Spam blacklisting is out of control
blog.roastidio.us
blog.roastidio.us
You can do everything possible, have a perfectly clean IP, have a good amount of outbound email traffic, only send transactional email, etc. Still, there will be edge cases where email does not go through. AT&T email servers would constantly blacklist me and not respond to requests to remove me, gmail/yahoo/outlook would silently put emails in the spam folder, and companies using email firewall products would blacklist me, with an IT Dept too inept to fix it.
The solution was to pay a small fee and proxy all outbound email through a transactional SMTP sender, like Postmark or Mailgun. It's easy to do, with one line of code in Postfix. You can be selective, and only proxy emails sent to certain troublesome domains. If you try an email provider and it's not working out, it's one line of code to change to another provider.
This allows me to still manage nearly all aspects of hosting my email server and control my email data, while not dealing with deliverability issues. I use Postmark and I have not dealt with a deliverability issue in two years.
Meanwhile I've seen people miss events because $BIG_MAIL_PROVIDER decided the invitation was spam, I've seen recruitment go wrong because a CV from an excellent candidate was blocked on its way to the designated email address for applications, and countless other examples where bad spam blocking was throwing baby out with bathwater.
The problem is that it's much easier to send email than it is to receive it. This puts the onus of spam filtering on the recipient. I'm sad that HashCash or some other PoW scheme was never adopted as a way to force rate limiting of mailers.
But since they have no accountability, their incentive is to eliminate anything potentially hostile regardless of the collateral damage from false positives. What are their users going to do? Take their business to another big provider that will probably be doing the same, and in the process lose access to a bunch of online services they like because their accounts on those services are tied to the old address? Move to a small provider, which is being forced out of the network by the giants so can't deliver mail reliably?
We are rapidly heading for a world where email is no longer a reliable communications medium for genuine messages between friends, family, work colleagues, etc. The giants are killing email through death by 1,000 cuts instead of a shotgun to the face but the end result is the same.
Even a national division of Microsoft got hands on my email and decided to sign me up for invites to developer events (I know I hadn't signed up for it, because they didn't even use my name in the mails, which they certainly would have known if they had gotten it through any legitimate source).
I assume it's bullshit KPIs, or signup schemes. Some poor seller at Microsoft or Walmart who's ranked according to how many people sign up to the opt-out company promotions, and buys harvested addresses to juice their numbers.
I even got some very obvious fraud spam ("Your [expensive product] is waiting for you!", trying to get you to sign up as a "product tester" only paying $$$ per month) with full DKIM signatures from a site hosted at a Danish hosting provider, and Danish business info. You'd think this sort of thing gets shut down quickly, but nope. The only times I've tried to address this sort of thing through the proper channels (domain registrar's abuse accounts, etc.), I've only gotten markedly more spam.
At least DKIM signatures makes it easy to autodelete mail. No need to even go into the spam folder, I'm happy to forbid Walmart from communicating with me by mail forever.
Probably 95% of the spam I receive is automatically filtered to a spam folder, with a negligible false positive rate. Every now and then some new pattern does make it through but once I've flagged a few instances the rest start getting classified as spam automatically.
Gmail's spam filtering is very generous to DKIM-signed mail - it has a tendency to let it trough even if I've flagged exactly that sender before (as I did with walmart). The one on my own domain has received too little spam to tell how good the filter is.
cryptocurrencies are not in any way going to solve the problem of people centralizing their MX all onto office365 and gsuite.
edit see this current front-page submission about how Bitcoin fails to provide this despite being centralized in exactly the same way as the decentralized SMTP: https://news.ycombinator.com/item?id=30224637
Then it should be only the SMTP server operators who have to handle the crypocurrency side of things, so email users never have to worry about it.
As you say, acquiring and using the cryptocurrency would be about as technically difficult as what the SMTP server operators are already doing, and there are a variety of 3rd party brokers they can choose if they want to simplify things and not run a node themselves.
The system I'm imagining is one where each newly registered domain has to put up a cryptocurrency bond if the registrant wants to send email from it. Existing domains would be grandfathered in (having already built up their reputation) and new domains would have their bonds burned if some N-of-M stakeholders agreed that they were sending (DKIM-signed) spam.
Choosing those stakeholders would be controversial, but hopefully no less controversial than the system we have today where Google can use the threat of a Gmail blacklist to make every SMTP server in the world follow its wishes. Ideally some of the stakeholders would be non-profits like the ISRG and Mozilla Foundation.
They could provide goods or services for it the same way people acquire dollars without going through a 3rd party broker.
This is why for any store over the size of a mom and pop operation in a neighborhood, you can't just tell the owner who you know by name to put it on your account, and instead for credit purchases you use a credit card company which acts as an intermediary and smooths problems over on both sides, and refuses to work with stores and people that are untrustworthy.
This is why there are mailing list (mass email) services and why mail servers allow them. They keep their customers working within the accepted bounds (they ensure removal works and fire clients that abuse), and this allows mass email for accepted reasons while still being able to come down hard in random exploited servers/accounts.
This is why big email services are very selective about what servers they talk to. I work at an ISP where our main outbound mail servers are on IPs that we try not to change because they've got decades of reputation attached. Even so, we recently brought up two new servers for email forwards, and shifted a small percentage of our mail queue traffic to them and ramped it up over a couple weeks, and that seemed to work "warming them up" to the likes of Gmail and yahoo, etc. It used to be there were lists of mail operators you could be part of and you could use reputation within that to get them to be lenient with you when you started. These days it's all so centralized in a few very large players that they really likely just talk to each other.
Yes, warming up a "fresh" IP definely works. If you suddenly send thousands of mails from a new server - sure, you'll be labeled as Spam. If you slowly increase the volume over time, have a good domain and recipients that interact with the email, things should be fine. GMail has quite helpful guidelines [1].
This has absolutely nothing to do with someone not wanting to talk to someone else. It has everything to do with some third party having the power to decide whether the other two may communicate.
These days it's all so centralized in a few very large players that they really likely just talk to each other.
And thus the single most important method of remote communication in the world today, the method that is frequently akin to root access to our online lives, became subject to arbitrary monitoring and interference by huge, powerful organisations with their own interests and negligible regulatory oversight, legal safeguards or accountability to anyone but their shareholders.
Do you really not see why this is a problem? You talk about the internet matching the real world, but in the real world we've had laws against monitoring and interference with things like postal mail and telephone calls for a very long time almost everywhere.
They have that power specifically people people outsource vetting of whether someone's worth dealing with. I'll repeat, this is all about reputation, and how without reputation you're subject to every anonymous person's abuse. That's not an issue when the total number of people you're dealing with is manageable. It is when the number of people you expect to have to deal with is the population of a small country or the entire world.
> And thus the single most important method of remote communication in the world today, the method that is frequently akin to root access to our online lives, became subject to arbitrary monitoring and interference by huge, powerful organisations with their own interests and negligible regulatory oversight, legal safeguards or accountability to anyone but their shareholders.
Only for those that opted into that system. Gmail does not dictate what mail servers accept your mail, they dictate whether they themselves accept your mail. They just happen to have a userbase in hundreds of millions billions, so a large percent of the people you might want to contact use it. You can blame Gmail all you want, but they're just doing what their users want, which is reliable email without much spam, and that's how they've achieved it. You're not going to get far telling people when you send stuff to them they're required to accept it. The system only worked the way you wanted it when it was small and offenders weren't as anonymous and social punishments worked against them. When everyone's mostly anonymous, that no longer works.
> Do you really not see why this is a problem?
It is a problem, but it's a problem of people choosing to use them. And there's been ways to keep people from reading your emails for decades at this point with GPG, and if you can't trust the other side to buy into that, or to not use a web based email client, then there's nothing you can do about communication with those people anyway.
Should they be monitoring email? No. Would we be better with laws preventing that? Yes. Is that really the same issue as them being so large that by dictating who they will talk with and how they are effectively dictating rules for running a public mail server? No. Separate issues.
People outsource that vetting? What choice did those people have?
If there is a small cabal of tech giants dominating email distribution, and a user's choices are between some of those giants that are essentially as bad as each other and some smaller competitors that might be better in terms of delivering incoming mail reliably but are frozen out by the cabal in terms of delivering outgoing mail reliably instead, there is no meaningful choice or consent about having this filtering done to their incoming mail. It is a situation imposed upon them by the tech firms.
That's not true. When you use Gmail you don't opt-in to a setting to "automatically send smaller non-commercial providers to the SPAM folder". Gmail is pretty bad at SPAM handling you still receive lots of commercial advertisement from "reputable" sources, you just don't see the non-scammy non-phishy email you were expecting.
In this specific instance, Gmail is not the worst out there. They're very protective but from what i heard you can get allowlisted in a matter of weeks of harassing their tech support. Meanwhile, Microsoft is a lost cause if you want to get your mail through to an Outlook mailbox...
Yet, if that store knows you, and wants to extend credit to you, they can. A difficulty with email is that it seems to be increasingly difficult for the recipients to have much control over the email they receive. I know of numerous instances where people are unable to get specific emails they want or need to receive to not be spam-foldered, or to be received at all. They may have no understanding of why this is the case, and even if they do, may have no ability to change it. The systems are often opaque and unchangeable, and involve organization- or provider-level choices before emails ever reach their accounts.
It is decentralized and has built-in mechanisms to fight spam. Also all messages are end-to-end encrypted.
Despite the name it has nothing to do with bitcoin or cryptocurrency.
I don't accept that. Nothing about defending against those threats requires the kind of guilt-by-association spam blocking we have been discussing.
There will always be bad people in the world but that doesn't mean we allow some random big business to tell us where we may go, what we may do or who we may talk to because it is in that business's interests and (coincidentally or otherwise) it might make us safer.
First, I will send emails to test accounts I have set up with major email providers (gmail, yahoo, outlook, etc). I used to do this often, but they have all been going to the inbox on my last few tests, so I haven't done this recently.
Second, is feedback I get from recipients. When I send an email and ask for a reply, I'll get a reply. I also used to hear a few times a month, "sorry for the delay. I found your email in my spam folder!". This has gone away.
It's usually worthwhile to remind businesses that when they use "free" services like privacy-paid outlook.com and Gmail, they'll get what they pay for, and if their communications really matter, they should find proper email providers.
We have had on and off deliverability issues for years (AT&T and Comcast being the worst).
As head of IT it fell to me to post whitelisting requests and try to get mail delivering again. I decided after awhile that this really isn't my job, and made a suggestion to the CEO which he took to heart:
There is another solution besides changing IPs, using a paid sender, or filling out whitelist requests into the void: Get your legal department involved. We have repeatedly been taken off various public and private blacklists by having lawyers do their job. Once we went this path, it was like magic. Same day responses from those companies, and we haven't been on any blacklists for a couple of years.
I also think that the likelihood of success in sending legal threats to somebody that demand they accept your SMTP traffic will not stand up in court, if you ever escalated it that far.
As somebody who runs postfix MX on the receiving side of things, I can guarantee you that the day I receive a legal threat from some unknown third party with which I don't have a pre-existing business/contract relationship, demanding that I accept their email, is the day that I blacklist their entire organization and tell them "okay, I'll await service of your statement of claim".
You actually think that the best answer to a network engineering problem is to make legal threats at third party ISPs? Companies with which you don't have a signed service order contract and/or master services agreement?
You say you're a mid sized company. I think you're running a huge legal risk of angering a Comcast or AT&T size entity that has much deeper pockets and legal resources than you. The day that one of those giants calls you out on your bluff is going to be very expensive.
On an ISP-to-ISP relationship level, this is not how you solve SMTP flow traffic problems. I can tell you that if I went to a NANOG conference representing my AS and proudly told other people "oh yeah, we've started sending threats from our lawyers to $OTHERISP1 and $OTHERISP2 because they won't take our mail traffic", that I would quickly be treated as a pariah.
It’s pretty easy to envision a situation where a lawyer sends a quite friendly and factual email to a company, that is literally identical to the one the IT head would have sent, but because it’s coming from a lawyer the recipient uses completely different internal routing to process the request. So someone actually takes the request seriously.
Seems both plausible and a reasonable thing to do for a company large enough to have a legal department.
People pay attention to lawyer letters. You’ve pretty much confirmed as much by noting that letters from a lawyer are so concerning to you that the mere mention of one makes you assume it’s a threat.
If you got a letter from an attorney asking for something nicely, and it was a reasonable request you would automatically reply “SUE ME” just on principle? What’s the principle?
The implication is that if you do not do whatever is demanded in the letter, the next step will be the client of said lawyer escalating the situation to paying their lawyer to actually sue you.
Of course there’s a lot of garbage out there too.
Your main premise seems to be that the recipient should take all this very personally. But it’s not personal, these are businesses, discussing a business related topic.
The VAST majority of business to business communications that involve attorneys don’t go anywhere near an actual filed lawsuit.
Like I said in my original reply here, we are talking about sending threats to third-party isps, with which the originator of the smtp traffic has no existing business or contractual relationship.
I would think that if you were the ISP owner blocking our mail to your customer, a single phone call from the customer or from us would convince you that our IPs were not a spam threat. That's certainly the way I prefer to deal with anything. But when dealing with something as monolithic and totally deaf as AT&T, lawyers were the only way.
It doesn’t mean they are thinking to sue you. Not even close. When you actually get served, that’s when they’re thinking about it, because actually serving you is a few bits of paper and requires a few brain cells. Sending you what amounts to a form letter? No brain cells required.
Make it a legal and business decision instead of a technical one. And find an audience that knows the practical benefits of quick resolutions to reasonable requests.
And I say this as a person that would likely respond in a similar punitive and/or principled way to what you described upthread. Which is exactly why I leave it to the lawyers.
I'd go further and say that choosing to unblock some IP addresses out of a block you don't like is always going to be a business decision. The whole rationale for blocking in the first place was that a sender is untrustworthy.
Untrustworthy senders generally don't have a team of lawyers, and if they do (and they're actually sending spam) you can show what spam they sent and keep them blocked.
All the legal letter does is force someone who's otherwise too busy to look at this particular case for the 5 seconds it takes to determine that it's not a threat. Sad to say that being head of IT for a company for a decade doesn't get you the same level of respect; it might if they even bothered to open your email, but there's something about postal and legal letterhead.
2. Possibly run all your outbound smtp through a trusted third party service that you pay for such relay. Leaves a bad taste in my mouth but that's where we are at in 2022.
3. Be absolutely certain that your own smtp, spf, dkim, dmarc configuration is flawless and you've never been a source of spam.
That's a nice idea. In fact, it's what my businesses have done for years. I have personally met several of the senior staff at the service I use and I know that they are both technically excellent and very serious about preventing abuse on their network.
We haven't been able to deliver mails to customers at certain mail services for years. 100% blackholed every time. And it's the same usual suspects that others have mentioned in this very discussion.
The problem is that the hosting service we use has many customers. Occasionally one of those customers makes a mistake and one of their systems gets compromised for a short time, until they or the hosting service detects the unusual traffic and intervenes. And that's enough to get someone's IP block on a blacklist it will evidently never leave.
This is of course absolutely no different to any hosting service like AWS or Azure, except that no-one is going to blanket block all servers from an organisation of that scale even though the exact same problems happen there too.
It's not laziness; the intention of collateral blocklisting, as with UCEPROTECT L2 and L3, is punitive. It's to incentivize the sending MSP to remove their spammer (or move them to address-space where they can be blocked without causing collateral damage).
Unfortunately the people it punishes are the legitimate users of both systems who only wanted the system to do its job and let them communicate. The bad actors will just move on and abuse another system instead.
"The system" you are referring to consists of a bunch of private networks. Your opinion about what the job of those networks is may not coincide with the opinions of the operators of those networks.
Email is not a public service, and there is no entitlement to send whatever "vital business communications" you like to anyone you want. It's not even reasonable to require a postmaster to state what their rejection policy is; that would just tell spammers what they have to do to evade your blocks.
If email doesn't work for your business, then switch to another channel, such as huge billboards or whatever. Ranting about blocklists isn't going to help, people have been doing that for two decades.
I take it you've never run a mailserver?
Yes, in that situation it's laziness, but on the part of your ISP. ISPs already spend huge amounts of time and money dealing with spam, hacking attempts, phishing attacks, etc. If your ISP is irresponsible and isn't doing their job keeping those things from leaving their network then your ISP is gong to find their entire IP space blocked and that's 100% reasonable. Why should we ever accept traffic from an ISP that refuses to keep their corner of the network clean when it's just going to cause problems for us and our users?
That's the situation for every ISP on the internet. Keep your users in line, keep trash off your network or else no one is going to accept traffic from you. You could call it laziness, but it simply isn't worth it. We'd rather spend our time cleaning up abuse on our own network and working with ISPs who are doing their job than dealing with the problems we get from bad actors.
If you own an IP surrounded by a bunch of spammers and it's giving you trouble step 1 should be to contact your ISP and tell them to get their shit in order or you'll take your business to an ISP who does their job. Step two is to switch to a new ISP if they don't. No one has the right to force us to accept traffic from anyone else. It's every ISPs responsibility to make sure the traffic leaving their network isn't more trouble than it's worth. Good ISPs are rewarded because users will give them their business and stay and bad ISPs are punished because users will drop their service when they see they are blocked.
The goal isn't to punish the poor sucker who signed on with an irresponsible host, but to cut down on the number of bad ISPs on the internet and the amount of work we have to deal with coming from them.
There is a reason that collective punishment is considered immoral by civilised cultures. It hurts the innocent and often fails to achieve its original goal anyway.
It's how the internet stays functional. If 0.001% of legitimate mail has to go undelivered in order to prevent overwhelming amounts of spam/attacks from an irresponsible network that's an acceptable loss to most people in our civilized culture. Bad actors have been ostracized from communities for as long communities have existed. If you want reliable service, choose a responsible ISP. Blacklists have enabled the internet and email to remain useful for most users most of the time. Without them, email wouldn't be usable. If you have a better solution for spam, the whole world would love to hear it.
[citation needed]
If 0.001% of legitimate mail has to go undelivered in order to prevent overwhelming amounts of spam/attacks from an irresponsible network that's an acceptable loss to most people in our civilized culture.
1. It's way more than 0.001%. Like, several orders of magnitude more.
2. What overwhelming amounts of spam/attacks? Those of us using traditional mail systems with traditional spam filtering seem to be doing OK not getting overwhelmed by incoming spam without the kind of "help" you advocate.
3. You don't get to decide what's acceptable to everyone else. Except that apparently you've decided you do, which is why regulation is needed to remove that ability from service providers who can't do their jobs properly and hurt others as a result.
If you want reliable service, choose a responsible ISP.
This is a poor argument. Any ISP that accepts significant numbers of customers will occasionally have a customer who is either malicious or operating with imperfect security allowing someone else who is malicious to exploit them. The decent ones will identify the problem and block it reasonably quickly, but there is plenty of evidence that they can still be blacklisted and it can still be difficult or impossible to get removed from those blacklists again after the problem is fixed.
The kind of policy you advocate punishes small ISPs just for being ISPs. I invite you to apply the same policy fairly and neutrally to larger organisations such as the major mail forwarding services and cloud hosts as well and see how long you survive in this industry.
Or maybe the rest of us should apply the same policy to organisations that do what you advocate. If they won't deliver mail reliably, we won't forward any mail to them at all, so their mail service becomes useless. Except of course it's some of the biggest mail services that do this, so just like no-one's going to block incoming mail from AWS or MailChimp, no-one's going to block outgoing mail to Google or Microsoft.
If you have a better solution for spam, the whole world would love to hear it.
Block actual spam sources and provide a reasonable method for removing blocks that are no longer necessary. Don't carpet bomb whole chunks of the Internet just because there are a few bad actors around. Don't fire and forget. It's really not difficult and plenty of small organisations operate just fine on this basis every day.
Nearly 85% of all emails are spam. source: https://dataprot.net/statistics/spam-statistics/
At times that number has been even higher with over 90% of all messages sent over the internet being spam. If 90% of all the messages in your inbox were spam how long would you continue to use it? Email systems can't bear the costs that spam forces on them. Even with tools like blacklisting which you think shouldn't exist that cost is measured in tens of billions annually. source: https://www.aeaweb.org/articles?id=10.1257/jep.26.3.87
If not for the ability to filter common sources of spam, email would never have survived as a viable means of communication.
> It's way more than 0.001%. Like, several orders of magnitude more.
Whatever the actual number, it's clearly acceptable to us because blocking irresponsible networks is standard practice. We depend on it.
> What overwhelming amounts of spa
Again, 80-90% of all mail is spam, costing billions. If you're able to run a mail system without blacklists that's great for you, but it clearly doesn't work for everyone.
> You don't get to decide what's acceptable to everyone else.
That's the beauty of the internet. I don't have the power to force an abusive network to do their job and prevent spam from leaving their network and that abusive network can't force me to accept mail from them. No one can force anyone to do anything. All we have is a loose set of standards and expectations and it's up to each network to decide what to accept or not based on how well those standards and expectations are followed.
> Any ISP that accepts significant numbers of customers will occasionally have a customer who is either malicious or operating with imperfect security allowing someone else who is malicious to exploit them.
A responsible ISP identifies those users and prevents them continuing to cause problems. If they refuse to do that their reputation suffers and they will get blocked. If they do their job too slowly or too poorly they will be blocked. Is it possible for a responsible ISP to end up on blacklists? Yes, it is, and there are blacklists that don't maintain their lists well. That's fine too because no one is forced to use them. It's still the case that every network has the choice of what blacklists they will or won't use and how they use them. They can whitelist blacklisted IPs they decide to trust and they can use blacklists to greylist instead of block.
> The kind of policy you advocate punishes small ISPs just for being ISPs
Nope. Even small very ISPs can staff their internet abuse departments adequately and implement anti-spam technologies to prevent their IP space from becoming a safe haven for hackers and spammers. If they choose not to do that they will and should be blocked.
> I invite you to apply the same policy fairly and neutrally to larger organisations such as the major mail forwarding services and cloud hosts as well and see how long you survive in this industry.
I'll agree that there are problems when certain services (either cloud providers or mail providers like Gmail) become "too big to blacklist". We've had that problem with AOL and we have it now with Google. Personally, I'd prefer to hold them to the same standards as everyone else, but the problem of the largest players throwing their weight around giving them unfair advantages exists in every industry and until someone comes up with a solution for it, we're all just stuck playing along.
> Block actual spam sources
If your ISP is a safe haven for spammers and hackers their IP space is the spam source.
> provide a reasonable method for removing blocks that are no longer necessary.
So your alternative to blacklists is just more blacklists that are run better? I think everyone who depends on blacklists would like those blacklists to be better at detecting spam sources and better at clearing unnecessary listings. The good news is that badly run blacklists don't tend to get widely adopted because they cause more trouble for ISPs than they are worth.
If some network won't accept your mail and you're convinced that your ISP is acting responsibility and that it's the blacklist that's wrong, you can have the person you're trying to reach contact their ISP to get your mail server whitelisted. If an ISP sees that a blacklist they use is catching too many messages that it shouldn't they'll adjust their thresholds or stop using that list.
It's not a perfect system, but it's the best one we have.
Did you actually read that, and the sources it cites, before posting it? If you had you might have noticed that it's full of the worst kind of junk stats. Several of the sources cited, the ones that supposedly support your arguments here, don't even say what the piece you linked claims. They literally have completely different numbers. Not that it matters since there is no indication of methodology used and the exact figures are clearly impossible for anyone to measure accurately. Some of the other "sources" are just links to organisation home pages without identifying any specific research or analysis at all.
If 90% of all the messages in your inbox were spam how long would you continue to use it?
As someone old enough to remember the time when that was actually the case, obviously we managed. But this is distorting the argument again because you are implying a false dichotomy where the alternative to overly aggressive blacklisting policies such as you advocate is all of the spam reaching our inboxes. Clearly that is not realistic as less aggressive defences are still highly effective and have consistently been so for a long time.
No one can force anyone to do anything.
Really? Then where can I sign up for a mail service that will reliably deliver both my incoming and outgoing legitimate messages without undue monitoring or interference with my own business? I contend that possibly no such service currently exists.
Personally, I'd prefer to hold them to the same standards as everyone else, but the problem of the largest players throwing their weight around giving them unfair advantages exists in every industry and until someone comes up with a solution for it, we're all just stuck playing along.
Which is exactly why some of us are in favour of statutory regulation to compel anyone participating in such an important technological ecosystem to be a good citizen.
So your alternative to blacklists is just more blacklists that are run better?
I don't believe I have ever suggested anywhere in this discussion that using blacklists to block traffic from proven spam sources was unfair or inappropriate. My objection, which seems to be in line with the submitted article, is to big mail services that think spraying fire into a crowd of 250 indefinitely because there was once one bad person there is a reasonable response to the problem. There is huge collateral damage being caused and the defenders of this policy are trying to sweep it under the carpet and use highly debatable arguments of necessity to justify their damaging policies.
This is not the best system we have. That's the point being made here.
I'm also old enough to remember that and we managed by blocking huge amounts of IP space. Even massively popular services like AOL have blocked the IP space of entire ISPs or entire countries from being able to send them email. Eventually spam filtering improved, things like SMTP auth, DKIM etc caught on and wide range blocking could be scaled back somewhat, but I doubt it will ever go away entirely.
> Really? Then where can I sign up for a mail service that will reliably deliver both my incoming and outgoing legitimate messages without undue monitoring or interference with my own business?
Use your own servers and you can do whatever you want. Again, you can't force others to accept email from your mail servers, but you can choose to accept or reject whatever you want from others. No one can stop you from sending mail from one mail server you own to another mail server you own.
> Which is exactly why some of us are in favour of statutory regulation to compel anyone participating in such an important technological ecosystem to be a good citizen.
You can't really regulate the internet. If you could enforce regulations on a global network made up of discrete but interconnected networks we could just make spam, phishing, and hacking illegal on the internet, enforce that law/regulation and there would be zero need for blacklists. Because laws and regulations don't work on the internet we instead have to come up with blacklists, filtering technology, and other tricks to keep the internet even semi-functional.
> My objection, which seems to be in line with the submitted article, is to big mail services that think spraying fire into a crowd of 250 indefinitely because there was once one bad person there is a reasonable response to the problem
It's the only one that works. I've seen with my own eyes ISPs who didn't care enough to invest at all in abuse handling, but were forced to because of being blacklisted and in order to keep their customers they had to clean up their network, pay attention to abuse notices, participate in feedback loops, and slowly rebuild and maintain their reputation as responsible network operators.
If you limit blocks to individual IP addresses than spammers just cycle IP addresses. ISPs that ignore anything sent to their abuse@ address (if they even have one) never have any pressure to invest in preventing spam and can just keep accepting money from spammers and hackers and give them new IPs whenever they need to.
IPv6 makes the problem much much worse since a single spammer would get a huge amount of IPs to burn through before they have to bother their ISP about it. Blacklists themselves could become so massive and cumbersome that restricting larger and larger ranges may be the only option.
We absolutely can regulate the Internet on this kind of issue. We don't have to regulate everywhere in the world to make a big improvement, just businesses above a certain size that operate a commercial email service. If our governments can effectively lean on social networks enough that they add warnings to potentially misleading comments about science, they can lean on email services to do better with this problem. They only difference is that there is an obvious and unambiguous way the mail services could do a better job.
And again, just to be crystal clear, I am not arguing for giving real spammers a free pass. I am only arguing for credible, realistic measures to try to avoid the huge numbers of false positives we get from mail filtering today.
The only reason we don't is because unlike email, it's the sender who pays not the receiver. Telecoms do monitor and block outbound international calls if the connection times are excessive, if they occur at unusual hours, or if they going to certain "blacklisted" countries where phone fraud is common. They do it because hackers will break into a business's PBX and use it to place a bunch of international calls and the business suddenly gets a massive phone bill. They call their phone company about the changes, the phone company waves the changes (once) but that leaves the phone company on the hook for them. When false positives happen, the business has to call into the phone company and explain the calls were legit and they will be whitelisted and similar outbound calls will be allowed going forward.
I wouldn't oppose using regulation in the US against US based mail services if it meant forcing them to do a better job preventing spam from leaving their networks, but I'd be hesitant to support legislation forcing them to accept more spam. Maybe the largest ones could be pressured to invest more money in handling the influx of spam after they accept it, but I'm guessing there would be costs to consumers such as long delays in delivery, or "free" services like Gmail suddenly requiring payment or closing their services for good. At the ISP I work for now we stopped hosting our own mail servers and outsourced email services to a third party because spam filtering was too expensive and time consuming, and now we're looking at possibly no longer offering an email product at all and telling all of our customers to migrate to services like gmail and yahoo. Killing our email service today would eliminate a lot of problems in terms of help desk calls, phishing attacks, and spam problems. Make it too much harder for people to provide email service and there may only be giant providers left.
Whatever the other guy thinks about it being "necessary" or whatever, there is not commonly a way for a user to whitelist a service. And services providing email dont normally take that sort of signal into account, either.
Once you are operating a large system that is used by many people, you become a public utility - furthermore, at that scale we can generally find where you live and come lock you up. This kind of thing is 100% regulatable.
Either let users choose what mail they receive, or implement regulation forcing compliance. If that doesnt happen, and you snub my lawyer like the irresponsible mega corp you probably are, guess thats one more reason for me to polish off my shotgun and takeout the dickwads running the megadoom corp.
The majority of all email has been spam, for more than a decade.
> You don't get to decide what's acceptable to everyone else
If you operate a mailserver, you do actually get to decide what kind of stuff you are willing to accept, and from who. "Everybody else" does not get an automatic right to inject data into my computer.
> The kind of policy you advocate punishes small ISPs
Not at all (perhaps you meant MSPs?) Blocklisting Google was a reasonable policy, at one time. Blocklisting MailChimp is perfectly reasonable now.
> Block actual spam sources
Of course, good plan. Unless the sender's ISP is in the habit of moving spammers from one address to another, so they can evade blocks. Then you have to block the ISP, or eat their spam.
Postmasters can't inspect every inbound spam!
Or maybe your policy?
> There is a reason that collective punishment is considered immoral by civilised cultures
Rejecting email submissions isn't punishment, collective or otherwise. It's something you have to do if you run a mailserver. In the same way, I'm not punishing trespassers if I secure my front-door with a deadlock.
Currently, countless people essentially commit countless abuses for free because the actor is hidden behind a machine or a process. But somewhere it's a humans decision to institute an abusive protocol, and it seems pretty fair fo me to make that human accountable for their action. Not just email but all kinds of things.
You are probably merely a dick but still a legal dick if you wantonly block email for yourself. But the second you are responsible for even one other person's correspondence reaching them, I say you should be legally culpable for any failure to deliver.
I question whether you or what other percentage of the commenters in this thread represent any specific ASN with its own IP space that it cares about keeping clean, and have bgp relationships with other ISPs.
Or whether they're actually end users only.
Have you actually encountered this problem as a service provider in the past and implemented solutions to it, or are you just sharing your opinion as a possibly-frustrated end user of email?
What I didn't do was ignore the problem or think there was no problem. Things still got blocked elsewhere but I (personally or my company) didn't do it.
I recognized that handling someone elses correspondence is a non-trivial responsibility. Yes the job is hard, and so you either decide that is your whole job, or you outsource it to someone whos whole job it is. You don't just do a poor job because it's hard and not your all day every day job.
So you think an MSP's advertised policy should be "We guarantee that anything sent to you will be delivered, including spam"? That no MSP should provide spam-filtering, at risk of legal culpability?
If that's not what you mean, then presumably you are requiring all MSPs to block only spam, and to deliver all legitimate email. But that is impossible, because nobody has figured out a way of reliably distinguishing spam from ham.
If I received such a letter from your legal department, I would laugh, and reply: "I am awaiting your writ." If I got the note from a recipient or their postmaster, I'd be much more inclined to try to help, but I never try to negotiate with lawyers brandishing threats.
If they say "We drop 10% of messages at random, and you knowingly choose to take that, then that is just a stupid arrangement you should never agree to, but they aren't doing something unexpected.
I decline to believe you are as stupid as that remark.
Good faith best effort is perfectly reasonable. Not knowingly and intentionaly discarding mail is all that's required.
If you're the mailman, you do not have to garantee that you will never lose a single letter in a car accident.
But you can certainly garantee, absolutely, that you never go through the bag and throw away all the spam and sometimes mistake a legit letter from a lawyer for lawyer spam.
You can ceetainly garantee, absolutely, that you never apply utterly thoughtless rules like "we got these scam letters from Nigeria so now we just throw away anything from Nigeria."
You should absolutely be responsible for other peoples stuff that is in your hands while it is in your hands. It's not yours to dispose of, even when part of your explicit job is to filter. If that sounds onerous, that's why you get paid money for the responsibility. If it's too hard to bear this responsibility properly, then you have no business doing that job. Do the job right or don't do the job. There is nothing unreasonable about those two choices. It is not at all required to do the job, but poorly or carelessly.
No.
Suppose the MSP uses bayesian filtering? How would one go about advertising a policy that depends on bayesian filtering? You'd have to publish the contents of your filter table. The only people who could benefit from that would be spammers, who could use the data to customise their spam.
In general, telling the world what your filtering policies are is just going to cause spammers to try to sidestep your policies. The only policy that you maybe ought to publish would be along the lines of "We filter out spam; that sometimes results in false positives. Sorry."
You likely wouldn't do that - just get a template version that gets reused, just like you pay once for a contract / t&c you reuse with multiple parties.
> You actually think that the best answer to a network engineering problem is to make legal threats at third party ISPs?
It's not a "network engineering problem" if administrators and managers are the ones making the decisions to provide no reasonable recourse for a ban, on top of actively ignoring or denying legitimate requests for removal. If the third party hasn't broken any rules, then the only resort is getting attorneys involved, because there are literally no other options if the provider refuses to act in good faith.
> Companies with which you don't have a signed service order contract and/or master services agreement?
If you are providing an email service and your customers are not receiving the emails they're expecting, all because you refused to acknowledge a removal request, then it shouldn't matter if you have a relationship with the third party. You've failed your own customers, and for no legitimate, logical or even conceivable reason, other than to assert some personal dominance over another group of engineers that you see as "lesser."
None of this is an engineering problem. It's an asshole problem.
And on this I concur with you, because if the sender of the email is actually sending spam and has ended up on some smtp-receiving deny lists for well founded reasons, they are indeed failing their customers. Failing them through their own lack of procedure and policy, such as not having a click to remove link in the mail, and not removing addresses/putting addresses into a scrub-before-sending list.
If that is the case, they have nobody to blame but themselves for that and should not be sending legal demands to other peoples' ISPs such as at&t or Comcast.
But the receiving MSP can't acknowledge the removal request, in the scenario reported by the author. They are not the ones operating the blocklist. That's UCEPROTECT, not Comcast or whoever.
> USING OUR BLACKLISTS, YOU ARE AT YOUR OWN RISK. WE WILL NOT BE HELD RESPONSIBLE FOR ANY DAMAGE OR PROFIT OR LOSSES. SHOULD THIS BE UNACCEPTABLE TO YOU, YOU ARE NOT ALLOWED TO USE OUR BLACKLISTS.
I don't know whether that is true; but I do know that it is usually used as a DNSBL - a DNS lookup for an IP address, that answers whether that address is or is not in the list. In general, mail providers do not "ingest" entire blocklists.
> responsible for maintaining their own list in a way that complies with international laws [etc.]
Actually, anyone can publish a list of anything; unless that publication amounts to a contract (statement of purpose, assertion of fitness for purpose), then I'm not aware of any international "law" that says you can't put anything you like in a publicly-acccessible list. And anyway, there's no contract without an exchange of considerations - baiscally, you have to cough-up if you want to assert a contract.
> even offers a very blatant disclaimer on their site
Have you ever read a FOSS licence? UCEPROTECT are simply stating that as a free user, you can't hold them responsible for the accuracy of their lists; in the same way that FOSS authors disclaim responsibility for fitness-for-purpose.
Completely false. They ingest lists[1] into their own local daemon. There is not a major mail provider on the planet who is querying a third party DNSBL every time an email comes in. It's also more than a little ridiculous that you admitted to not knowing if something were true, but then decided to confidently (and incorrectly) explain it anyway, rather than taking ten seconds to look it up.
> Actually, anyone can publish a list of anything; unless that publication amounts to a contract (statement of purpose, assertion of fitness for purpose), then I'm not aware of any international "law" that says you can't put anything you like in a publicly-acccessible list.
Both wrong and irrelevant. I explained how Comcast is responsible for maintaining their own blacklist, which is NOT public, and as an ISP they are absolutely under all sorts of legal regulations, in addition to having service agreements with end users and other network providers. Yet, you responded with a "point" about how UCEPROTECT (a completely different company) is allowed to post whatever they want in a text file. Do you seriously not see the disconnect between what people are actually saying and how you interpret them? Because it's beyond frustrating.
> Have you ever read a FOSS licence? UCEPROTECT are simply stating that as a free user, you can't hold them responsible for the accuracy of their lists
Entirely irrelevant. The UCEPROTECT disclaimer is not for the end user -- it's for the ISPs. If someone sues Comcast, they're not trying to hold UCEPROTECT responsible for creating a text file, they're trying to hold Comcast responsible for acting in bad faith by not allowing reasonable recourse for removal from Comcast's copy of the text file. So, the end user in this situation is not a "free user of UCEPROTECT," but instead a paid user of a network provider with a service agreement in place.
If the remote mailserver is explicitly bouncing your mails, it's not a network engineering problem.
If you're able to complete a TCP 3 way handshake, it's not a network engineering problem.
If you're being prevented from completing a TCP 3 way handshake because of a filter list on the MX server, it's not a network engineering problem. Admittedly, you can't tell that one without asking a network engineer (on the remote network) to validate that for you, or asking the application owner to advise.
Contrary to popular advice (thanks MSFT), you probably shouldn't contact your network administrator. We probably have priv 15, but we dont necessarily have root.
The number of people using anything other than google, apple, microsoft or yahoo for personal email these days is vanishingly small; and of the remainder, almost all of them are on major broadband providers. If you were some local ISP blocking us, that wouldn't be worth the trouble of sending a legal letter. I'd just tell that particular customer to contact you and ask why you weren't delivering our mail.
This only leaves other corporations that provide their own mail services to people who work there, and if those people want to get their mail they can call IT.
Somebody created a github containing domains he thought should be blocked: https://github.com/chadmayfield/my-pihole-blocklists
A year later I bought a domain that had expired under my country TLD. It turns out that domain for some reason was previously added to that list.
Now, as you can see, the man behind that Github repo has decided to archive that repo and therefore make it read only.
As you can tell from both pull requests and issues on that repo, people has asked him to remote legitimate domains (e.g. *.urbandictionary.com). But those calls remain unanswered. It is fruitless to contact the author.
So this random dude causes real problems for legitimate business and individuals and we should just accept it?
Obviously he doesn't act on friendly geek requests. It seems lawyering up would be the only recourse in this situation. I find it analogous to somebody standing on a soap box in a village and announcing: "Don't trust James. Don't trust Mary either. There's problems with Charles" and James, Mary, and Charles have no way of stopping his libel.
I don't have the funds for legal action, but it is obviously wrong that he can announce "these domains are bad" and offer no way of fixing mistakes. He should take down the repo, but oh that sweet sweet Github karma probably discourages him from doing so.
You're speaking of some random domain-list on github - that isn't even maintained? Taken from someone's private pihole?
Anyone using a list of that kind to block is simply incompetent. Even as part of a scoring system, it's pretty silly. Before adding a blocklist, a postmaster needs to familiarize herself with the list's policies. Are list entries aged-out? How quickly? Do they use spamtraps, or user-reports? Or is it just the whim of the list-maintainer? Do they block individual addresses, whole domains, or entire allocations?
> So this random dude causes real problems for legitimate business and individuals and we should just accept it?
So you're having problems sending mail to a domain where the postmaster cares more about rejecting spam than she does about receiving legitimate email. That's a matter for your recipient to take up with their MSP. And if the recipient wants to receive mail from small-time domains, they need to accept that they're going to receive some spam as well; but maybe they need to switch to an MSP that only rejects on strong evidence.
My point is that it's your recipient's choice to use an MSP that blocks using some crazy list they found on github.
Some postmasters will block everything from selected countries; at one time I would block everything from Romania, because none of my users had correspondents in Romania, and email from Romania at that time was 100% spam. But I wasn't providing service to the public. I knew all my users.
Different MTAs have different users, and different patterns of abusive email. So if you want to use a custom blocklist, make your own, based on your own incoming spam (and then you can honour removal requests yourself). Otherwise use a public blocklist, based on multiple spamtraps in multiple ISPs.
So yes, you should just accept it. You don't have a right to have mail delivered by any MSP you send to; they're private organisations or individuals, and they're entitled to determine what their own policies are. In the world of email, nobody is entitled to protection from the foolishness of others.
"Tortious interference is a common law tort allowing a claim for damages against a defendant who wrongfully interferes with the plaintiff's contractual or business relationships"
A blacklist is not directly interfering with your business. They just provide a list that contains IPs, your IP, and say we have seen spam traffic from it in the last x hours. The mail receiver, who trusts and uses the list might be interfering, but it is his right to pick and choose who he is accepting email from. Same right you have to pick and choose who you let into your bar, apartment, club, house, ... what so ever.
Safe your 500 bucks per hour. A lawyer will not do anything
But I do know the "enemy" of my "enemy" is not necessarily my "friend".
I've seen the nightmare of the next Internet. Whether it's micropayments for digital stamps, or slowly refilling quotas, or hard hierarchical controls, it's more power to central authority.
I know I'll never win this argument.
The Internet will keep fracturing and normal citizens will get more frustrated. There will be a change. But just like the hatred of social media, lurking underneath it is an insatiable thirst for power.
I'm not saying this situation is _good_. It results in massive consolidation and gatekeeping by Tier 1s for mail. The problem is that it's much easier to send mail over SMTP than receive it, so the cost is on the receiver to filter mail. That means the receiver has every incentive to be overly defensive. I think a hashcash/PoW mechanism or a cryptocurrency deposit mechanism could be a great way to fight some of these problems. Pay a crypto deposit and receive a tunnel to send mail through. If you violate any spam policies, your deposit is lost and the tunnel is closed. Or charge cryptocurrency (quadratically?) per mail sent on this tunnel.
TOS are gonna TOS, but if the small print reserves the right to block spam, or traffic from spammers, then proof of not being a spammer, or at least proof of a legitimate use, might create an obligation ATT has to its users to not block those addresses, even if there's no obligation to the 3rd party making the request.
The knowledge affects ATT's obligations to others, not the requestor.
Idk where that gets you in a practical sense, but its at least a line of reasoning that's beyond "ATT can block whatever it wants".
That’s what tortuous interference is, getting in the way of third parties doing business. You could argue about technical details around the server but a judge will properly understand the social constructs at play.
What's the particular law that makes those curators of blacklists pay attention to your company's lawyers? Do you have example text of those legal requests?
I have a contract with the customer which involves interacting with them by email. Customer uses EmailCompany. EmailCompany unfairly blocks me from emailing customer. EmailCompany is interfering with the contract. Both myself and the customer have a reasonable expectation that we should be able to carry out our contract by emailing back and forth, so EmailCompany is obstructing the contract.
Any of this sort of "advocacy" only works if the customer actually cares about receiving your emails though.
In my experience, the customer will just switch email providers when I tell them we're blocked. I had an email provider just the other day who wasn't accepting my emails. I tried to get them to unblock us, but to no avail. I told my customer and he went and registered a new email elsewhere.
Obviously, legitimate use-cases exist for such services, provided that they are operated by faithful people/entities with some level of credibility.
This causes a human with authority to actually look at your case and likely go "yeah, this is silly, I've taken care of it". Lawyer is very happy because something that looked like a lot of expensive work just went away.
An email from legal@companyA.com to legal@companyB.com will be read.
You need to involve legal.com.
Or you send certified mail with return signature requested. Snail mail to the rescue?
ETA: We refer to it as "6mins and a stamp".
Or just write and snail-mail a letter from yourself. People respond surprisingly well to requests and offers in a personal/non-bulk letters.
The easiest way to close a ticket is a quick wontfix. The easiest way to avoid a possible six-figure minimum legal issue is to spend 15 minutes figuring out what they want and giving it to them. The law is almost entirely irrelevant; the cost-benefit analysis doesn't change enough when comparing a frivolous vs meritorious lawsuit.
Excellent idea, thank you.
1. I don't know how the lawyers worded the letters. Obviously it's not illegal for a host to block inbound mail. But clearly the legal team made someone look more closely at our particular case. If you're not doing anything nefarious, it's easy for someone mid-level to clear you off a blacklist if they want to. The company I work for isn't a household name outside a few states, but it's large and respectable enough that getting a cursory glance from someone with the power to say "take them off the blacklist" is probably enough.
2. We've had these IPs for > 5 years. Our reverse DNS, DKIM, SPF are all in order. The mail comes from the same IPs as the company's main website/app services. The company uses Constant Contact for marketing emails, so our outbound server only sends things like receipts for purchases, service agreements, confirmations of renewal/cancellation, verification codes for login (password reset / new signup / new device), and intra-company mail which sometimes goes to employees' personal addresses. Being able to firmly state that we never send marketing off this server, and we regularly assess exim logs and we are sure we've never been hijacked has usually gotten good results, even without lawyers.
3. When we first got these IPs, around 2015, there were more problems. I don't know who they belonged to before. For the first year it was whack-a-mole with the blacklists. But truthfully, I've never had a problem personally getting us taken off any of the public blacklists shown on mxtoolbox. It's the private blacklists of mail providers like AT&T, Comcast and AOL that caused repeated issues. There's no public way to check if you're on these, and as far as I can tell their removal forms are black holes. The removal form for AT&T looks like it hasn't been updated since the '90s, and I've never received even an automated response from using it.
4. We did, for awhile, advise customers to get other email addresses. Unfortunately, the customers with an @comcast or @att account are also probably the least tech-savvy. It's just not feasible to have customers calling their local retail outlet and having a manager spend an hour trying to walk them through why they didn't get their signup verification code, and three days later this bubbles up to corporate and finally to me. And it isn't a good look for us as a company to say "sorry, Comcast doesn't like us for some reason."
5. For awhile we had customers write to Comcast and AT&T asking why they were bouncing important emails from us. Overall this had basically no effect. The customers would get back automated emails telling them how to configure Outlook. However, one such customer was a lawyer who I ended up corresponding with personally (through my gmail account, since he couldn't get my emails), who got such a bug up his butt about it that he seemed to get AT&T to unblock us, which only lasted a month or so. This was where I got the idea to approach the CEO about unleashing the legal department.
6. Variations of "Don't waste $500/hr. on a lawyer". IMHO lawyers are pretty terrible to have against you, and pretty awesome to have on your side. I'm glad I work for a company whose philosophy is to keep as much of our infrastructure as possible in-house, and which also refuses to pony up something that feels like a ransom, even if it would be more expedient. I personally don't have $500/hr. to pay a lawyer, and it would be a much worse uphill battle for a sole proprietor or small company these days trying to manage their own email server than it was ten years ago, or is now even with the backing of a larger company. But it's important to us that we don't get pushed around by big providers, and retain as much autonomy as possible in our business dealings with customers. That's part of the philosophy I brought to the company, it's why we run dedicated hardware, and I'm the one who pushed to spend the money to have the lawyers get involved. If more mid-sized companies took this approach, and if other IT leads pushed for running their own metal and their own mailservers (granted, yeah, it's an endless pain in the ass), then big email providers would be forced to actually take things on a case by case basis the way they did ten years ago, rather than block blacklisting whole /24s with the assumption that every company and individual will eventually cave and be forced into their monopoly. So to whoever says that decline is inevitable, save your money, lawyers suck, etc, without offering any positive solutions: you're welcome.
Edit; to be clear I'm not saying this results from some legislation, although you could make that case. Just that scale has many benefits and the principle of regulatory burden obtains!
Providers like Postmark and Mailgun let you forward and send email through them. Essentially, you are paying these companies to deal with IP address reputation management and ensure high deliverability.
Are you though? Aren't you giving your outbound emails to Postmark or Mailgun in that case?
https://postmarkapp.com/eu-privacy#security-and-privacy
It's a decision I am fine with. Others may not be.
... cheapest is $10/month. Not paying that for my meager email traffic. Are there other options for personal domains?
I'd rather pay a little and get unmodified emails. Guess they all cater to marketers not personal use.
Both my personal domain and rsync.net are on a distinct subnet, but that subnet is smaller than a /24 and someone on a different subnet has, apparently, behaved badly.
Enter "abusix" ...
One of my engineers had an enlightening webchat with one of their engineers where we were shown the "offending" IP and it was explained that they have no ability to distinguish subnets (and no interest in doing so). So if you're not wasting an entire /24 (we only need ~10 IPs at this location) you're in danger of this misclassification.
We were also informed that our normal, business communications with paying customers should have unsubscribe notices appended to them. Which is to say, you're a paying customer of a service and we send you some kind of alert or critical announcement ... and it should have an unsubscribe link.
Unbelievable.
You should have an unsubscribe link. You should also have your business address and identify yourself.
Even if it's not required by the letter of the law, you should add it.
As an example: Amazon automatically opted me into an "alert" when a wishlist I viewed had a new viewer. Since it's an "alert" and a "business communication" it has no unsubscribe. This is spam - this is an ad hidden as a notification.
there seems to be some conflict of interest here.
Not all blocklist providers are the same.
thou every security-for-profit scheme suffers from afromentioned conflict of interest.
Once the unsubscribe is activated -- and it's supposed to be very easy to activate -- then it's permanent. There's no "un-unsubscribe", "oops I clicked it again", "some other service glitched and clicked it for me".
Further, there's a distinction made between "commercial" and "transactional" messages in both law and etiquette. The unsubscribe link is expected in commercial messages, not transactional ones.
Abusix didn't know what they were talking about.
Most of the junk that gets through my spam filter are transactional or other "mandatory" messages intended for someone who fat fingered their email address. If those senders don't want to be marked as spam, they need to provide a way for me to make the messages stop.
Unsubscribing from transactional emails eventually causes the following support conversation: "Hi, uhh, rsync? Yeah, so, I'm having trouble logging in to my account and we really really need our backups, our intern just nuked a database. Yeah, it's uhh... cto@company.com. What do you mean my account's not active? ... ... Why didn't you just tell me my card expired? Well yeah, of course I unsubscribed, but I still wanted to know my account was being shut down!"
There's a scale of headaches happening here. At one end of the scale we have "nuisance", as in, "I'm getting too much email, or I have a stupid email address, or I don't know how to filter messages from reputable senders", and at the other end we have "job-ending cockup", as in, "I'm just now finding out that a critical part of our disaster recovery plan hasn't been working for a long time because somebody somewhere was inconvenienced by a notification, and I'm finding this out now because today happens to be the day we really need that disaster recovery plan".
Pushing the needle away from the nuisance end moves it closer to the disaster end.
It is not fair to the rest of us to be inundated with endless spam just so some screwup can be kept from doing something stupid.
Now someone will likely reply shifting the definition of what “spam” is to include Rsync’s critical service emails, and now the term spam is so wide as to be meaningless.
At that point it’s on you to manage your own spam filter if you truly feel “your critical backup service is down” is spam. I haven’t been inundated with endless spam for about a decade.
Abusix don’t know what they are talking about, and basically all services that let you manage your email notifications still send through critical “your service is about to be turned off because your card details failed” emails regardless of how many checkboxes you disable — and for good reason.
I solicited them to store my data.
Did you solicit every nag and advert Amazon sends you?
That you bought something from someone does not mean your email inbox is now free game.
I get emails for some dude's Chevy when it needs servicing. I can't unsubscribe. I am stuck getting emails about a car I have never owned from some dealer in Pittsburg. I need an opt out that lets me communocate "hey, some dumbass fatfingered his email, stop spamming me."
I'm honestly curious.
I think what you're picking up on, is that some folks don't differentiate between commercial email filtering services, and personal spam filters.
There's conflation of email 'I don't care about and don't want', bulk UCE, shifty list operators with shifty 40 page terms, etc.
1. Some senders intentionally send ads and actually useful notifications from the same address, making filtering more difficult
2. Better yet, some senders constantly shift their address so they are almost unblockable without going to arbitrary keyword blocking
I would like to live in a world where instead of me doing stupid amounts of work to not get a flood of spam.. people just don't spam.
I am lucky - I can literally throw money and tech at this problem. Most people are not lucky. I would much rather spam be elimited as a whole.
At least based on my understanding of rsync-the-company and rsync-the-hn-commenter.
What would that even look like ?
You're a paying customer of a service - they charge you every month - and you use that service ~daily ... and then you unsubscribe to emails ...
So then what ?
We just keep taking your money and when the service fails or there is an outage or critical notification we ... just don't send it ?
It's still nice to have granular subscription though even for rare things you think 95% of users may like to hear about e.g. I've been using a similar service since 2015 and I have 0 interest in receiving their downtime or scheduled maintenance notifications as I don't care enough to take a special action for a failed sync or two in the first place so... I don't opt to receive them and I appreciate that option. I don't get the invoices emailed so I haven't had to think about it one way or the other there.
What are they supposed to do in response to an outage notification, apart from wait for things to recover on their own?
- The service-based economy means that entities (individuals and businesses) have numerous relationships. For the typical individual the number of password-based accounts crossed the 100 threshold years ago, at a doubling rate of every 2--3 years.
- Responsibilities can be transferred. The person who signed up for your service 5 years ago may no longer be at the company.
- List purging is a Real Thing. A few years back I'd worked for an organisation that had ... numerous relationships ... with individuals and corporations. These received regular email messages. Nominally, requested. Included amongst these was a major Wall Street financial firm whose implosion years earlier hit lead news and headlines worldwide. Despite not existing for years, there remained hundreds if not thousands of addresses being sent email on a regular ongoing basis.
- Mail can be forwarded. It's quite possible that you're sending mail to one address that is is being forwarded, manually or automatically, to others. This raises issues in unsubscribe requests, but might at the least be an opportunity to reach out to your customer to clarify the situation.
- I don't know if revisiting email contact approval on a regular basis (say once every year or two) is yet a recommended practice, but I'd strongly suggest that it be so.
Your hat may be less blisteringly white than you presume.
We have that. You can change contact info, set owner/technical/emergency contacts, alert thresholds, etc.
But unsubscribe means something totally different:
When I click on unsubscribe I want it to be the end of all communications. Period.
In this case, that makes no sense. Ceasing communications for all purposes implies service cancellation and, in our case, service cancellation implies a human interaction confirming data destruction.
How would we confirm data destruction for your implied cancellation if no further contact is permitted ?
You, and the blacklist operators, have become so jaded by the abuse you've suffered that you've forgotten that legitimate, paid services exist. I'm sorry.
People should have every right to unsubscribe themselves from that, and thus should have some sort of feedback loop attached to the email being sent (to the detriment of your bottom line I fully understand and sympathize with).
If this was indeed a marketing email, then I don't think it matters if it comes from a "legitimate, paid service" - the receiver still should be able to choose if they want to receive sales related emails wouldn't you agree?
I agree.
We have a flag for such a thing and set that flag when people ask us to. They ask us in a nice email exchange between human beings. We're very responsive to this since they are our paying customers.
That's the big disconnect here: it's inconceivable to many people (including abusix, et. al) that healthy, straightforward interactions like this occur in 2022.
In their mind there are nothing but robots and newsletter subscribers forever locked in an arms race.
I'm sorry that is the case.
You are wayyy to smart not too see the abusive asymmetrical theater of that scheme.
So it's not enough to allow them to opt-out with an e-mail exchange or an unsubscribe link, you must allow the user to opt out when you initially gather the contact information.
If that wasn't done, it's illegal (and unethical) to send marketing e-mails, even from a paid service.
All emails should of course contain enough information to make it clear who the message is from, why the message is being sent, and who it was sent to.
But there is no point in adding unsubscribe links to messages and notifications that are essential to the service.
I mean, what are you going to do if the user accidentally clicks "unsubscribe", and then a payment doesn't go through? Should you just cancel their account without informing them? That's absurd.
I'd be really pissed if eg. my backups were deleted because I accidentally unsubscribed from emails from a cloud service provider.
Agreed. rsync alluded to it below as well.
'unsubscribe'... from what? If I just bought something from service ABC, and I get an email from ABC saying "you just bought foo from us"... what would an 'unsubscribe' even mean? "Do not ever email me about this purchase again?" "Do not ever email me about future purchases?"
Please send me the order, just don't send me the PDF invoice :)
My recourse is to just flag them as spam in gmail.
The company won’t talk to me, and the sender sends a lovely message, but no contact information.
Wtf is wrong with putting contact information in the unsubscribe link, or reach out productively on request? Why would you presume somebody clicks it by accident vs. the much more likely case of it being a legitimate request? Are you afraid they really want to cancel your service? Or are you afraid you can't send spam under the guise of transactional messages? Or worse, listen to customers about how best to alert them? Truly ridiculous!
By law, depending where you are, a unsubscribe link has to be instant. So there can not be an intermediate screen asking for confirmation or showing contact information. Well, you could show contact information, but then the unsubscribe (of in this case service critical mails, thus the service itself) had already happened.
It is neither dire nor prevents clarification, nor prescribes a specific experience related to unsubscribe links. It's odd to hear the only options are between receiving messages and having service terminated. That sounds pretty user-hostile tbh.
[1] https://www.ftc.gov/tips-advice/business-center/guidance/can...
That implies a level of manual effort on the part of the recipient that's unreasonable. I have no relationship with these companies. They did not verify the email address before starting to send a stream of supposedly transactional messages to it. They should be happy that I'm willing to click unsubscribe when available, because the alternative is to set up a mark-as-spam filtering rule that'll hopefully tank their sender reputation.
Writing them via a contact form begging to be removed is not an option.
In my case users are sending estimates and invoices and monthly statements to their clients and while fake invoices may be a spammer thing those clients know who's sending them an invoice, and why and what for, so an 'Unsubscribe" link would be completely out of context because they are not subscribed to any email list.
I've had the same domain name for over 20 years now and none of my users have ever used my apps to send spam. And as spam and email volume go my server isn't even close to sending out a lot of email.
When I set up a new email server last year, with a new IP address, I had to go through the process of getting white listed. All the big email service providers have ways to do that. Google made it very easy. They gave you a unique string to add it to your DNS records and that's it. Microsoft is so convoluted I've still not gotten anywhere with them. Comcast and others had a few hoops and ladders but nothing that got me stuck.
Personally, while it's a bit of a PITA to setup and manage an email server, it's been worth it.
I used "Mail-in-a-Box". It's pretty easy to set one up with that. It has a built-in DNS server and that's a really great thing to have for managing several domain names and as many email addresses as you want. I've setup email accounts for family and friends as well as throwaways for my wife, who signs up for everything she sees on the internet.
I can move the IP address of my email server to the top of the list in my Mac's System Preferences for DNS and start testing new domain names and changes to the DNS immediately. I don't have to wait for those to propagate to whatever my access provider is using.
So I have 3 servers. An Email/DNS server, a database server, and a website/webapp server running on DigitalOcean's "Droplets". It's a bit of work for a small shop but it's much easier to manage once it's setup and I don't have to worry about any 3rd party service selling out or going under or changing their API to something entirely different. All of which has happened to me in the past.
But even that does not help: some people just do not understand how to manage their email: they will consider as “spam” any email they receive and they cannot act on it immediately. Like storage quota reached, cc expired, or similar. There is some strange feeling that if they ignore the email and mark it as spam that the problem will magically disappear.
We have so many cases like that: and they then contact us with questions like “why my service is not working” etc.
I bet Netflix has same issues…
If they are an established customer, it is legal to do that in all jurisdictions, even under GDPR.
It’s much better than having a customer mark it as spam, as some people will definitely do if they can’t see how to unsubscribe.
In some jurisdictions there is information that businesses are legally required to provide to their customers in a permanent form and email is the conventional (and potentially the only) way of satisfying that requirement.
IMHO, it is not helpful for anyone to have a system where recipients may not understand this and may treat that mail as spam, yet businesses are compelled to send it anyway.
If I don't like an email I never unsub. I add it to the spam filter.
I had one email address that for some reason often ended up being mistakenly used for other people, and ended up enormously frustrated by the usual combinations of no unsubscribe links, no-reply senders, and notes that any changes should be made by logging in. In the worst case, the US Department of Education, for several years, sent the address private personal and financial details about a complete stranger at a university with student loans. There was no way for me to unsubscribe, and no monitored address to reply to: in their view, they were important emails for the recipient, and they were, but I was not the intended recipient at all.
You absolutely should. The amount of junk I get because someone else signed up for something and fat fingered their email address is ridiculous. "Mandatory communication" with a company I've never dealt with gets flagged as spam.
But it's not a good reason for adding unsubscribe links unilaterally to all email communications.
Remember, unsub links are machine-automatable; Gmail at least offers to follow any embedded unsubscribe links for you if you mark a message as spam. (Which, with hotkeys enabled, is one accidental keypress away.)
So consider the extreme case: what if the user fat-fingers an unsubscribe (without realizing) to their local electric company's e-invoices, which is what they've been relying on to prod them to log onto the site and pay the bill?
If it's clear that "bills you need to react to or your power will be shut off" shouldn't have an unsubscribe link, then clearly there's some sort of line that must be drawn somewhere.
(Note, I'm not arguing against the use of "Manage your Mail Preferences" links in these cases — the kind that act as magic sign-in links and take you directly to a page on which you can un-check a "mail me about X" checkbox. It makes sense to include those. I'm just arguing specifically against unilaterally including "Unsubscribe" links — the kind where following the link unsubscribes you with no further confirmation needed.)
To name a specific example of this problem, I want Gulf Power of Florida to stop sending exactly the kind of email you speak of. Bills. Nastygrams when the person falls behind on the bills. Unwanted power saving tips. Calling the company and sending them postal mail has not helped. It all gets marked as spam these days. If they had an unsubscribe button, it wouldn't.
If the email is so damn important, they can go back to sending postal mail to the service address when someone unsubscribes.
They should have confirmed their user controlled the email address, too, but why not go with both?
And this is hardly confined to Gulf Power. Verizon, Spectrum, countless banks...
Unsubscribe is the alternative to the spam button. If you don't provide it, you are asking to be marked as spam, no matter how important you think the email is.
it happened to me - someone charged a bunch of stuff to my cc and then registered my email at thousands of sites to bury the email receipts (it didn't work since I have simple filters for that sort of thing) but it has been impossible to unsubscribe from all the junk. livemail's bulk optout was roughly 50% effective. the dark patterns around optout are outrageous and it's worse when you have to use google translate just to find it.
But in the cases where there is authentication, isn't it enough (in most cases) to reset the password and change the email to something disposable?
Of course that's not really practical for the case where you get subscribe-bombed, but maybe for the general case it is, no?
1. Forget I had a monthly power bill for a couple of months.
2. Ignore the e-bill that gets sent to my bank bill payment service - ebills have been a thing for almost two decades. I worked on some of the early implementations.
3. Ignore the physical snail mail warnings for a couple of months.
I actually unsubscribed from my provider's invoices. That's because I have activated direct debit from my bank account so they're always paid, and I can view my past invoices on the website.
However you make a good point. I'd say the one thing where it doesn't make sense to have an "unsubscribe" at all is on "bill unpaid" emails.
In the early days of the internet I was a teenager in a prank war with someone so I signed them up for all kinds of spam and also free trial magazine subscriptions.
I never flew with them but somehow they still sent ads disguised as reassuring messages every other week.
Support constantly denied help, since I was never a customer. Only a GDPR complaint solved it.
But once the email is confirmed, I think it's totally fair for a company like rsync to say 'if you're a paying customer of this service, then we need to send you certain information to fulfill our obligations in the contract, if you truly don't like it cancel your account and take your business elsewhere.'
You might not be able to go this route if your customers have some expectations about how your email is handled, in which case this recommendation is here for anyone else that might read this and need another thing to try.
As such the reputation score of a subnet is the reputation of the entity advertising itself as publicly controlling and maintaining that network, not the reputation of individual sub entities inside that subnet (which is known only to the controlling entity). If that entity is constantly allowing bad actors onto their block then that block is considered poor reputation.
Nonsense. SMTP is not internet routing, and there is no reason at all why some "reputation" system should be constrained by the same /24 limit that the global routing table ended up with.
I find it telling that these "reputation" outfits tend to serve up this sort of poor excuse to justify their businesses; whereas they're often plain old protection rackets. Like the one described in the article clearly is. They take money to solve the problem they created!
These domain based lists are also a thing but you need to have a pretty sizable volume to become trusted in those lists as anyone can register a domain in seconds, none of the above will have gained trust on them.
The whitelist payment option for arbitrary hosts (which Abusix doesn't offer) is a workaround to create a publicly registered identity with some stake to tie the reputation to. I mean it can be other things to, scammy rackets exist too, but you can't just claim it's a poor excuse without offering a better answer that solves the same problem.
Speaking of spammers are the ones that created the problem because people started using individual network block lists before companies started aggregating and mediating them. You're welcome to make your own community replacement solution but the problem is the replacement has to be reliable enough other mail administrators want to use it not just nice for the occasional legitimate guy that wants to run email out of his house and be trusted by default. Also it's a lot of work to mediate.
Here's the actual chain of events in question:
- You recently switched ISPs and that meant you moved to a new IP block.
- The IP block in question is owned by Hurricane Internet and unfortunately contains a host which persistently sends out a lot of junk (https://lookup.abusix.com/search?q=216.218.240.46)
- Without going into massive details on how our infra works, but when we have the most serious level of listing (e.g. hitting our most secure traps as in this case), we treat the same /24 more aggressively than we normally would (because of things like snoeshoe spam that would normally spread traffic across a wide range of IPs).
We use /24s only where we cannot determine different ownership of the IPs by looking at the abuse contact registered at the RIR e.g. if the IPs are different contacts then we don't bundle them into the same bucket. In this case Hurricane Internet owns the entire range, so the /24 is used.
We do this because there is no other way to do it that isn't completely abuseable by a bad actor e.g. rDNS is completely trivial to forge and to claim multiple fake entities. If someone has a fool-proof way to do this, then I'm all ears.
Then we get to your failings in this case:
- You don't have proper, working bounce handling.
You were repeatedly sending mail to an old customer and we were rejecting 100% of the email you sent to that address - stating that you should stop sending to it in the rejection message. We always reject traffic on traps we are building so that bounce handling removes them automatically over time.
- You decided to send a marketing message to a bunch of users, including to the address described above "We (rsync.net) are experimenting with a lifetime prepayment option". This message provided no List-Unsubscribe option at all, so I could not unsubscribe it without exposing the trap to your engineer (which is my primary concern as it takes years to build traps properly).
- Your engineer said to me "we took down 600 old accounts and are reviewing our contact policies going forward" and "there are still plenty of other customers who are listed generically as bouncing so we will have work to do here".
That tells me that you knew that you were sending to old accounts and that your bounce handling was either not great or non-existent.
I take our role very seriously and I and my team go out of our way to help anyone who finds themselves listed by providing evidence and advice and we always try to find a good resolution for all legitimate senders.
That is exactly what I did here with your engineer, the problem is resolved as the specific account was removed and you know now what you were doing wrong and how to fix it so that it never happens again.
Your engineer was appreciative and I said if there are any further issues in the meantime whilst you fix things on your end, then we would help by exempting your traffic whilst you did those changes.
I can't see how we could have done anything more in this case.
On our website, we provide blog posts and videos, we take part in conferences and workshops to give advice on how to do things properly so you never have blocklisting issues.
I can easily summarise here what everyone should do to avoid issues:
- Have proper rDNS for your sending IPs that is part of your administrative domain (e.g. don't use your providers generic rDNS that contains the entire IP in the hostname). If someone visits the domain name, make sure there is a website present that has contact details as a minimum.
- Make sure your abuse@domain and postmaster@domain accounts actually go to a responsible person.
- If you send any marketing mail at all, make sure it has a working List-Unsubscribe header (preferably HTTP that allows someone to unsubscribe without having to contact you). Important note: If you use a mailto: unsubscribe, then I cannot unsubscribe a trap, even if I wanted to.
- Make sure you have working bounce handling. If you repeatedly send to an account and it either bounces or is hard rejected, then you need to stop sending to it and mark it as bad. No excuses.
- Don't send to addresses where you have not contacted them or had any interaction at all for > 1 year. If you haven't kept in touch with your customers, then that is on you.
- If you have a web form that when submitted, sends a message to an external user, then you MUST a) validate all of the input fields and disallow URLs unless the field required it and b) prevent automated submission via the use of a CAPTCHA.
- When collecting email addresses for mailing lists, always use confirm opt-in e.g. send a message containing a link that they have to click to activate the subscription. Do not send any further messages to them until this has been completed.
- Make sure you separate IP addresses being used for outbound mail from those used for outbound NAT pool. Block outbound port 25 from the NAT pools and make your firewall notify you of any port 25 activity from any hosts as this could indicate they are infected.
- When provisioning a new mail server IP, don't send more than 30 messages per minute (e.g. 0.5 messages/sec) for the first day and then increase the volume over the following week.
I'm sure some will disagree with some of these, but I guarantee that if you follow all of these, then you'll never have an issue with a blocklist like ours.
I hope this helps.
The comment about "lifetime prepayment option" is exactly something that MUST have an unsubscribe option to it. I understand everyone is trying to increase business, but this is something @rsync missed mentioning, insisting you don't need one.
Bollocks.
> - Don't send to addresses where you have not contacted them or had any interaction at all for > 1 year. If you haven't kept in touch with your customers, then that is on you.
That seems odd though. "Keeping in touch" in regular intervals is exactly the kind of thing customers may want to unsubscribe from, so a working unsubscribe function means there are customers where I cannot keep in touch. How is that on me? That means I can't send e.g. a password reset email to them if they try to log in two years later?
My point here is "consent to send", e.g. I think most people would find that someone importing a years old list of email addresses and then suddenly sending marketing messages is unacceptable and constitutes spam.
If you're "doing it right", then "touching" your customers with a once a year "Hey, we haven't seen you in a while" message is absolutely fine and generally a good thing to do as it maintains consent, because they can unsubscribe if they're no longer interested (as they might have forgotten they signed-up or had an account).
If someone hasn't logged into your "service" for >1 year and now wants to do a password reset several years later, then I would say that is entirely different.
We generally try our hardest to handle this case without it causing a listing if you hit our traps with something like this.
One question though, what do you mean by 'traps'? I feel like I'm missing some interesting context
One of the ways most blocklists work is by employing spam traps which can either be individual email addresses or entire domains.
It's quite a large topic, but I'll try and summarise for you.
You can't just take a spam trap and use it to block anything that hits it - that would be incredibly unfair as you might not know the history of the email address or domain and you'd generate considerable false positives if you did.
If you buy a domain or register an email address and you know for certain that it's never been used before, then this is typically called a "pristine trap" and there are then various ways to then "seed" this so it starts to receive spam and malicious traffic. This is the exception where it could be used immediately. However this process usually takes a very long time (usually years) to receive enough traffic to be useful.
Any other address or domain where you don't know the history would be called a "recycled" trap. There are various opinions on how these should be managed, but generally it's accepted that you need to reject ALL traffic on these for at least 2 years to allow genuine senders to work out they are no longer valid (which is why bounce handling is important!).
The other type of trap is a "typo" trap, which is typically an entire domain (or an email address deliberately close to another) which could easily be typo'd. I have a rule here that says we never ever use these for any blocklisting (other lists will not have the same rules!), however they can be useful to see trends and detect compromised hosts (as these typos frequently end up in compromised databases that get sold around the dark web). They can receive considerable volumes as well.
Typo traps are especially why you should always use confirmed opt-in and employ CAPTCHA on your sign-up forms (although this also helps with all traps in general).
We carefully monitor all traps to see how they are performing e.g. detections .vs. traffic from whitelisted sources and false positive reports and will immediately remove a trap if it begins to perform poorly or has been made public in any way.
It takes years to build a trap network and we have to discard all of that work should they become discovered (as they could be used maliciously), which is why a blocklist operator will never reveal the addresses to you as they are a closely guarded secret. In fact on our own network, even we don't see them - we monitor them via a hashed name! (only I and a few other select people can convert the hashes to the actual domains or addresses).
I think that covers the basics - I hope that explains it sufficiently for you
This is very interesting to me. Could you elaborate further?
- Some providers (IIRC outlook being the main one) only support mailto: and most prefer it. I have never seen advice that you "must" provide HTTP before.
- What do you mean by unsubscribe without contacting me? Isn't HTTP or SMTP just as much contacting me?
- Why can't you unsubscribe mailto: even if you wanted to? Earlier you hinted that it may have to do with revealing the trap, but while a HTTP url may not directly include the trap address it would be linked anyways. (Otherwise how would the unsubscribe work?)
Thanks for all the valuable insights. It does sound like you are taking a responsible approach here.
I didn't say anywhere that you MUST support HTTP unsubscribes, but I just said it was preferable (to me at least).
HTTP typically means there is some sort of form which will typically auto-fill the email address from the database (because of the HTTP GET parameters included in the link) and all the user has to do is generally click 'Unsubscribe' and the address is automatically removed from the database.
The reason I prefer this is that if I or one of my colleagues want to unsubscribe a spam trap, then we can - but only if the above is true. (Disclaimer: we only do this very occasionally, usually it's only if we've spoken to someone and are convinced they are legitimate, have fixed any issues we've identified, but are hitting traps because of a single address - if we didn't do this then we'd either have to force the sender to reconfirm their entire database by asking everyone to opt-in again, or we'd have to whitelist the sender, neither are great options - so if we ever do this, then it's done hours/days after the event so as to not tip off anyone on what the trap was).
Because spam traps have to stay secret, we only reference them by a hash value internally (and only I and a select few can convert hash to trap value to do the unsubscribe) and as the messages are received on our trap network the traps themselves are obfuscated to a know value before they are stored in our evidence system.
mailto: means we can't do this at all - a trap can never originate traffic, that's cardinal rule of them, it would be grossly unfair if they did.
> - What do you mean by unsubscribe without contacting me? Isn't HTTP or SMTP just as much contacting me?
Poor use of words from me.
What I meant is that I don't trust messages that have an unsubscribe that does not encode the recipient in it in some way e.g. HTTP method without GET parameters (as this means I would have to do considerably more work to fill out the form) or a mailto:foo@domain.com?subject=unsubscribe - both to me scream our that there is no automatic connection to a database and that someone on the receiving end has to manually remove the address - which could take weeks, or be never.
Because spam traps get very different traffic to a real mail server, you can make a lot of assumptions that would be impossible to do on a real mail stream.
So generally speaking, if your domain name isn't in the top million domains (and there are still some exceptions to this - but I'm not going to list them all here) and you hit one of our pristine traps (see my post about the different trap types) with a message containing your domain, that will usually cause the domain to be listed.
Other lists and spam filters will have their own methods for this.
My belief is that Google and Microsoft only use AI for this (which is a bit of a nightmare if it gets it wrong).
(snip)
"That tells me that you knew that you were sending to old accounts and that your bounce handling was either not great or non-existent."
Yes, that is exactly right.
I have instructed everyone to be as liberal and forgiving of non-payment and failed contact as possible. These people, who were paying customers, have data stored here for safekeeping and we're not going to trash it because we haven't heard from them in 3 months or their email bounced.
I can give you hours of stories of customers who came back from military deployment, came back from depression, came back from prison, came back from financial ruin ... that contacted us, beyond all hope, to see if we still had their account. And we did.
After 21 years of doing this are there a few hundred accounts that we are giving extreme benefit of the doubt to ? There certainly are.
Bottom line: our duty of safeguarding customer data trumps this weeks fashionable spam heuristics.
I haven't suggested that you do anything differently to that.
Keep the data, keep the accounts, do whatever you feel is best for you and your business.
All I've said is to be smarter when it comes to sending email to old accounts that are repeatedly bouncing. I've outlined what was wrong and I worked with your engineers to resolve it.
The screenshot I got of your conversation with (Dave) was, indeed very well done and was both informative and actionable. You responded to us very quickly and with a high level of professionalism.
Further, participating here in this HN discussion is noteworthy and impressive.
FWIW, we immediately complained to he.net about the bad actor elsewhere in that /24 ... we'll see.
> Further, participating here in this HN discussion is noteworthy and impressive.
Thank you - much appreciated.
The quality, morals and perception of our product and how we treat people matters to me greatly which is why I'm always happy to get involved in discussions like this.
As is evidenced from other post on this topic, there are a number of competitors that don't behave in the same way and that leads people to think we are all the same and I absolutely want to challenge that perception when it comes to us.
> FWIW, we immediately complained to he.net about the bad actor elsewhere in that /24 ... we'll see.
Thanks, feel free to refer them to me and I'm happy to provide as much evidence as they need.
If you want to know why there's a new thread each week on HN about why it's impossible to host your own email service, this is why.
I run a mail server on AWS, and we use some blacklists to drop mail. It's quite effective and that's why people keep using them. A properly curated blacklist is a powerful tool, and more accurate than the machine learning mush that people have come to rely upon.
But this is by design[0]
> This blacklist has been created for HARDLINERS. It can, and probably will cause collateral damage to innocent users when used to block email.
And it makes for a perfectly usable blocklist. If you use postfix, the postscreen_dnsbl_threshold and postscreen_dnsbl_sites parameters let you create a simple scoring system:
postscreen_dnsbl_threshold = 10
postscreen_dnsbl_sites =
zen.spamhaus.org*5,
bl.spameatingmonkey.net*5,
dnsbl.sorbs.net*4,
bl.spamcop.net*4,
dnsbl-3.uceprotect.net*3
But it isn't really a problem with uceprotect, it's about how DNSBLs are used.
[0] https://www.uceprotect.net/en/index.php?m=3&s=5 [1] http://www.postfix.org/postconf.5.html#postscreen_dnsbl_site...
The article sums it up nicely: IP blacklists have their place, however, please don’t use the overarching neighbouring blacklists such as UCEPROTECTL2 and UCEPROTECTL3.
People have been complaining about this for a very long time; as long as I can remember. In the past they've also just permanently added people's IPs after complaining their IPs were wrongfully listed (not sure if they still do that).
The UCEPROTECT blacklist should be blacklisted by everyone. Yes, we need a blacklist of blacklists.
To be clear: I DON'T think blacklists are the problem. I think the problem is that a half-dozen major blacklists are controlled by unaccountable organizations who make up rules willy nilly, and privilege major email senders while punishing smaller senders / home hobbyists.
No. These blocklists are employed by the actor receiving the email. They have a perfect right, even on "the open internet", to decide that they want to limit who can send them messages.
There are tons of checks on the people who provide those blacklists, in the form of their users complaining about lack of mail delivery and ultimately not using their list anymore. We vote with our wallets, and as a group we have decided that these blocklists are useful.
> and for what?
To make email usable. Full stop.
If I send an email to alice@example.com, Alice is the recipient, not Bob the sysadmin for example.com.
This is the way that email is designed to work and the fact that users and service providers have choice is a feature, not a bug. If you don't want your service provider to do this type of filtering, that is your choice. But you have absolutely no right to say that Alice can't get the service she wants from example.com because it's inconvenient for you.
Edit: and just to head off the inevitable "but what about if it's at work and I can't change providers", the domain and email accounts belong to your employer, not you. You are welcome to work somewhere else if you don't like how they configure their systems.
Alice is not beholden to example.com, she can use any hosting provider she wants. Moving providers isn't even that difficult. It's also totally irrelevant, because again spam is inevitable.
Yep, that's the one.
If your hoster doesn't care about spam spreading from their IP space, you should take your mail server elsewhere. There's literally nothing to think about.
And if they do care about this issue, they are likely to be taking steps to remove any of their IP space from the blacklists, without being nudged.
PS.
I've been running a mail server for close to 20 years now and I do blacklist by /24 netblock on the second offense. This doesn't bounce emails though, just tags them as spam. So I had a quick look in the logs and Hetzner, Digital Ocean, OVH and LeaseWeb are all spamming a lot. LayerHost, Colo Crossing, Liquid Web, Host Winds and Servion are also close to the top. Anecdotal data, obviously, so caveat emptor and all that.
As of last night, the number one source of spam for both the personal and company servers I maintain is Digital Ocean.
Which is a shame, because otherwise I'm a happy Digital Ocean customer. But because of this, I would never move any of my commercial projects there.
By blocking outbound connection to port 25 outright. I don't want my VPS hosting company to "clean up" this way.
Most ISPs submit their domestic ranges voluntarily to certain blocklists, e.g. the Spamhaus PBL.
I used to have a bunch of blocklist services in my MTA configuration, with different lists having different weights and so on. Eventually it became clear to me that nearly all true spam that hit any one list would also hit Spamhaus Zen. So I scrapped my list of lists, and relied entirely on Zen. My life became easier.
Note: Zen includes the PBL; so if you block using Zen, then you will block mail from domestic ranges. Zen is pretty safe (unless you are expecting mail from domestic ranges). It's safest to use Zen as part of a scoring system, with something like Spamassassin to add heuristic content scoring.
It's a race to scraping the bottom of the barrel on per customer profit margin and pricing.
That seems excessive and abusive, I am not aware of any commercial ISP that is giving out /24 anymore. /29 is most common, I had to practically beg to get a /28 so what is the justification for banning an /24???
If they did, what would be Google's justification for blocking senders that send too few emails?
Good thing it's my own mail server then, isn't it?
The practical reason is that virtually all Whois lookups of offending IPs return blocks of /24 or larger, so that's a reasonable default. Besides, as I said, this doesn't result in a "ban", just tags emails as spam and passes them through. At my scale an occasional false positive is not a big deal.
Lets be real there is spam coming from gmail and hotmail/outlook as well and places like abuseix specifically state they don't block these ranges. So the large providers get excused for clean up because they are too big. Sure blocking colo crossing probably won't get any one to complain, but Digital Ocean is probably going to get some collateral damage. For your own mail server fine, don't accept it, send to spam - but there is a reason real RBL lists are very careful to skip the big providers or make sure they expire. Spamcop always had the best method - expire when the spam stops. Does it keep getting listing? Keep it longer. Rspamd also has a good method where an RBL increases the score. The hard line approach gives gmail and microsoft a large share of the email market and hurts smaller providers when they are not held to the same standards as everyone else. If gmail emails start bouncing when they have a heavy spam hit, then maybe gmail users will change isps and help gmail clean up. These are two trillion dollar companies that also have spam problems.
As far as UCEprotect. Their level1 is actually reasonable, especially for spam traps. The timestamps easily allow for you to find exactly what the spam is from with the smtp response and time frame. Their scanning methods are less so. Dos prevention measures can get you listed there and are not valid. The level2/3 lists are utter shit.
I only block AS 4134 and AS 4837, some AS that host services like shodan, and aggressive crawlers like semrush.
Anything that sends packets to my server get ratelimited quickly. Still barely noticeable for occasional human interaction. I also started with /24, but I am now up to /12.
PS. By the way, has anyone ever seen spameri@tiscali.it in the logs? It shows up almost on a weekly basis as RCPT TO address from literally all over the world.
Yes, I see it all the time in my logs.
Interesting. This tracks with my experiences too, but is there a good source for this that I can reference in the future?
I got out of hosting email last year entirely because of intractable deliverability problems with the big three: Google, Microsoft, and Comcast. Comcast's issues mostly appeared intermittent and the result of incompetence. Google and Microsoft have clearly been competing to see who can kill small email service providers the fastest.
Google can be tough though. If their Algorithms don't like you, good look.
I was responsible for a new IP/Email setup and while all other relevant providers liked our mail and Google Postmaster tools didn't show anything bad, all our mail went to Spam in GMail. I found out why when even mails from completely different senders would usually go to Inbox but hit Spam once our Website or email was included:
Despite showing it as "high reputation", GMail must have considered it as utter trash. Changed to a different, older domain and all has been fine since.
You can't make this up. They pass all abuse reports throuh their spam filter and this affects your server's reputation. Then at some point they start rejecting emails from you with a 5xx code that stands for "your IP is on our blacklist". Contacting "Hotmail sender support" will yield "We see no problems on our end, kindly fuck off". You have to then go through a ritual of summoning a manager, taking anywhere from 2 to 5 days, at which point they will say "we put a mitigation in place, will take up to 48 hours to propagate", but it works right away. If it doesn't, it means that they mistyped your server address and the next chance to ask them to fix this will be, you guessed it, in 48 hours. So a week of bounces in total.
Ran into this 3 times in a course of a year, then took a closer look at the logs and - yep - in all three cases it was triggered by sending a report to abuse@outlook.com.
It's just such a magnifecent clusterfuck it's jaw-dropping.
A few years ago I did have major problems with hotmail and icloud, but recently it doesn't seem to be an issue. I think I didn't have DKIM set up when I was having problems. I have noticed that google will auto-junk your email if you don't have SPF, DMARC and DKIM set up, and you must set DMARC to "reject".
On the other hand, some VPS providers are still allocating multiple customers to the same IPv6 /64 using SLAAC by default, and this will make it impossible to deliver mail on IPv6 since reputable RBLs always blacklist the whole /64.
As far as the argument about spamming being immoral but not illegal, I’ve never seen a reputable ISP that didn’t prohibit unsolicited bulk email in their terms of use, so the grounds for reporting it is that a customer is violating the terms that they agreed to follow when they signed up.
And to answer the question of whether or not RBLs are useful: in my experience, yes, they are quite useful. The biggest problem I’ve noticed with them is not typically false positives on small providers, but false negatives on giant companies like Google who cannot ever end up on an RBL because they process so much mail but don’t do a good enough job of preventing their servers from being used to send spam.
And my hosting provider basically said "we're warning against using our servers for outbound email", which amounts to "we're not gonna do anything about your bad neighbors that landed you on the uceprotect lvl2&3 lists"
Been running my own email for 18 years, and outlook is the only place that i have problems delivering to.
Just to be sure, you are using the Outlook form[0] and not the Office365 form? Have you signed up for SNDS?[1] All these hoops are pretty dumb, and I don’t think they have updated any of these systems since the mid-2000s so they are also pretty terrible to use, but here we are.
My limited personal experience with Outlook’s removal process is that it’s just slow and confusing (sometimes 12 hours for a response, sometimes the response emails are empty). The last time I had to do this, their TLS certificates had expired (and they use HSTS), so I could not even get into SNDS for half the day until someone there finally noticed and deployed new certificates.
[0] https://support.microsoft.com/supportrequestform/8ad563e3-28...
My email server is very small, so it's not a big deal... But thanks for the links.
I do wonder if the guy behinf this scam is randomly blocking whole ip ranges to make a living, having enough people agreeing to the racket.
Google's recent message was more helpful - apparently forwarding emails from the accounts I setup for my kids to my wife's Gmail account triggered some obscure rule that ruined my server's reputation with Google, and I think Google & Microsoft collaborate because the issues cropped up within a week of each other.
The interesting thing was I discovered the Outlook issue by trying to reply to an email sent from an Outlook customer. Yes, my reply to an Outlook customer's email to me was blocked because of my server's reputation.
To be clear, I run no mailing lists nor solicit any business with my email server. I use it for personal use only and my consulting work which involves know contacts. The forwarding I spoke of before is solely to our own personal accounts.
I use Mail-In-A-Box, for what it's worth, on a Linode VPS.
1. Set up on https://sendersupport.olc.protection.outlook.com/snds/
2. Read everything at https://sendersupport.olc.protection.outlook.com/pm/
3. Make sure everything is fixed, then use the link to the form hidden under Troubleshooting > "Sender services, tools, and issue submission" (the link's label is "here") to contact support. Make sure all fields are provided, including a website. It may take a few days to get a response. You may have to try multiple times until someone actually helps you.
Google:
Pray
a) not worth paying
b) should never be used by a production mailserver to block messages.
From the beginning there have been enterprising RBLs that are clearly overbroad, and offer to accept money. The money is not for getting off the list, it is always for something else so as to appear legitimate and a side effect is getting your domain off the list. This model is unethical at best, and is right up there with companies that snail mail over-priced domain renewal notices.
It's not UCEPROTECT either, big vendors like Cisco with their IronPorts are just as bad.
Of course they offered to unblacklist them in exchange for payment, or wait.
Waited 2 weeks and they dropped off. I've yet to hear about anyone using their DNSBL for anything serious in 2020/2021/2022
I only knew about the listings because a monitoring service emailed me about it.
While I personally don't use UCE (and personally think that they're not good at what they're doing), unless you've get that IP range before 2012, I doubt it's a new one. Many spammers do often exploit RIPE's unallocated IPs for their spamming operations (either using BGP hijacking or just asking RIPE nicely for a range), which unfortunately is a perennial problem.
The answer to this is the other option, leaving the hosting company. In this manner, every hosting company gets a choice - either they will kick out legal-but-immoral things like spammers, or they will not and rightly lose their above-board customers.
This is essentially how the global e-mail community self-polices by establishing a norm that a host either has to work to exclude bad actors or will get boycotted/excluded for allowing them.
Why is he considering—with incomplete information—the hosting company's legal options and the severity of the spamming?
Edited to add: He could also consider whether it is his hosting company with the problem or the orgs using UCEPROTECT; although as someone who needs their emails to go through "no matter what" that can be more difficult.
This assumes that the RBL in question is a good actor. It is as bad as the spammers.
> This is essentially how the global e-mail community self-polices by establishing a norm that a host either has to work to exclude bad actors or will get boycotted/excluded for allowing them.
Most of us have ignored UCEPROTECTCTL lists for a long long time as it appears to be a money grab.
It's description is not subtle:
> This blacklist has been created for HARDLINERS. It can, and probably will cause collateral damage to innocent users when used to block email.
So if the mailsystem you are trying to reach employs it, is either experiencing spam levels that justify it's use - OR they made a mistake in using it, if this is the sole reason you are being banned.
The first order of business is of course to complain to your hosting provider. Nobody wants spammers on their networks - but if they do: then this is kind of exactly the reason for this list. The policy describes in detail what made it possible for this netblock to end up on the list, that should be enough for them to take action either pre-emptively or by notifying their offending customer, and if neccessary kicking them off the network.
The next thing you can do, instead of paying for whitelisting, ist to contact the mailserver-admin at system you are trying to deliver mail to. This can be a bit of a hassle - seeing that your mailserver just got blocked - but it usually works. The same way systems don't want to receive SPAM they also don't want to overblock, after all they want their users to receive emails as well. If you are the mail admin of a sending system and you're reaching out to the receiving system this is usually a pretty good indicator that you don't want to spam them.
I have had success doing this even at some larger ISPs, where you would expect this to be more difficult.
I very much enjoy these blocklists - simple, transparent. Loads better than the kafkaesk black holes that are the major mail providers who barely care, and who do not give you easy recourse if you are mistakenly blocked.
That's nice in theory. In practice, UCEPROTECT level 3 is used by, for example, all Microsoft properties (including hotmail, etc). And UCEPROTECT level 3 lists a lot of netblocks.
So, if you want to send e-mail from, say, Digital Ocean, to any Microsoft managed e-mail domain, you can just forget about it. DO doesn't care, Microsoft very much doesn't care. Good luck trying to contact a postmaster there!
... attributes which they achieve by (1) selling services to anyone and anyone and (2) not dedicating any resources to fighting spam.
So you got what you pay for.
In the end, even with 3rd party blacklists networks still maintain control over what incoming traffic/mail they block or allow.
Internet censorship is a real issue, but blocking hackers and spammers are not an example of internet censorship gone wrong. ISPs black spam over email. They block malicious web traffic too. Every network gets to decide what to accept or reject. It's about as fair a system as you could ask for.
You analogy is flawed. In the case of IP blacklisting, it is an intermediary preventing legitimate email delivery on shaky ground. My correspondent want the mail; yet she never got to see it. This is censorship.
The 3rd party hosting the blacklist is an intermediary, but the network operator on the receiving end isn't. They still have total control over what lists to subscribe to, how they'll be used, and what to block or allow. That's not censorship. As a network operator it is there right to block or accept whatever traffic they want. If you, as a customer/end user don't like how they run their network you're free to operate your own or choose to give your money to someone else.
Some platforms rise to that level I think. Youtube is one example. Email doesn't really have that problem. Anyone can install an email server, and there are a huge amount of options (many of them "free") for those who don't want to run their own mail servers. Email is still a competitive space with different companies offering specialized services targeting people who want privacy, low cost, mass mailing capability, tracking features, etc. Email itself has a ton of competing communication protocols. There are plenty of folks who prefer to communicate over chat or by text. You can even keep an old email address and forward anything sent there to another one or use a separate email address for sending, but get replies at your old address. Switching email addresses is about as easy as it gets.
First, there actually does exist a strata of spam which is plainly illegal. All the phishing spam and all the messages that claim to be from networks and/or addresses that they are not, for example, are illegal. The problem is that it's not enforceable.
Second, sending unsolicited messages when you do not have the permission of the sender is unambiguously wrong. Large sources of spam get around this by mixing bad messages in with good ones. This is why we get tons and tons of spam from Gmail, from Outlook.com, from Sendgrid, and so on, even though they really should know better.
The point is that your "bad neighborhood" doesn't just have regular spammers - it's almost certain there are illegal and egregious spammers that your ISP is doing nothing about. How do we know this to be the case? Because many of the blocklists also run honeypot email addresses. If email addresses get harvested and someone sends spam to these addresses, you can be 100% certain that no permission was ever given, so the behavior that leads to this is definitely wrong.
ISPs make too much money to punish anyone but the worst of their clients, and that's definitely a factor that contributes to the affordability of the author's ISP.
However, the author left out one big, simple, obvious option: pay significantly less money than the cost of the blocklist extortion to smarthost through an ISP that has good email reputation. You get what you pay for, so when you save on your ISP, don't be surprised if you have to pay a little more to make up for their shortcomings.
I've dealt with real time blocklists as long as they have existed. They are not going away any time soon. I agree that the paid exception lists are a bit shady but I also see the validity of their methods of temporarily punishing everyone on a hosts network to put pressure on the ISP/platform provider to police it's own network and remove spammers. The best one can do today aside from securing ones own server is to research an ISP's IP space ahead of time to see how dirty they are. There are plenty of providers that cleaned up their act some time ago. Linode is a great example of change. New accounts can't even send email unless they open a ticket and prove they made some effort to comply with can-spam. More providers need to follow that example so that we don't run into this problem of dirty networks that real time block-lists like UceProtect have listed. It's an imperfect solution to an old ugly problem.
- A dedicated server provider that has been around for a while and has a strict AUP and is known to enforce it.
- A hosting provider that is not entry-level in cost. Spammers gravitate towards cheap throw-away ephemeral solutions.
- A hosting provider that verifies identity of its customers. e.g. Dunn and Bradstreet lookup for commercial customers. Video conference meeting for individual customers and commercial customers and that have mutually signed contracts.
Short of that if you just want to use a VPS provider then I would look up their AS number of a prospective provider, get all their CIDR blocks and start validating their IP addresses against the numerous RBL/RSL sites. AFAIK there is not a good database of this. Good RBL/RSL sites will remove listings after a week or two. One could even open a ticket with a VPS provider and state your intentions to run a mail server, explain your process to deal with spam and ask for an IP from a clean subnet.
Another method would be to look at colocation providers. They can give you blocks of IP space. On the low end would be Hurricane Electric if you are on the west coast. A step up from there would be Sonic. There are a myriad of colocation providers that scale up in price from there. If you wanted a presence in many locations then Equinix may be a good option though they can be expensive when you are first starting out. All of these providers have partnerships with network providers and can help get you peering arrangements, address space and much more.
Blocking outgoing port 25 unless going through an extra step is IMHO against the spirit of the internet, and would make operating a personal email server harder than necessary. The world of email is already too centralized as it is.
I completely agree. I believe the commonly used phrase is and this is why we can't have nice things. The open and decentralized internet is also open to people with ill intent. We can solve things with technical solutions, monetary solutions and/or legislative solutions. All of these are double edged swords and have varying degrees of effectiveness and unintended side effects in my opinion.
I don't really know what the right solution should have been that would have made everyone happy minus the people with ill intent.
I found some old lawsuits [0] that got injunctions and even damages awarded from being included in black lists.
I hate how the big tech customer nonsupport is bleeding into small firms attitude. “Sorry, you did nothing wrong but might one day so get fucked.” is really not something that should happen very much.
[0] https://www.techdirt.com/articles/20051228/1349229.shtml
I've had an IP for about a decade that never once sent spam, but has ended up on blacklists from time to time (hosting EICAR on a web server apparently gets your mail server banned, 15-year-old me found out). SPF nicely says that this IP is supposed to be sending email for this domain. I don't think I'm on blacklists anymore, but email still ends up in spam folders nine out of ten times. People are giving off signals that my messages aren't spam all the time (it's my personal email server).
Years of non-spam emails count for nothing whereas a single spam mail from an adjacent IP can get you on such a list. Somehow it's a bit imbalanced.
Obviously, legitimate use-cases exist for such services, provided that they are operated faithfully by people/entities with some level of credibility.
Whether or not Spam is legal, many (or most) service providers list it as prohibited in their SLA (Service-Level Agreement). Depending on the type of setup they run--the clientele; company focus/prioritization of ROI--they forward complaints to the account owners, with the understanding that the activity must be stopped. Again, this depends on the service provider, especially as you start thinking on a more granular level. E.g. they can have automated systems handling much of this, or they may have employees handle the communications and provide help to their clients in thwarting such activity. Generally, though, the account owners need to balance the security with compatibility--or the uptime and short-term performance of their services.
Ultimately, however, some of these spam block-lists--UCEPROTECTL3 in particular--are blatantly using a sledge-hammer approach for little benefit to any of their users, in the hopes of spurring controversy; or fear that leads to customers; or outrage from customers toward the service providers. In the end, the intention is purely to get paid, either by clients who sign up for their service or by the internet service providers that can't wait multiple days or weeks until their network gets de-listed. Their accusatory and aggressive language conveys to people that they must be associated with either some amateur company or one that uses, and knowingly profits from, nefarious practices and shady characters if they somehow ended up on their list.
Depending on where you stand, it can be seen as deplorable or underhanded behavior maybe even rising to the level extortion, or it can be considered a shrewd business tactic.
Practically though, if you want to get your mail delivered, you should use Amazon SES or the like to send it - setup is really simple, and they have the clout to not be blacklisted EVEN THOUGH they are definitely being used to send spam. At $1 per 1,000 mails, it is unlikely you will even feel the cost.
(I commented on this a few weeks ago at https://news.ycombinator.com/item?id=29713030.)
Really we probably need some sort of anti-RBL system. To keep good actors honest and force bad actors out of business.
Maybe something like that could be done with email, but without a culture around it, figuring out what addresses you need to add to your contact list when you add new services could be a pain.
I got listed as well (layer 3 only), because there were more than 32 cases of spam in an ISP that maintains more than 4 /16 networks across the whole country. With the whole AS blocklisted, I fail to see the point of UCEPROTECT and agree with the OP that they are mere thieves.
On topic of protecting people, they have a trivial XSS right on the input field where one checks if their IP address is listed - yes, a GET variable is printed directly to HTML.
</rant>
Hosting companies can kick customers off for basically any reason they want to. High-end hosting companies will absolutely kick a customer off if the customer’s activities result in spam blacklisting.
When I first signed with Rackspace, it took almost a week to negotiate their AUP (acceptable use policy). They were basically unwilling to give us any solid assurances about account suspension. They gave themselves enormous room to determine what was acceptable behavior on our part.
If you haven’t carefully read your host’s AUP, it might be enlightening. If your host doesn’t have an AUP, or doesn’t enforce it, then yeah maybe that is not the best neighborhood.
When I worked with a company who ran email systems like this we'd always steer customers away from lists like this, in cases where we couldn't deliver mail because of it we blamed the other provider. This actually worked a lot of the time and the customer would wind up contacting the sender another way to let them know they had a problem, and fairly often they'd change lists.
Our standard messaging was something like, "Google and Microsoft are receiving our emails just fine, something is wrong with the receiving service"
While I do not use any blacklists, I do make my own. I've found that blocking individual IPs is useless, there is always another one. However, if I block the entire IP range I have much more success.
This is your hosting company's fault for allowing spamming and doing nothing about it.
"Allow cookies" modals are horrible but GDPR gives people in the EU rights to be forgotten and not contacted in the future. Reminding companies about GDPR regulations works very well.
As soon as they do, email might as well be dead.
The GDPR already caused some sites just to block access to EU countries.
[0]: https://noyb.eu/en/noyb-files-422-formal-gdpr-complaints-ner...
Edit: The nice thing about running the mail server personally and without a relay (like mailgun) is that mail is to-my-end encrypted. If the other party is running its own mail server, it could even be E2E encrypted. Considering the vast amount of personal information that going through email, this makes me feel good in terms of privacy.
I have never heard of UCEPROTECT and fortunately, I never had to deal with it. The language on the webpage somehow reminds me of Kryptochef...
A small inconvenience is that I had to unblock the IP on Spamhaus PBL every month. By now, it feels as if they know me, because I now only have to do this once if I get a new IP...
Many mail servers are nice and provide the reason for the block even with hints how to unblock it. I successfully unblocked it on Abusix and Microsoft. Never had an issue with Google.
GMX on the other hand will never accept my email because they require a proper PTR record. They are the only company I have come across and I find that scandalous.
Spamhaus PBL is build based on your ISP telling Spamhaus which IPs are dynamic and which IPs should not send email. Your ISPs seemed to be nice enough to allow you delisting from it, which not all ISPs will allow.
Abusix has a similar list, but its completely build based on dynamic looking or no PTRs. You can create an account and delist without any issues.
Never the less there is way more services than GMX that block based on dynamic or no PTR. A lot of smaller solutions have this option checked by default. And it actually has been a best practice for decades to have a proper PTR.
If you can, I'd set one and be done with it.
A PTR match is a very strong signal that I am looking at a hosted server or business connection which is configured to send email and not a compromised windows PC on consumer dsl which used to be a huge problem. ISPs supply those dsl ranges to blocklists as well.
When I first started hosting my own mail it was on an old repurposed linux PC on ISDN in the early 2000s and even then my ISP offered configurable PTR records along with static IP in a small business package.
I moved my family email vps to a different region recently and it took me a few seconds to update the ptr records for ipv4 and ipv6. It seems a very low bar for such a strong signal that you are dealing with a mail server and not some random compromised machine.
NOTE: By using Level 2 blocking, be prepared to lose a few mails too. DO NOT BLAME US, YOU HAVE BEEN FOREWARNED! > https://www.uceprotect.net/en/index.php?m=3&s=4
Use of Level 3 for blocking is recommended only if you are a HARDLINER and you want to cause service providers and carriers that have spammer / abusive clients to be quickly and effectively blocked and it does not matter to you if regular email is also occasionally rejected. > https://www.uceprotect.net/en/index.php?m=3&s=5
This has been the subject of hot debate on one of the popular mailing list operator's lists, and the providers that find themselves in Level 2/3 reliably are working on strategies to deal with that but it becomes complicated to do anything that affects thousands of customers. You'll find the providers that aren't on these lists are the ones that have an equally strict policy for allowing outbound Port 25 from their customers and a severely punitive abuse desk.
IMO, I run a personal mail server and deliver directly and if my mail is rejected or lost then I use another mechanism to contact that person (usually telephone) and inform them of the failure. I also provide usable feedback to senders for why I may have chosen to reject their message. And when it really matters, I also have a relay I can use for troublesome transports.
You don't run a mail server for other people who are not so careful with their e-mail addresses, let alone for a large organization. You don't run a mail server which hosts public mailing lists.
We should wake up and raise the standard. We could even make it a public utility, or a utility that contracts to competing companies, or just develop standards for delivery and user rights - nobody can read my mail, nobody can interfere with it, delivery guaranteed within X time, etc. One challenge is to esure that individuals are still able to run their own servers (e.g., develop standards that the FOSS email software developers can meet). But we could have similar laws to regular mail, protecting delivery, confidentiality, and against fraud.
Why do we endure this awful system? Recently I feel like the IT world has shifted from an innovative, creative, we-will-solve-any-problem, build-it-in-our-garage community, to one that is defeatist (nothing we can do), status quo, depends on mega-corps (or VCs) for everything, and which prizes a how-much-can-I-take narcissistic business mentality rather than a creative, change-the-world engineering one.
To some extent, this also applies to legitimate business mails going to Spam. That was a big reason for me to switch away from Outlook to a provider where I could configure the spam filter to the equivalent of "Viagra scams etc".
Of course it's possible (and quite easy) to set up your own mail server, there are tons of guides out there for that, but this is just the tip of the iceberg! Once you have it up and running, you'll find yourself in a constant battle between those who spam you, and those who consider you spammer, real or not.
Using blacklists to filter out the bad incoming email is one efficient tool, but you need to rely on good, reliable and *honest* blacklist systems. Sorbs and Spamhaus comes to mind, compared to uceprotect and backscaterer. The last two asks you money to delist you regardless of what you did, and this is typically bad player playing on top of bad players. They do a race to the bottom by offering a racketing service.
But the worst is still that popular mail server uses these! Microsoft uses UCEProtect to filter email! So, as a mail provider like OP's, you find yourself in a pickle where you need to decide if you want your email to be delivered to Microsoft's users - that means paying, or not.
That is why managing emails has become a million dollars necessity if you want them to be up and running properly. You need to be close to the big ones, have a seat at the big tables where the decisions occurs and hopefully have your IPs added to whitelists.
To plug one very useful link in all this, I'd share Hetrix! (https://Hetrixtools.com), they monitor your IPs and notify you when they are listed on almost all the existing blacklists, and automatically delist them whenever possible, or provides you all the info to do to delist them. As a mail provider, this is on of the most useful tool there is.
(as described e.g. here https://www.howtoforge.com/tutorial/protect-your-server-comp... )
(in both modes, download & upload)
My root server hosts as separate VMs at least a website and an email server => both are magnets for all kinds of scans and intrusion attempts => I've got logscans + honeypots etc... set up (fail2ban then closes the source IP's connection for a while) which seem to be working.
I wonder if things could improve (e.g. preemptive FW-drop) by making fail2ban use "badIPs", and to make fail2ban feed back to badIPs the "bad IPs that I identify"? Any personal experience in this area here? How dynamic/reliable are "badIPs" lists?
Either we can concede defeat and relay personal correspondence email through a commercial service or we can just accept that some recipients will not receive our emails.
I've adopted the latter stance because I don't operate a business that uses my personal mail server and therefore have the luxury of not caring a whole lot. My stance is that if a recipient is using an email service that blocks me for no good reason, they don't want to receive my email. They can complain to their service provider if they choose to. In a few cases, I've asked them to complain to their service provider and have seen some minor corrective action. But that's pretty rare.
I gave up trying to run SMTP ourselves years (decades?) ago. Services like Sendgrid et al all made it very easy.
However, in the last couple of years I've noticed even the basic paid plan on Sendgrid, Sparkpost, Mailgun (I've tried them all now I think?) still gets constantly blacklisted. Going to support gives you one answer: upgrade to a dedicated IP, which is close to $100/month, and really quite wasteful on the IPv4 space (if anyone knows a cheaper/better alternative, please let me know!).
The problem is I am often building systems which require very little (but really important) email - the odd password reset request perhaps.
For small webapps you're going to be spending probably more for this dedicated email IP than the rest of your infra combined.
That blacklist is listed as "Suspect RBL provider". That says enough about it...
The people doing the actual blocking based on the list have the responsibility for their actions. If, say, one of the largest email providers in the world is found to be giving preferential treatment to the email of other large email providers then they can't use some list as an excuse. Their actions are still anticompetitive and generally harmful.
We shouldn't forget to go after the entity that runs the email server that is blocking email from our servers. You don't have to care that there is a list. Blame the right people and involve governments if required.
And of course the side-benefit for Gmail to enforce more anti-spam provisions is that it increases the reliability of the data they can mine from your mail, and my mail too since so many people still use Gmail.
It is in the interest of big email hosts to shut out small, independent senders to strengthen their position of monopoly. Again, I have no proof.
Wait, what? That's like off-handedly mentioning you eat a pine cone for breakfast every day and glossing over it like we're not going to wonder if you're crazy.
I've been running my own mail server for 9 years now and I don't use any blacklists.
Blacklists are the red line districts of the web. They are just a large blunt tool used in a very inappropriate way. The system of IP address space ownership assignment where the vast majority of the IP address users will never own an IP address. It's like renting a business suite in a large office building and being accountable for some unscrupulous neighbor.
> If my understanding of the law is correct, spamming is legal, albeit immoral
But the FTC takes a rather clear stance (at least in my eyes) [0]:
> Despite its name, the CAN-SPAM Act doesn’t apply just to bulk email. It covers all commercial messages, which the law defines as “any electronic mail message the primary purpose of which is the commercial advertisement or promotion of a commercial product or service,” including email that promotes content on commercial websites. The law makes no exception for business-to-business email. That means all email – for example, a message to former customers announcing a new product line – must comply with the law.
> Each separate email in violation of the CAN-SPAM Act is subject to penalties of up to $46,517, so non-compliance can be costly. But following the law isn’t complicated*
On top of that there's GDPR [1]:
> After the GDPR passed, some people said it would be “the end of email marketing” or “the end of spam.” But it will be neither. Spam has always been outlawed or against the terms of use of most email providers. Those who send unsolicited or malicious mass emails will probably continue to send them. Did your spam folder dry up after May 25, 2018, when the GDPR took effect?
So, it seems the laws are in place, it's just that the jungle of unwanted & unsolicited email continues. It sucks when it's your server that get's blocked in a dragnet because you've a noisy neighbour. But like others have stated, move neighbourhoods.
[0] https://www.ftc.gov/tips-advice/business-center/guidance/can...
If your outbound mail is being blocked by some server just because your MTA is in one of those lists, it might be worth contacting the postmaster at the other end, and asking them either to drop those two lists, to use them only for spam-scoring, or to whitelist your MTA.
We have been with the same provider for 15 years, no problems, ever. It seems that email delivery started to become unreliable about six months ago. After repeated attempts to fix it we had no choice but to move elsewhere.
Zoho mail. Very happy with them so far. The webmail client is excellent. Deliver is fast and without problems. We usually setup users with multiple email addresses. Using email aliasing works great. In other words, you could receive email from you@example.com, sales@example.com and vendors@example.com. This only consumes one paid user account.
Ultimately, we decided to move email for three businesses to Zoho.
The prior host was GoDaddy. I can imagine someone saying something like: "You deserve the problems you had". My counter is that we've hosted email with GoDaddy for more than ten years. No problems at all. Rarely an issue. In my opinion, most of the GoDaddy hatred is truly unfounded. If you are running a more sophisticated infrastructure you need to go to something like Linode (which we use for exactly those applications) or something else.
In the last 6 to 12 months, something changed. Once reliable email delivery became questionable we had no choice but to make a move. They said they could not do much more than request the servers in question to be removed from the various blacklists. In other words, play whack-a-mole with that entire ecosystem. A loosing proposition. So, we moved our business. Perfectly sensible decision. No hate involved. We are buying a service that, in my opinion, was no-longer being offered. Done deal.
It's illegal in the US: https://www.ftc.gov/tips-advice/business-center/guidance/can...
The US firm providing anti virus/antispam to UK politicians wont even confirm if they have blocked email sent from constituents to UK politicians.
Some websites also block emails from some European email providers, but it always traces back to US tech firms.
I figure that places like spamhaus are a good part of why gmail took over.
When a very big company blocks nearly all email from another company that they compete with in hosting services I think it should be investigated as a possible abuse of market power by the EU and DOJ. I think there are important questions to be asked about the damage this does to competition and consumer choice.
I wouldn't worry about UCEPROTECT listing. They have an option to pay to remove IPs which to me looks like a protection racket. If anyone blocks solely on the basis of those lists instead of taking signals from more respectable block lists they are only hurting their own customers by blocking legitimate email.
I have multiple servers at multiple providers for various things so it is relatively trivial for me to add a rule to a transport map and have my server securely login into another machine to relay the email to a misbehaving receiver. But that requires slightly more than minimal knowledge and many people are pushed towards a small number of gatekeepers who will ultimately be purchased by a cartel of mega companies and nobody will know how to run a mail server anymore. Which is a shame because it really isn't very hard or very technical. What makes it tedious is the behaviour of some companies. The technical workaround to a block is as easy as routing email via another IP and takes a few minutes.
But then there is the process of getting the original IP unblocked, sometimes via multiple processes in businesses with a single owner, seemingly none of them knowing what the others or even their own tech people are doing. And/or swapping out the IP for one in a clean block.
What a ridiculously hostile message. Just because a website or mail server or whatever they are talking about got compromised does not mean that the owner "did not care about their security". This sounds like a case of a blocklist operator on a power trip, talking down to people.
"Sorry, Your IP is listed in some other Realtime IP Blacklist."
I am looking for somebody that already has a sophisticated mailing operation, and I have the rest. Anybody interested?
Rather than defaulting to allow-all, it should default to deny-all. Content is let in on a case by case basis.
I found an email service the operates this way, Hey.com. Unfortunately, it’s pretty sparse in mostly every way but this. However, it’s practically impossible for me to receive spam. You can’t get into my inbox unless I’ve approved you.
It doesn’t make any sense (to me) to maintain these hilariously complex deny lists that results in infinite cat and mouse. Just deny everything and only allow in known good actors. Voila.
That being said, I don't know how SMTP hasn't been replaced yet. It shouldn't have lived much longer than the early 2000s, in my opinion.
(I certainly agree that we should have a new one, I just can't see how that would happen)
Brave loves Verizon so much they even listed them as a featured advertiser. https://brave.com/brave-ads/
Unless they're South African :)
Set up your email wherever is convenient, encrypt stuff that matters, and move as much communication as possible elsewhere.
Network-specific messaging tools are another option. I strongly prefer open protocols.
The concept that anyone anywhere can intrude on anyone anywhere else at no financial or reputational cost is ultimately flawed. It works only so long as those with that access are few in number, generally of mutual interest, and act in a largely principled manner.
As numbers increase, and levels of interest and level of principles fall, the system will collapse.
Usenet was the first such network to fall to this dynamic. Email is well on its way. Telephony is into its first years of general intolerability (any direct-dialed universal access, wired or otherwise). Facebook faces this threat less through its lack of filters than the defection of high-affinity users.
Postal mail has its own issues with quality, but the associated costs do in fact impose a minimum bar to malicious content.