[Declude.JunkMail] SmarterMail vs iMail

2005-08-08 Thread Spaminator
I've long been evaluating both SmarterMail and iMail.  While offended at the 
iMail price increases, I am generally pleased with the maturity of the product 
(trying to track down the cause of mail routing failures, etc. would seem more 
difficult in SmarterMail).  However, I have not used SmarterMail in any 
production capacity under load.

Does anyone have an opinion regarding whether SmarterMail will easily handle 
800 users / 50,000 emails daily on the same dual 2.8 gHz/4gb RAM hardware that 
our iMail install currently runs on?

Declude, invURIbl, and Message Sniffer would be running on both platforms.

Thanks... 
---
This E-mail came from the Declude.JunkMail mailing list.  To
unsubscribe, just send an E-mail to [EMAIL PROTECTED], and
type unsubscribe Declude.JunkMail.  The archives can be found
at http://www.mail-archive.com.


Re: [Declude.JunkMail] SmarterMail vs iMail

2005-08-08 Thread Darin Cox
I haven't used SmarterMail beyond testing, but the main drawback I've come
across is that SMTP AUTH info cannot be passed to Declude from SmarterMail
as it is in IMail/Declude.  I understand that SmarterTools is working on
that, but until it is available it is a showstopper for us.

Otherwise I've heard good things about performance and webmail for
SmarterMail, but I understand there are a few other features missing that
may be important to you as well.  See Matt at mailpure's postings mid July
on SmarterMail shortcomings for more info.

Darin.


- Original Message - 
From: Spaminator [EMAIL PROTECTED]
To: Declude.JunkMail@declude.com
Sent: Monday, August 08, 2005 1:25 PM
Subject: [Declude.JunkMail] SmarterMail vs iMail


I've long been evaluating both SmarterMail and iMail.  While offended at the
iMail price increases, I am generally pleased with the maturity of the
product (trying to track down the cause of mail routing failures, etc. would
seem more difficult in SmarterMail).  However, I have not used SmarterMail
in any production capacity under load.

Does anyone have an opinion regarding whether SmarterMail will easily handle
800 users / 50,000 emails daily on the same dual 2.8 gHz/4gb RAM hardware
that our iMail install currently runs on?

Declude, invURIbl, and Message Sniffer would be running on both platforms.

Thanks...
---
This E-mail came from the Declude.JunkMail mailing list.  To
unsubscribe, just send an E-mail to [EMAIL PROTECTED], and
type unsubscribe Declude.JunkMail.  The archives can be found
at http://www.mail-archive.com.

---
This E-mail came from the Declude.JunkMail mailing list.  To
unsubscribe, just send an E-mail to [EMAIL PROTECTED], and
type unsubscribe Declude.JunkMail.  The archives can be found
at http://www.mail-archive.com.


RE: [Declude.JunkMail] SmarterMail vs iMail

2005-08-08 Thread Panda Consulting S.A. Luis Alberto Arango
I am currently in your same scenario.. But one step forward perhaps.

I am currently starting to test smartermail in a production enviroment (Xeon
dual 2.8 Ghz/2Gb Ram). I have studied it for a long time.

I run 350 domains with around 4500 users (25K messsages per day). Declude
junkmail, antivirus (f-prot) and sniffer (test license)

Smartermail lacks of some features compared to Imail but is has a few
interesting ones that Imail doesn't have.

Major hit is the webmail interface and administration capabilities for each
domain.

I am runinng Imail and smartermail now. Some domains in smartermail in a
production environment. I will let you know how it goes next week. If things
goes right I will migrate everything this incoming weekend.

I understand from a declude email that the largest smartermail installation
they are aware of handles 4 millions messager per day running declude
antivirus.

-Luis
 

 -Original Message-
 From: [EMAIL PROTECTED] 
 [mailto:[EMAIL PROTECTED] On Behalf Of Spaminator
 Sent: Lunes, 08 de Agosto de 2005 12:25 p.m.
 To: Declude.JunkMail@declude.com
 Subject: [Declude.JunkMail] SmarterMail vs iMail
 
 I've long been evaluating both SmarterMail and iMail.  While 
 offended at the iMail price increases, I am generally pleased 
 with the maturity of the product (trying to track down the 
 cause of mail routing failures, etc. would seem more 
 difficult in SmarterMail).  However, I have not used 
 SmarterMail in any production capacity under load.
 
 Does anyone have an opinion regarding whether SmarterMail 
 will easily handle 800 users / 50,000 emails daily on the 
 same dual 2.8 gHz/4gb RAM hardware that our iMail install 
 currently runs on?
 
 Declude, invURIbl, and Message Sniffer would be running on 
 both platforms.
 
 Thanks... 
 ---
 This E-mail came from the Declude.JunkMail mailing list.  To 
 unsubscribe, just send an E-mail to [EMAIL PROTECTED], and 
 type unsubscribe Declude.JunkMail.  The archives can be 
 found at http://www.mail-archive.com.
 __
 [Email scanned for viruses]
 [Email escaneado contra virus]
 

__
[Email scanned for viruses]
[Email escaneado contra virus]


---
This E-mail came from the Declude.JunkMail mailing list.  To
unsubscribe, just send an E-mail to [EMAIL PROTECTED], and
type unsubscribe Declude.JunkMail.  The archives can be found
at http://www.mail-archive.com.


Re: [Declude.JunkMail] SmarterMail vs iMail

2005-08-08 Thread Matt




I have actually moved my hosted E-mail over to SmarterMail despite my
displeasure with the lack of ability to block non-authenticated
messages and the way that they handled the answers. I came to the
conclusion that SmarterMail just wasn't very well set up to handle the
deluge of requests from their customers (or didn't respond
appropriately), but other companies in this space will mostly not do a
very good job themselves. For me, since I am not doing any Declude
stuff on SmarterMail, and I don't have a ton of hosted E-mail business,
I figured that I could tolerate the shortcomings for a period of time
and so I took the leap. Having scanning and hosted E-mail on the same
server presented bigger challenges for me and I need to rectify them
for QOS reasons (scanning can be bursty).

SmarterMail 2.6 is definitely a 'green' product, though certainly not
as green as it's predecessors. I would imagine that it all depends on
one's specific requirements, and how willing they might be to wait for
a new version of SmarterMail that should address some of the issues.
3.0 is rumored to start development at some point in the near future.

After moving over to SmarterMail for hosted E-mail, I started taking a
list of it's shortcomings, both in comparison to IMail, and also
generically (some of which might also be shortcomings of IMail). If I
was using Declude on my SmarterMail box, the list would be different.
There is one big issue for me with Declude and SmarterMail being that
it has no capability for WHITELIST AUTH, but the value of this will be
different for every administrator. So anyway, here's a list of things
that I have found and that people should watch out for if they matter
to them:
1) Mailbox sizes can't be locked down. Domain
admins can override the default value set by the system administrator,
so in effect there is no control over what your domain admins might set
for mail box sizes.
  
2) Built-in antispam whitelists are based on the Mail From address
matching a local user instead of something that checks to see if it was
authenticated. I consider this to be a beginner's error in spam
blocking technique and ultimately this was one of the things that
prevented me from constructing a work-around for restricting E-mail to
only what was authenticated or came from my scanning server.
  
3) Built-in antispam will count any A record returned from a blacklist
query as a positive hit regardless of the value returned.
  
4) No capability for Program Aliases.
  
5) Autoresponder can't be removed from the webmail interface. I don't
allow autoresponders from the server due to looping and backscatter
issues, and I could only break the functionality and change the label
to show that it was disabled. This has already resulted in customers
asking me to re-enable it. On IMail I was able to remove the option
entirely.
  
6) Catch-all (nobody) addresses can't be disabled from the domain
administrator's interface. I think we all know how bad catch-all's are
these days, and while the system admin's interface allows you to
disable it, it still is functional, or at least the interface to it is.
  
7) Uses a proprietary mailbox format. Mailbox files are a mix of
binary and ASCII data. This limits options when editing a mailbox
manually, and it also presents challenges for migrating to other
systems from SmarterMail. I'm not sure to this day how to approach
editing a mailbox in the same ways that I did on IMail.
  
8) IMail import utility produces inconsistent config files in
comparison to the ones generated when adding domains directly through
SmarterMail. This makes doing global search and replace operations on
the XML to fix issues following the import a delicate operation.
  
9) Logging default path is always appended with "\log". Using a
default path of "D:\" will result in logs being saved in "D:\log".
  
10) No capability to track the last login date for a user's account.
  
11) No tool for deleting messages by date similar to IMail's
immsgexp.exe. This isn't as big of a deal since mailbox files are
split into different files based on the date so scripting the process
isn't that difficult, but it's not built-in.
  
12) Webmal doesn't display message list in the same screen as the
message content. This is an important feature if you have a lot of
E-mail in an account that you intend to review in webmail as you have
to either just use the Next button or go back to the summary list after
each message view. This is a major issue for me in an otherwise nice
webmail interface. It makes spam review in SmarterMail's Web interface
completely impracticable.
  
13) Lacks a 'chaining' mechanism similar to IMail where the SMTP
service will call apps like Declude prior to handing things off to the
queue. SmarterMail relies on a configurable delay and a call to an
executable, but it doesn't manage the process. This delay is the same
for every message and so you must for instance wait for test messages
to pass through the system, and 

RE: [Declude.JunkMail] SmarterMail vs iMail

2005-08-08 Thread Markus Gufler



Excellent list, Matt.

Some of this I've allready discovered durring my 
tests.
Hopefully people at smartertools can read 
this.


At the moment I hope they will address at least the most 
important things.A wrong sorted send folder is nothing against something 
that will bring us admins critical errors or problems, especially in ISP 
enviroments.
As I've seen Smartermail seems to be working great and 
beeing developed in a more actual and future-save ".NET-style" instead of 
"CGImail" but at the moment I can't switch to Smartermail 
without
a.) keeping some of our users on IMail
or
b.) take away some features that are important for some of 
our users and also spam detection.

For example consider SMTP-AUTH:
Most of our - and I believe not only our - customers are 
using Outlook as EMail Client. 
There is an excellent test in spam filtering called 
CMDSPACE. It's very simple and resource-friendly. It's also pretty reliable and 
last but not least it's catching a significant part of all incomming 
spam.
The problem: Any message comming from one of our customers 
using MS Outlook will also fail this test. As an ISP we cannot whitelist a 
certain IP-range and need the ability to whitelist all users that has 
authenticated before sending out messages trough our server. Simple cause and - 
Ibelieve - simply to solve. Let's see what will happen with 
v3.

Markus





From: 
[EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] 
On Behalf Of MattSent: Monday, August 08, 2005 9:54 
PMTo: Declude.JunkMail@declude.comSubject: Re: 
[Declude.JunkMail] SmarterMail vs iMail

  I have actually moved my hosted E-mail over to SmarterMail despite 
  my displeasure with the lack of ability to block non-authenticated messages 
  and the way that they handled the answers. I came to the conclusion that 
  SmarterMail just wasn't very well set up to handle the deluge of requests from 
  their customers (or didn't respond appropriately), but other companies in this 
  space will mostly not do a very good job themselves. For me, since I am 
  not doing any Declude stuff on SmarterMail, and I don't have a ton of hosted 
  E-mail business, I figured that I could tolerate the shortcomings for a period 
  of time and so I took the leap. Having scanning and hosted E-mail on the 
  same server presented bigger challenges for me and I need to rectify them for 
  QOS reasons (scanning can be bursty).SmarterMail 2.6 is definitely a 
  'green' product, though certainly not as green as it's predecessors. I 
  would imagine that it all depends on one's specific requirements, and how 
  willing they might be to wait for a new version of SmarterMail that should 
  address some of the issues. 3.0 is rumored to start development at some 
  point in the near future.After moving over to SmarterMail for hosted 
  E-mail, I started taking a list of it's shortcomings, both in comparison to 
  IMail, and also generically (some of which might also be shortcomings of 
  IMail). If I was using Declude on my SmarterMail box, the list would be 
  different. There is one big issue for me with Declude and SmarterMail 
  being that it has no capability for WHITELIST AUTH, but the value of this will 
  be different for every administrator. So anyway, here's a list of things 
  that I have found and that people should watch out for if they matter to 
  them:
  1) Mailbox sizes can't be locked down. Domain 
admins can override the default value set by the system administrator, so in 
effect there is no control over what your domain admins might set for mail 
box sizes.2) Built-in antispam whitelists are based on the Mail From 
address matching a local user instead of something that checks to see if it 
was authenticated. I consider this to be a beginner's error in spam 
blocking technique and ultimately this was one of the things that prevented 
me from constructing a work-around for restricting E-mail to only what was 
authenticated or came from my scanning server.3) Built-in antispam 
will count any A record returned from a blacklist query as a positive hit 
regardless of the value returned.4) No capability for Program 
Aliases.5) Autoresponder can't be removed from the webmail 
interface. I don't allow autoresponders from the server due to looping 
and backscatter issues, and I could only break the functionality and change 
the label to show that it was disabled. This has already resulted in 
customers asking me to re-enable it. On IMail I was able to remove the 
option entirely.6) Catch-all (nobody) addresses can't be disabled 
from the domain administrator's interface. I think we all know how bad 
catch-all's are these days, and while the system admin's interface allows 
you to disable it, it still is functional, or at least the interface to it 
is.7) Uses a proprietary mailbox format. Mailbox files are a 
mix of binary and ASCII data. This limits options when editing a