Maybe 10 and reused on many servers. Tedious? Umm, no.

Re:

[hawkinspeter]

I prefer the break-early model of LetsEncrypt. Set up your test system with free LetsEncrypt certs and then test the cron script (one-liner) for renewing. Also, the certbot client has a dry-run feature so you can check what it's going to do if you do want to do proper testing.

With long expiry dates, you'll never get around to automating renewal and then you'll probably forget all about it and/or move to a different job and not care. Someone is then left with a ticking time-bomb of embarrassment for a doma

Re:

[bill_mcgonigle]

Letsencrypt will continue to lack any credibility until they abandon this retarded policy.

Dude, you are lacking credibility here if you don't understand why long-lived certs are a problem for security. For small businesses, the main reason not to do a short cert, given letsencrypt's cron jobs, is for a wildcard cert, which is expensive, and now that is being solved. For personal websites, wildcards are generally not used. Enterprises have the option of syncing their client and server certs, for authentic

Re:

[Anonymous Coward]

The reason for short-lived certificates is that certificate revocation does not work and is broken beyond repair.

Re: 90 day certificates

[guruevi]

300ms? That's for the network latency on some CAs alone. Checking revocation, if at all possible since many CAs simply don't have the infrastructure up and running, can take several seconds to verify the entire chain which often contains 3-5 chained certificates, if the CA doesn't respond, this could easily be a full minute before your certificates have been verified without even a proper response on the condition.

If Google were concerned about latency, they could simply do the lookups on their end and cac

Comment removed

[account_deleted]

Comment removed based on user account deletion

Re: When LE announced, but no wildcard...

[Anonymous Coward]

Cool story bro. What else do you predict will come to pass? Will you be my oracle?

Good idea, but...

[ErichTheRed]

LetsEncrypt is a good idea because it makes certificates accessible to a wider range of users. I've been doing systems engineering work for quite a while, but haven't really concentrated on web stuff. When I got involved with a public-facing web project at work lately, I noticed there really is a lot to the TLS system and certificates once you get beyond internally-trusted certificates. Most places did the legwork for certificate acquisition years ago, but setting something up from scratch requires that you

Re:

[tepples]

Why not just fucking use cacert.org?

Last I heard is that they didn't have the finances to do the sort of third-party auditing that the CA/Browser Forum requires.

Re:

[tepples]

Let's Encrypt offers certificates for as long as your automatic renewal cron job continues to run, provided that your domain also remains paid-up.

100% HTTPS??

[spire3661]

There is no fucking need for EVERYONE to be running HTTPS.

Re:

[Anonymous Coward]

Yes, yes everyone should be running HTTPS. There is NO reason for any internet connected device to be communicating using HTTP. HTTP is a primary target for "enrichment", redirection and other payload manipulations. HTTPS is the only way to go.

There is no reason not to use HTTPS. The days of low CPU devices are LONG gone. Recent technology improvements such as QUIC and HTTP2 (over tls) are encrypted by default. QUIC eliminates the round trip time for TLS setups- it's easily as network efficient as H

Re:

[spire3661]

ITs not about CPU usage. Certs add a huge administration overhead, and need to be maintained. A static webpage with some contact info does not need HTTPS. Further, i REALLY dont like the idea that is starting to shape up that if you dont have a cert, you shouldnt be on the web. HTTPS is a tool for SOME jobs, not all HTTP, everywhere. That is just plain retarded. I shouldnt have to get permission from a third party to run a dead simple webpage. There ABSOLUTELY 100% are reasons to not need HTTPS. You are a f

Re:

[account_deleted]

Comment removed based on user account deletion

Re:

[hawkinspeter]

In general, that kind of page doesn't need to be encrypted.

However, encrypting connections to websites makes it harder for bad guys to sabotage someone's connection to the website and injecting malware/ads etc. A free and easy to get and use SSL cert provides some protection for very little cost, hence the push to get as much of the web encrypted as possible.

There's also an issue where people might be trying to analyse traffic and it could be of some advantage for them to know when you're visiting "secr

Re:

[spire3661]

I dont believe in an internet where you need blessing from a third party to participate.100% HTTPS as the system is currently implemented is outright folly. Let me easily self-sign and ill be more on-board.

Re:

[hawkinspeter]

Self-signing is easy enough but has security issues. The client has no way to determine who did the signing - it could be the website owner or it could be a man-in-the-middle.

Using HTTPS everywhere is more about protecting client computers (and their data) rather than needing a third-party's blessing. LetsEncrypt is a major step in lowering the barrier to let everyone run HTTPS easily and for free. It's designed to be easy to automate, so all you have to do is set up your web server to allow the specific

Re:

[spire3661]

Again, i dont believe in a web where you need a third party to vett you to participate. Its an INCREDIBLY ugly road. I like the HTTPS initiative, but i hate hate hate people pushing 100% HTTPS. We are all trained in absolutes and exceptions here, dont you think 100% HTTPS could have some nasty downsides?

Re:

[hawkinspeter]

Nope. A simple analogy would be to use the postal system. Imagine that HTTP is like people sending each other postcards. Anyone can read them whilst in transit and also alter them. HTTPS would be equivalent to everyone sending letters in sealed envelopes (maybe with old-time wax seals on them). Now I understand that you don't want to be funding the BIG envelope corps, but here's an initiative that provides free envelopes (although they bio-degrade after 90 days which some people think is awkward).

The bigg

Re:

[account_deleted]

Comment removed based on user account deletion

Re:

[hawkinspeter]

I've seen MITM attacks at several wifi hotspots. Airports are a particular favourite place for people to set up a rogue hotspot and grab loads of credentials. To be honest, the safest way to use hotspots is to encrypt everything by using a VPN, but at least HTTPS will give you some warning (invalid certificates) if you do connect to a rogue hotspot without using a VPN.

I'm not understanding the "vetting" issue with LetsEncrypt - they don't do anything except determine that you have control of the domain. I

Re:

[Anonymous Coward]

According to the snowden documents, because it used strictly http and avoided https, for a long time three letter agencies manipulated the slashdot.org website when it was viewed by network administrators at large corporations when they were on their break. They planted exploits in the traffic to infiltrate the admin and their network. They were specifically targeting I.T. administrators in that campaign in order to slip exploits into products and services used by Americans and increase spying capability. T