Probs sending e-mail (SMTP)?


Hi –

for the past 8 days, I’ve had problems, on and off, sending e-mail from all our client machines. (Using outlook). When I’ve reported the problem to Dreamhost support, I receive, after a time, a message that says many have reported this outage and that it’s being looked into or worked on or otherwise investigated.

Most of the time I’ve gotten a reply that indicates that it’s a mass e-mail because so many people have reported such a problem.

(Webmail does work).

And then I get a follow up messsage from dreamhost that indicates the problem is “resolved” - but almost always it’s not. Today, I had a brief reprieve and mail got sent.

The error I get in outlook is:

Is this a dreamhost problem or my problem. I guess I’m the System Administrator though I’m certainly a poor, ignorant one! I thought dreamhost was our ISP but when searching previous threads, I’ve noticed some from dreamhost suggesting that one contact one’s ISP, which is confusing, since I thought dreamhost was the ISP.

Network Solutions is where I registered the domain, but dreamhost is the site upon which we run our website and e-mail.

Anyone have any suggestions on how to proceed? I’ve browsed the Wiki (and tried to understand it!), checked, and sent messages to Dreamhost Support.

I remember a year or two ago, there were a lot of problems when dreamhost was expanding, but it’s been smooth sailing since then and until now. But our users at our small business are starting to get irritated that outgoing e-mail is getting held up in the outbox and from the wording of the replies I’ve gotten from dreamhost support I assumed it was a problem on their end.


ISP = Internet Service Provider, and usually refers to company providing your local connection to the internet. Today that is usually via a broadband connection such as cable or DSL, many of those companies also provide some form of hosting and/or email.

Dreamhost is a web services provider or webhost, and does not provide local connection to the internet.

You should read this Microsoft article

I’ve had this particular problem myself before, and it in fact turned out to be malware (also or formally known as a virus) related in my case. It’s been quite some time since my incident, but as I recall I concluded the malware must be attempting to send a large number of outbound messages all at once and the server slammed the door shut to prevent it.


Thanks for the help! Much appreciated. Will read the link.

I did check each outbox of each outlook to make sure that there wasn’t a lot of stuff being sent or that sending limits were exceed or jamming up our quotas (this has happened before and I confirmed with dreamhost that this wasn’t the case – i.e. at least that sending quotas weren’t exceeded). But I’ll certainly check again. Last time, it was some overzealous user trying to send 3000 email at once, but in that case, dreamhost sent a message saying sending quotas had been exceeded. I guess the virus could be on the server or any client. Or might act differently than someone actually trying to send 3,000 e-mail at once!


It is not you, it’s them, most likely. They are having issues with load balancers, or that is the last story they have given me.

I’m guessing your email server is in the homiemail cluster. Mine is homiemail-sub5 and have been having same issue for the same time period.

FYI, if it were a virus, you probably wouldn’t find the outgoing SPAM in your email client - it might not even be a virus on your system.


Thanks for your reply! Yes, My email server is homiemail-sub5. Let’s hope they resolve this soon. I wonder why they’re not posting this as updates on – I also notice that there are no comments from others in the existing threads that were open.

Anyway, frustrating!

Here are (some) of the different responses I’ve gotten, so far:

27 NOV:

[quote]This message is in regards to the intermittent SMTP connection issues you’ve been experiencing. We apologize for the mass response, but found it necessary due to the nature of the problem. Many of you have received this notice yesterday, and unfortunately, problems remain intermittent at this time.

Our administrators are working diligently on trying to find the cause and solution. As you may already know, this is also a part of a larger scale mail infrastructure improvement that is already in the works. Unfortunately, the issue may arise temporarily again, but we assure you that it is not going unnoticed and we’re doing all we can to alleviate these issues.[/quote]

26 NOV:

[quote]Earlier today we notice several of the mail servers were having highly fluctuating CPU loads. The CPU loads on these machines were spiking up and then dropping back down on their own. In addition we found that webmail was taking longer than expected to load. Our admin team determined that the Apache service for webmail was not running properly. After Apache was rebooted webmail began working quickly again and the CPU loads across the mail servers also appears to have stabilized. We believe that extra CPU load was likely responsible for mail client connection drops. Email connections should be working correctly now.

25 NOV. Same reply (wording) as 27th.

24 NOV.

21 NOV.

20 NOV.

Thank you for contacting DreamHost Support. I am sorry that no one has
gotten back to you yet. We were doing some maintenance on on your mail
cluster today, this would have caused this error, but was finished up a
while ago. You should have had access a while back. I just checked over
your mail service and verified it was working.[/quote]


It seems to be going around. I’ve been dealing with this issue for about a week, and the only solution DH gave me was to change to port 587. But that doesn’t seem to help either, since my current machine is using port 587 and gets the same “The connection to the server was interrupted.” error.

Since our mail is on homemail-sub5 as well, it really looks like a cluster issue, doesn’t it?


I’m sorry to hear you’re still having trouble. :confused: I’d encourage you to submit an official support ticket for this, noting the affected emails, so we can have it on record. We’ll also be able to dig a little more thoroughly into your mail server, and update you over email if we find anything.

If you’d like, feel free to let me know your ticket’s message ID number and I can track it for you. :slight_smile: Thanks!


we have been having problems for 2 1/2 weeks. DH said to change all ports to the magical 587, then it stopped workng for some users. so last communication from DH on response to new ticket, was to keep changing ports, 25 or 587 which ever worked. most days, before 9:30 am MST sending works fine. after that, it is intermittant. my guestimate is that once all parts of the US are up and sending emails, the outgoing mail servers are getting over loaded. but it is so darn intermittant! telneting to yeilds a differnt server name each time; homie.axxxxxxxx w/ x changing depending on the computer that i am running tests from. personally i think it is a problem on DH outgoing mail server “system” but i don’t know enough about load balancing and large server farms to do more than guess.
yes this is exactly what we are experiencing. been submitted tickets almost every day.
same response i got today; load balance system. note that in system status, the maintenance to homie sub 1 was delayed due to undisclosed issues. we are on homie sub 5 as well and this is driving the staff nuts. preasure to jump ship to another provider is growing. If you get any sort of solution please post it here!


hesperus, I see your open tickets. The issue is still being investigated; you’ll get a reply from our support team when there’s more info. I apologize for the frustration - we’re working on it!


This has been going on for weeks… why no updates on Dreamhost Status Site? I can’t even email you guys @support right now. What is going on with Homie… can we all move to a different email cluster? Thanks.


[quote=“shollis, post:6, topic:60858”]
It seems to be going around. I’ve been dealing with this issue for about a week, and the only solution DH gave me was to change to port 587. But that doesn’t seem to help either, since my current machine is using port 587 and gets the same “The connection to the server was interrupted.” error. [/quote]

I’ll look into trying Port 587 myself.

I’m getting that pressure too - but I’m all the “staff” there is :slight_smile: We originally went with DH 3 years ago because our web site developer favored DH. There were those hiccups about 1-2[?] years ago, seemingly weekly (or every other) for a while. I was told to jump ship, and then it was pointed out that all this major infrastructure reinvestment might just be a good thing and would probably end up making everything more stable. After 2 years, I went on the $10.95/mo hosting because I was considering a change but then everything’s been so smooth for this past year that I’ve let it be. What I realized then – and am reminded of now – is that a problem like this can easily cost more in time, lost opportunities and grief then a few years of hosting fees put together!

I don’t know if these types of problems are just part of the way any of these domain services work. I noticed when there were probs a few years back in the status comments, people would get angry and threaten to leave and another person would reply: “try actually paying for a better quality service” or “what do you expect for what you’re paying!” But, at the same time, being ignorant of how these things work, I’m not sure where the tipping point is where you could say “OK, I can go with this service for twice or three times (or X times) the amount and really have no headaches and that fact will pay for itself easily. (If anyone does, let me know!) I’m also going to have another look at services like gmail for business. (Maybe that’s another thread). I don’t know enough to know whether these are also dependent (to whatever degree) on DH or not, but I’d love to really be sure of reliability and consistency.

Thanks! Much appreciated!

I’d love to see updates too. Communication is important so that you know what’s going on! I’m grateful though that DH provided for this thread and responded to it. At least that’s something!

Edit: I see that a few have posted similar types of enquiries or reported problems of what appear to be the same nature on Dreamhostcare twitter feed. I wonder how many people use homie-sub-5? And how many are affected, in total, by this problem?


Hi, am I glad I’ve found this thread because I am also on sub5 and am losing my mind. Outlook as well. Multiple clients, on different ISPs, one was on port 25, the other was on port 587… I’ve been having them switch back and forth, now they are both on port 587 and they are able to send mail… for now. The only other thing I was told to try was to change the port to 465 and SSL but then you have to get a nag update to verify a certificate every time you open the email client. Whether or not this does anything to fix something any better than 25 and 587 I don’t know. For now, the only thing I can really think of is to remind people that they can still use webmail, and to keep switching the ports back and forth until this is resolved but people are getting annoyed.


email blog this morning. after last evening w/ everyone in our office unable to send (they had been able to send for 2 weeks while other offices could not). this morning i can send but another office can not even after switching ports. I am spending all my time answering staff’s questions and trying to help them instead of working on other projects. At this point i would happily switch providers if i could find a stable one.
Ellie, Can you be more specific than “working on it”! please, enquiring minds want to know. i need answers for the managers at our company. I’ve found DH incredibly easy to work w/. config new accounts, lists, edit users… the control panel is a piece of cake. but we just can’t be down w/ out knowing what the schedule for fixing this is?


This is pretty clearly with one cluster. It has been going on FAR too long. Trouble tickets go ignored (at least unanswered). Live chat request rings:

“All operators are currently assisting others. Thanks for your patience. An operator will be with you shortly.”

every 2 minutes ad nauseum.

Is there not enough redundancy in the system to relocate us to another server, at this point? (admittedly, I do not know if this is as easy as it seems to be)

The owner of our company is putting on the pressure to switch providers, and when I compare prices, I find reputable hosts offering 1/2 the price we have been paying. Not looking good for DH.


Are all these problems related to Outlook or other mail clients. Thing is, I’ve tried testing out sending out mail from Outlook accounts when these problems come up and am never able to replicate the send problem which is why I’ve been confused if this is more to do with Outlook than DH. Is everyone here having this problem with Outlook?


Outlook. I thought Iphone as well, but some are telling me that is not the case. Finally got an answer through chat:

“Sorry for all the trouble with this. There are intermittent routing issues preventing you from sending mail using Outlook. The issue only seems to appear with Outlook users. I’ll have our network administrators look it over again… as a temporary fix you can try sending using webmail while we work through this.”


are we all on the most current version of outlook? my users are on 2010 professional.


We have users on Outlook 2003 through 2010 having the issue. Not sure about 2013.


Yep, our support ticket hasn’t been answered either and there’s no update to the site. Yet.

I’m coming around to thinking that maybe way to go (though I’m not qualified to implement it) is to use google apps for e-mail and keep DH just for the website itself.

So I’m reading up on whether the stability and continuity of service via dreamhost affects those who’ve switched over to Google for e-mail. But if I did switch everything, I was told to switch to Media Temple a long time ago. Heck, as I said, even something at 3 or 4 or 5x the price would’ve easily paid for itself already. How can you put a price on this if you’re having problems and have a business?


we have users on outlook 2010 to 2013. seems to be more related to the actual office location (domain), than the version. smart phones are not affected and interestingly users who connect thru the vpn are not affected. which makes me wonder if the cisco routers are also having issues dealing w/ the data? sometimes rebooting the cisco fixes the issue for a while. other times it does not.