Grizly's questions

We have an offsite hosted email server, which works (but is slow), and all our users connect to it to receive email.

Internal users connect to my postfix server to send email, I want to leave that there and create a new email server that can act as a inbound relay (Not sure if that is the correct term) so that internal users can connect to it instead of the remote one:

Ext Server: -> POP/IMAP/?? -> Int Server -> IMAP -> Users

Possible?

I'd basically like to mirror all email from the hosted server internally, so users can access it via IMAP. Allowing me to back it up! (Hosting company organized by consultant who manages it through a third/fourth/fifth??? party, gah) And filter spam/virii/etc..

Is there a way for users to log into my server, which then logs into remote server and downloads email to itself, then delivers via IMAP to them? Or am I overthinking/confusing/breaking something here?

What would happen if I sent an email to myself? Would it go out to the hosting company and then back in? Ideally only certain users would need to be on the hosted server, could I set it so that internal email gets delivered locally, but certain users get delivered to hosted server? Its 1630 on Friday, so maybe my brain is fried. I'll probably think of a hundred things wrong with this by Monday, but I want a second opinion.

Just thought of something, I can set a forwarder for all the users who are ONLY EVER INTERNAL, and get their messages forwarded to my server.. hmmm.. that could work! I'll have a play and see what happens to a few test users.

EDIT: Nope, either DNS is fubar, or the remote server won't send email back to the server that sent it.. would probably create a loop anyway. Bad idea.. :-(

The graphic (below) shows the fault I am seeing in VMWare Infrastructure Client, the drive0 light is also flashing. Not sure why, the other drive lights are green. Can't see any errors in the logs and VM's are running sweet. I got up super-early and rebooted it this morning (production box, so I'm paranoid). It came up and all the status lights went green.. but I get into the office and its back on red.. Is there anything I can do to troubleshoot this? I think reinstalling ESX might fix it, haven't tried that yet.

Main points:

• Server has 4 x 149GB SAS Drives in front hot-pluggable bays. AFAIK, no other drives.
• They are RAID5'd into one big datastore, which is reporting no errors. (See below, Logical Volume)
• The warranty has just expired.
• SAS 1078 MegaRAID controller
• Just checked physically, and the machine is reporting the following error on its tiny little status LCD: "System Software Event: Drive Slot Sensor: drive fault was asserted" Which is about as helpful as the other message. Can't seem to clear this.

I haven't found any iLO/DRAC interfaces on the back of the box, and I've only been here for a few weeks, so still struggling to adapt to ESX's weird ways. Why is IC showing 8 drive faults? I can't find anything about this online.. I thought I was good at googling!

EDIT: Now its back: (And you can see there are 4 drives in there, 1 hot-spare, didn't notice Drive0 was missing before)