Only through filtering and you can only put these in the new mail folder of the other account.  You would filter on the From: field of the sender to trigger a forward to the other account.

If MercuryE is working, I suspect the ASSP logs will show the outgoing mail is bypassing ASSP.  That's not necessarily a big problem.

Your logs showed ASSP was on the same machine as Mercury - that's a setup I'm familiar with, and MercuryE will only forward through ASSP is you force all mta DNS queries to localhost and use another relay beyond ASSP.

The best place to understand more about ASSP is on its mailing list or via the wiki/forum at  .  There is a diagram if you follow the link to 'configuration'.

[quote user="kjell"]

Hi,

I just started with Mercury and installed the version 4.52 provided with win-xampp. I then added some local users and started working local. (MTA and MUA on the same machine) Everything worked fine, but I realized that the E-Mails are note deleted after poping.
 
I checked the POP-Server configuration, I included local settings:

Mark read : Y
Show read : Y
Show status : Y
No delete : N
Delete is final : Y

Everything the same. I took a look into the logs:

18:41:17.296: >> DELE 1<cr><lf>
18:41:17.296: << +OK Message deleted.<cr><lf>

But the Message still exists. The only help is to delete the .cnm-file in the user folder.

I didn't find a hint searching this forum and the internet. Has anyone an idea about my mistake?

[/quote]

This indicates that the message is marked for deletion and will be deleted if the POP3 session ends without error.  If there is an error, all these deletes will be removed ands the mailbox restored to it's original state.  Make sure that the POP3 session ends normally. 

I understand! Thanks Peter.

Graham
 

In addition, I should try to increase the NTFS journal size with the chkdsk /L:kbytes option.  You may simply be blocking too early on writes - Mercury doesn't give up a good fight easily, but a bit more breathing room for delayed transactions would surely be welcome in your situation.  (Use the /L option to find the current size.  You'll have to reboot for the change to take effect.  Try double, and watch your free disk space which you really want to provide more of if at all possible.)

Cheers,

Sabahattin

[quote user="gandalf458"]..but then I got: 553 We do not relay non-local mail, sorry.[/quote]

Check the 'Connection control; section of the MercuryS configuration.  You need to have the first two 'Relaying control' boxes checked and your internal IP addresses you want to relay from as Allow entries in 'Connection control'.  If you have php running on the same machine as Mercury you may need localhost (127.0.0.1) in there.

As Thomas says you need to upgrade you version of Mercury.  Either install the latest version in the same location (it will keep all your settings), or patch your version to 4.01c level.

Well I was unable to make it crash after that.  I tried moving emails between local and remote, (and in doing so I stumbled across the Outlook Express duplicate message issue, badly), I tried sending and receiving and moving files on the same account at the same time, hitting it with multiple users on a shared account, and everything else I could think of.

So far, I am becoming to believe it is one of two things;

a) Some kind of unexpected response from the ISP's SSL-enabled server that gums it up (because it crashes fairly consistently when it fails to deliver as described above), or

b) There is a user-detector that only makes it crash when a user is using it, which turns off when they go home and I slave away at night to try to figure out what the deal is.

Next step: 

I made arrangements with another company to let me relay mail through their SMTP servers, no SSL and no restrictions.  And I will grab the mail off the encrypted POP3 accounts with a regular client for the time being.  Thus, no SSL.  And I will see if the program is stable that way.

 I'll let you know how it turns out.  Thanks for your help.

--97T-- 

P.S.  I've given up on running a mailing list using someone else's smtp and pop servers, so I'll start a more appropriately named thread for the crashing issue.

 
P.P.S> -- For those who find themselve chasing bugs, this might be a nice diversion: 

The MX record was wrong and is now fixed. My e-mail server is working, I can send and receive, even using a DHCP address. Thanks to you guys that responded. The help was greatly appreciated!

 Cruise
 

No sorry, I haven't used Netware for many moons, but this would be my first step to rule out this as a co-incidence or pinpoint the problem.

Just to let you know, Updating the software to the latest version seem to have sorted the problem. Once again thanks for your replies  [Y]

[quote user="gideon220"]When  I test this with an email client what would be my server name?  Or how do I find out the server name that needs to be in my client?  Is it just going to be smtp.server.domain.org?  Or should i use my internet IP address?  Selecting the option in core/general did work, but does that mean that anyone can send through this server?
[/quote]

Depends on the system.  If your email client is running  on the same system as Mercury/32 you can use "Localhost" or "127.0.0.1" as the server host name.  If you're running on a different system on your local lan then you can use the internal IP address of the Mercury/32 system.  If you are running from offsite then you need to use the hostname of the router or system connected to the internet.

If your server is actually connected to the internet then yes people can connect to port 25 to use your MercuryS host to send mail.  You have to turn off relaying in the MercuryS setup. 

[quote user="jdiglidis"]

Hello everybody,
is there a way to discard messages for non local users (no alias and no local user existing)
and to send an informational mail back to the sender of this message?

Thanks for any help.

[/quote]

Nope, it's already been delivered via SMTP and so anything you do about sending the mail back to the sender will have to be outside this protocol via a separate message.  This could be done with filters from the default user account.  If this is mail sent by spammers you are confirming that the message was in fact delivered to someone.  This is especially true if you return the message with all the headers.

its a Windows 2003 R2 Standard Server

Okay, thanks for clearing that up. I was successfully able to get around the outbound mail issue by just using MercuryC with my ISP's smtp service. Looks like I will have to use an MX service like you desscribed. Curse ISP's that block port 25 from legitimate mail server usage!!!

[quote user="fxmaker"] Forgive me for being dense, but I could not find out how to make a user specific to a single domain so that U@Domain1 and U@Domain2 are different users.[/quote]

Create the local maildrops as UserA, UserB, UserC

Then add aliases, so that f.ex.
info@domain1.com == UserA
info@domain2.com == UserB

this is how we run lots of domains and aliases. People who f.ex. retire, but keep their domain, have asked us to change their addresses from john@domain.com to f.ex. john.a@domain.com and they only get emails from people they really want to get email from, without having to alter usernames or passwords.

Mercury doesn't discriminate on specific interfaces.  The greeting string will be the same, no matter what the local address of the connected socket is.  As far as I can tell, there's in fact no way to tell Mercury to do anything more than bind to specific addresses for use as server; client sockets are decided as usual by the OS according to address reachability (that's behaviour you can change in Windows itself if you've a mind, but by default it's based on the IP number of the destination).

Cheers,

Sabahattin

Thanks everyone. You have all been very helpful. It appears that the Trendnet app has a serious security flaw. It seems to be using the local host's administrative password in its authorization.

It look's like I'll have to write my own app.

I can't think why the stalling is occurring only intermittently.  It can't be a filesystem permission problem.  Perhaps the jam occurs due to networking conditions.  Have you tried to change to MercuryC (if you want to)?  Alternatively, try increasing the number of MercuryE delivery threads - say, to 50 (dangerous if the host is doing anything time-critical).  See if that helps you at all.

If the jobs in the queue have errors (QIF files), it'd be nice to know what those were - if, perhaps, many open connections are occupying system resources, we might be able to explain why restarting the server made them pass through.

Cheers,

Sabahattin

Filtering rules, the core never sees transaction filters.

You can't change the subject line, this filter only adds a header, it cannot replace a header.  With what you said you should have a new header in your message showing test: hallo.

You also cannot at this time move messages via filtering to a specific folder.  What you can do is add a specific header to the message like "SPAM: True" and then have Outlook filters looking for this specific header in the headers to move it to a folder.  This type of filtering is available in Outlook 2002 I assume this is available in your version of Outlook.