550 5.1.1 Recipient address rejected: User unknown in virtual alias table.

Howdy -- were you performing a migration to a new server? Or were you restoring several domains on your existing server?

What exactly occurred to your server during the restore process -- are you saying it locked up?

Are you using a dedicated server, or a VPS? Can you paste in the output of "free -m"?

Oh I understand now -- you're not saying the restore process caused the crash (which could be a sign of a RAM problem), you're saying that you're restoring due to a crash... and after performing that restore, email for some accounts isn't working properly.

Just to clarify -- the users who are having these problems -- they're able to log into your server using Usermin or some other email client... so authentication works, it's just that email being sent to them fails?

If you go into Limits and Validation -> Validate Virtual Servers, what output do you receive if you perform a validation of a domain that has an account with the problem you're describing?

Would you mind if I logged in to take a look?

If that's okay, could you email root login details to eric@virtualmin.com -- and also include an example of an account that's not working? Thanks!

I received your email, thanks.

Reviewing your Postfix config, I don't see any of the configuration data for email users in the andzyme-bioscience.com domain.

Although the system account information exists (that is, the user can login), the configuration for that user's email address isn't setup in Postfix, which would cause the problem you're describing.

That's an unusual problem though...if the email feature is included in the backup (which occurs by default), it should be restored as part of the restore process.

What you may want to do, is to attempt restoring that domain again.

When you do so, can you keep track of any output it generates, and paste that in here?

Hopefully, restoring that domain again will fix it.

But if it doesn't, seeing the output it generates may help us figure out why that's not working.

Yeah that looks correct now... restoring it again appears to have resolved the issue.

So you may need to do the same for other domains you have that are experiencing that same problem -- if you restore those backups again, that will hopefully resolve the problem for those domains too.

Hmm, although your email settings appear correct, there appears to be a problem with your DNS settings.

It appears that your server is running behind a NAT server... so the external IP is different from the internal IP (that part is okay).

However, it doesn't look like Virtualmin was configured to handle that -- so when your domains were imported into Virtualmin, the DNS for them are incorrectly set to use the internal IP, rather than the external IP.

To use NAT, what you'd need to do is click System Settings -> Virtualmin Config -> Network Settings, change "Default IP address for DNS records" to "Automatically detect external address".

Once you change that, you would then need to either re-import the DNS feature from your backups, so that it re-generates your DNS records using the correct external IP address.

To see your current DNS records, you can go into Server Configuration -> DNS Records.

When doing that now, you can see the internal IP address listed in there... you'd normally want those DNS records to contain your external IP addresses.

It doesn't appear that the MX record is returning your actual IP address yet... so there may be multiple DNS problems that need corrected.

Whenever I perform a DNS lookup on the domain andzyme-bioscience.com, the IP address being returned for the MX record (email) is 87.201.247.254.

And that IP doesn't appear to be online now -- perhaps that's your old server?

When I perform a DNS report for that domain using intodns.com, I see this information:

http://www.intodns.com/andzyme-bioscience.com

That shows that the IP addresses for your nameservers, as listed at your domain name registrar, are pointing to 87.201.247.244 and 87.201.247.245 -- is that correct?

As it's that nameserver that's returning the IP addresses, not your actual Virtualmin server.

It sounds as if you may either need to go through your registrar to correct the IP address of your nameservers, or alternatively you may need to correct the IP address information being returned by those nameservers.

The errors you mentioned in comment #17 are errors returned from Google; it appears that your server is trying to send an email to a Gmail account that may not exist. Those errors appear normal.

There was an entry in Postfix which was causing all email defined for the domain "onyxar.com" to be bounced.

However, since "onyxar.com" was also part of your server's hostname, and is used during the delivery process, that was causing email to other domains to bounce.

What I did is remove the catchall for "onyxar.com" that was causing those bounces, and once I did that, I was able to send a test message to marathonsealing.com (I deleted that test email).

The issue is with a catchall address in the domain "onyxar.com".

You can see a similar one in the domain "alahlamcharity.com", by going into Edit Aliases for that domain.

There, you can see a catchall address that bounces all email messages.

That issue would only occur when a catchall alias setup for "onyxar.com" though.

There are a variety of failover options, though setup for most of them can be fairly complex.

We can certainly discuss your options though.

Since that's a different topic than your initial email question, could you open up a new request regarding failover, and we can go over your options in that new request?

Thanks!