On 2010-02-10 16:49, Chris Wilson wrote:
Hi Rocco,

On Wed, 10 Feb 2010, IT-Doc24 Ltd. - Rocco Radisch wrote:

  
Same here. I do not understand the issue of port 25 in conjunction with
spamming!
    
An ISP that allows port 25 out of their network, allows their clients to 
indiscriminately spam the world.
  
Agree, that is possible. But rating spam is done at the receivers site. Even if your email server's IP is as clean as a whistle you still find another email server classifying your email as spam. There is not a single solution and in most of the cases SPF doesn't work very well. If you look deep down in a antispam software's configuration, if a SPF record does not exist it will be rated with 0 or has only a slightly negative impact. So it affects mostly that user who owns a domain, the hosting provider has configured a SPF record correctly, but then a random ISP forces the user to relay through his email server. The average user won't be aware of changing domain dns settings in this situation. I assumed the latter, since Kyle's email address is with a private domain, not yahoo etc.
I am not recommending to drop SPF. Its one of the many approaches to fight against spam. I am separating what is the task of the ISP (providing internet services -> connectivity) and what is the task of a internet/email hosting provider or an email server administrator. Blocked smtp ports (25) is new to me, only occurred in Uganda yet.
An ISP that allows port 587 out, but not port 25, allows their clients to 
send mail to submission servers (e.g. their employer) but not to the 
general population. Submission often requires authentication and is not 
part of the normal mail delivery system (MX records).
  
Again, port 25 can be used for mail submission. It was redefined in the IETF but look at the exact phrase:
When conforming to this document, message submission uses the
   protocol specified here, normally over port 587.
http://tools.ietf.org/html/rfc4409

Due to compatibility most email providers allow mail submission on port 25. I personally don't know anyone who blocks email submission on port 25 as a email provider and yet most of the email clients are using that port for email submission in the standard configuration! Only Outlook 2007 implemented now an auto detection for the mail submission port.
  
If the ISP forces the user to use their email servers for email delivery:
a) causes confusion and breaks the idea behind SPF. According to the
sender policy framework you will have to add the ISPs email server in the
DNS domain settings (spf record) as Reinier has mentioned. Or you leave
out the SPF record completely, then there would be no point of having the
SPF.
    
Forwarding and mailing lists already break SPF. 
The LUG lists actually sends the message from the list's server and changes the email sender address to the [email protected]. Very friendly for the anti-spam software, thumb up. So what you are saying is not entirely true.
The only thing it's useful 
for in general is for declaring that your domain never sends email, unless 
you control all of your outbound mail servers. In this case you would not; 
also if you use a blackberry or an ISP that blocks port 25 outbound you do 
not. This is a limitation of SPF, not the fault of the ISP.
  
You can list all your outgoing email servers easily. There are different ways to do it. Not only per IP, also per MX records, subnets and combined with all of them etc. Being forced to change this settings due to a port blocking of the ISP makes SPF administration a hustle and therefore minimises the value of the SPF.
  
Anyway, which average email user knows all that?
    
Anyway, which average email user writes SPF records?
  
That was the meaning of the question. No average email users would know how to set the SPF. That is task of the hosting provider, respectively the domain's dns admin.
  
b) If the ISP's email server relays without smtp authentication THAT is
the actual evildoer. Giving spammers a free gateway to send emails
without any identification.
    
Very few ISPs allow open relaying as they would be blacklisted by 
everybody by now. Most authenticate by IP address which is a start. 
Authenticating by username makes life3 more difficult for your users, so 
is often not implemented. I agree that it would be better, but most ISPs 
don't do it, so it's hard to argue that it's evil not to.
  
Yes I know that most of them do not use open relays any more. I wanted to emphasize that that is/was the real spam distributor.
Authenticated via IP? You mean in terms of, if you are a customer of that ISP you can automatically use the email servers as relay without any authentication? What if you are connected through a public place using the ISPs network. You still cannot identify the spammer. Also, Trojans use this hole to distribute spam, assuming certain ISP provider settings. Intercepting the users smtp credentials is a bit more difficult, but also not impossible for a virus installed on the users machine. So what can you really do as an ISP? It always comes down to the receiver.

  
c) Every time the user changes the ISP/network he has to change the
settings or maintain two smtp server profiles. I.e. one uses UTL at home
and at work MTN or Orange. Each ISP forces to use his own email server.
Congratulations. Good thing we still have port 587 and 465 for SSL. I had
a number of people working at different locations with a laptop facing
this exact challenge.
    
Then stick to a single mail server that accepts email on port 587 with 
authentication and relays to everywhere. Job done. This is what most 
people do.
  
I cannot control the customer's email service provider. What if the provider only offers port 25? I only helped the customer reconfiguring their mail client to meet the ISPs and their email service providers configuration.
  
c) Most of the spam comes anyway from dynamic IP addresses of an ISP's IP
pool. In that way its even easier to distinguish the sender for the
antispam software. Like:

 pool-71-108-40-184.lsanca.dsl-w.verizon.net
551-1-60-93.w86-192.abo.wanadoo.fr
213-168-8-183-dsl.est.estpak.ee
    
Yes, these are not hard to block with Exim and then Spamhaus RBL, but a 
lot of spam comes from free email providers like google, yahoo and 
hotmail too, and that is hard to block.
  
Agree. Wanted to show that the approach of blocking smtp port 25 didn't help a lot looking at the sample dynamic IPs, so what is the point? An ISPs task is to give data connectivity.
  
Or did you ever receive a Facebook pishing attack from a Facebook server?
Very unlikely, more likely you got it from estpak.ee or similar.
    
I got my first Google Docs spam today, and I've received a lot of spam 
from people signing me up to Google Newsgroups without my permission or 
confirmation for the express purpose of spamming me.
  
You misunderstood. I even receive spams (rated as spam) from myself, how is that possible? Also mentioned on the LUG list before. Look into the email header, the actual IP of the sender's email server won't match the same IPs Facebook's email servers are using. This is just a trick to manipulate users and make them click on various links, which also look genuine. If you open thee email body (code) of the email, those links for changing the password always go to another randomly created temporarily site. There are some antispam softwares capable of detecting and rating this as a pishing attack. Some antispam softwares also detect certain reoccurring IPs and senders and put them on a temporary blacklist. Or they delay the sender on purpose, closing the connection and inform them to come back later. According to the IETF every email server has to retry within 24 hours. Most of the spam servers or spam clients do not try to send again. This technique is called greylisting and sorts out already up to 60-70% of all spams.
If you are unsatisfied with your current email service provider you can always look around and learn otherwise.
  
Plus, the whole port blocking idea of ISPs actually violates the freedom
of internet usage and doesn't really make sense either.
    
Perhaps we should ban firewalls then?
  
I was looking at the angle of an ISP. Nobody said anything about banning firewalls. Its an audacity to pay for a internet connection only working half because of restrictions. Its like leasing a car but you are only allowed to turn right on certain streets. :-)    And I am not accusing any particular ISP, just in principal. There were times when the www started where the connection was for free.
Although I support freedom of speech, I wish ISPs could be forced to 
declare which of their IP ranges belong to dynamic customers so that I can 
block them. Failing that, I wish they would block outbound port 25 
completely. Spam makes email useless. That's not freedom of speech, it's 
drowning out the useful speech, so better spam filtering means more 
freedom.
  
Again, spam rating happens on the other end. If you have issues with your current provider or an insufficient anti-spam solution you can always look for various hosting providers keeping a focus on decent anti-spam services. If you are using yahoo and other public providers, nothing to complain about since it is for free.
  
Its not my intention to start such a discussion now, the net is already 
full of these: http://torrentfreak.com/search/isp+blocking
    
Ditto, but I couldn't help replying at the risk of starting a flame war 
that I don't intend to participate in.

  
Or like Kyle did, using a random port. The port-service associations are
recommendations, nobody said we have to stick to them. The internet is
the world wide wild west.
    
It doesn't have to be random. Port 587 was allocated for exactly this 
purpose.
  
Maybe Kyle didn't know that his MSA supports port 587, then the discussion wouldn't have started at all. But that is still not the point in this discussion, it was about blocking port 25. It is a delicate subject due to the lack of preventions against spam. My point is why do ISPs block traffic.
  
Where there is a restriction you create a market.
    
Who said we can't do this in Uganda? Maybe we offer port 25 services for all UTL/MTN/Orange & co clients. For x amount $ per month.
How does that square with your comment above that "the whole port blocking 
idea of ISPs actually violates the freedom of internet usage and doesn't 
really make sense either."
  
It is complicating things, but doesn't make them impossible. I am sure those remote connectivity providers had to fight first in court until the ISP realised that would be another market. Technology develops, more bandwidth is available, but to change now how they started is too complicated for a big mobile ISP like Vodafone. They rather recommend another provider who has build a solution or a workaround.
  
Which raises another question, why do ISPs offer site-to-site 
connectivity for that kind of money here in Uganda? Calling it a 
corporate network data plan or similar, charging each remote site big 
sums per month?
    
It's easier for companies to work this way, and there's no guarantee that 
the persistent connection trick will continue to work, which means they 
could wake up to a nasty surprise one day. Some companies prefer peace of 
mind (like insurance) and pick the expensive but guaranteed VPN option.
  
Its enough to have lets say a 1 Mbit line on your headquarters office with remote/incoming traffic enabled. That has nothing to do with a persistent connection going over a middle server in order to workaround incoming traffic restrictions on provider site. Again, an ISPs main task is to provide layer one. I do understand the need for big corporates to outsource deployments of VPNs and that the next ideal partner would be there corporate ISP. But yet there is no other option in Uganda and people are miss-informed about the possibilities.
Cheers, Chris.
  
_______________________________________________
LUG mailing list
[email protected]
http://kym.net/mailman/listinfo/lug
%LUG is generously hosted by INFOCOM http://www.infocom.co.ug/

The above comments and data are owned by whoever posted them (including 
attachments if any). The List's Host is not responsible for them in any way.
---------------------------------------

Reply via email to