Slashdot Mirror
Infrastructure for One Million Email Accounts?
cfsmp3 asks: "I have been asked to define the infrastructure for the email system for a huge company, which fed up of Exchange, wants to replace their entire system with something non-Microsoft. I have done this before, but not for anything of this scale. Suppose you are given a chance to build from scratch an email system that has to support around one million accounts. Some corporate, some personal, some free. POP, IMAP, webmail, etc are requirements. The system must scale perfectly, 99.9% uptime is expected... where would you start?"
I would start by talking to Kerio , their mailserver is very scaleable. www.kerio.com
I'd start by contacting people who know how to do it and can actually help you. A few responses on slashdot aren't going to help you along the entire process. Maybe even bring in a consultant.
take a look here: http://www.webhostingtalk.com/showthread.php?threa did=441925 .. the post by slidey is possibly the most useful.
I would have to say use Qmail on a freeBSD/Linux system. If you look at yahoo they have millions of email accounts and use qmail wich is very stable and very portable.
Wow, That is pretty huge scale but if Google, MSN and Yahoo have supported that many, and many more users all along open the back doors to see what they are doing? If it were me Linux obviously, Hi-Availability Clusters, some kind of solid indexing. Its still email :)
At IBM we use Lotus Notes which has saved us LOTS of virus hassles. Every employee has an account and we're something like 320,000 worldwide. The mail "databases" are spread among Domino servers but I don't know what platform these run on, or what hardware specs they have. I imagine it's either Windows or Linux... but who knows, maybe we're using some of our PowerPC-based iSeries servers. These are the boxen formerly known as AS/400.
However, I'd personally ask Google. They've done it and even their search engine has information. I found an interesting link from there detailing the deployment of a large hundred thousand user mail system, from the architecture to the software located on Linux Journal.
I've heard surprisingly good things about Communigate Pro, though I have no idea if it scales that high.
Mirapoint is probably _the_ vendor to speak to, though.
I'm sure other commercial vendors have it but I do know that large companies like ATT et al use it to handle their email. It's a shrinkwrap product that does it all and then some but it's very pricy.
I'm sure you could hack together something to do this much like what google did. Might take some time but it's totally doable.
2 years and no mod points. Join reddit. Because openness is good.
www.stalker.com
Is able to run clusters, and clusters of clusters, and theoretically scale into the hundreds of millions of accounts. Offers all the things you want, and more. LDAP, ACAP, etc, etc, integrated webmail. Intelligent directory creation structures, etc.
There are three parts to your system: sending mail, receiving mail, and storing mail. Keep them separate.
Your receivers will be a bank of servers running sendmail. They will do appropriate spam processing to reduce the amount of mail actually received. They feed the data into the storage servers.
The storage system has the data partitioned out so that all the data for one user would go to one server while all the data for another will go to a different one. The storage system also has to provide POP and IMAP access. You may want a special setup where the IMAP or POP service known which server to go to. Investigate having one giant virtual filesystem so that the system isn't too complicated.
Your webmail access will use IMAP to access the actual mail. It can be a completly different system.
The sending system will be a chokepoint for all outgoing mail. You are going to scan it as it goes out to look for virus-sent emails or unauthorized messages. For instance, you may want marketing email to be processed differently than inter-office email and such.
All of these systems will be running sendmail. I know sendmail has a bad rap for being insecure, but the insecurities have been found and since fixed. It is by far the most manageable system when it comes to large-scale deployments with heavy customization.
The radical sect of Islam would either see you dead or "reverted" to Islam.
www.myrealbox.com is a tech demo of NetMail and eDirectory.
"[We'll be] really getting inside your head and making it an unpleasant place to be" -- Trent Reznor
Here's Slidey's post. (Disclaimer: Copyright blahblahblah appropriate people yadda yadda fair use etc etc don't sue me, thank you)
---
ok i work for a large uk isp in the messaging (email) operations dept. we currently have 2.5-3 million active accounts (and a load of suspended), and manage anywhere upto 12-16million mails per day
our setup is like this (this is simplistic though):
front line - anti abuse mta's - these do dnsbl type lookups (spamcop, spamhaus and sorbs). we have 9 incoming
next we have mta's. they farm mail off to brightmail servers, which do similar to spamassassin. we have 6 incoming mtas, and 8 brightmail servers (not enough - high load)
after that they farm off to vscans (6)
after that any mail that gets through is delivered to mail stores (8 + 2 hot spares)
what you want to be doing is similar to this above - chaining hte mail from one level to the next. the first level should be the rbl's - these are less processor intensive, and can remove a fair whack of your mails in one swoop. spamassassin is going to be more cpu intensive, since it has to open each mail and read the first x many bytes
id have separate machine(s) holding your master directory, and if you can get directory caches then do that too (to take the load off the master directory) - ours run oracle
i dont know what your budget is, but split up hte different tasks as much as possible. that way if you need to add more to any pool (rbl lookups, spamassassin etc) you just add another machine..
one last thing - we also have a separate box just for postmaster mail (with exim + spamassassin funnily enough) - it tends to get busy
Last edited by Slidey on 09-08-2005 at 11:19 PM
--
(end of quote)
I know it'll get blasted, and I thought it did suck originally, but I am surprised by its scalability and reliability.
It's not free, but it's not dependent on the Linux community either. There is a concentrated and very very dedicated support and development crew. Message store size can be up to 1/3 the size of Exchange, and moving servers around is a cinch.
I'm not a Groupwise admin or anything, but I have been and Exchange guy, and I feel your pain.
Using this as a reference point (and from recommendations I've heard)...
I recommend CommuniGate.
E-Mail Server Setup Advice?
From my experience postfix scales the best for sending and receiving email. Use postfix+(mysql or ldap) + amavisd-new + clamav (or some proprietary alternernative) + spamassassin. Cyrus is probably the best for pop and imap access. Squirrelmail for webmail.
1) It'll run on anything - Win32, Linux, BSD, Solaris, x86, XServers, Alphas, Power5
2) It'll scale as big as you can dream - over 5 million accounts with clustering
3) MAPI support
My number one suggestion is hire someone who has built scalable mail systems, and written tons of code to support them: Matt Simerson
You can learn about him, and his mail projects at http://www.tnpi.biz/internet/mail/toaster.shtml
-Chris Knight
-- This sig is only a test. If this were a real sig it would say something witty. --
I once installed a 13 Million account mail system on a Linux infrastructure. As far as I know, it is working nowadays (I left that company).
The keys were:
- qmail (but postfix will work better nowadays)
- smtp (4 machines)
- pop/imap (4 machines)
- separated webmail 1 is enough (2 to high availability)
- NDS (Netscape Directory Server) which is now owned by RedHat and opensourced.
Hope that helped.
My slides relevant to this discussion can be found at http://www.shub-internet.org/brad/papers/dihses/ and http://www.shub-internet.org/brad/papers/sistpni/.
And yes, Nick Christenson has been a long-time friend and co-author of mine.
Feel free to contact me directly if you want some referrals.
Brad Knowles
http://daily.daemonnews.org/ -- if you're not
Contact IBM. A mainframe running z/VM is your solution here.
0 /
99.9% reliabilities is more then normal for those machines. It is modular enough to expand to what ever you may need in the future, and it has the dataprocessing horsepower to actually hand the 20k or so concurrent users at a time and have the harddrive space to match that many users as well.
Run linux or unix on top of VM and you should be fine.
Product Page for Z990:
http://www-03.ibm.com/servers/eserver/zseries/z99
(1) Plan an server setup which can handle the load. The requirements may change, but one million users is a fair bit. How much average incoming and outgoing emails is that? Figure that out, using a network sniffer or sniffers on existing traffic if need be (although logs should work). Then use this to calculate a number of servers needed for an outgoing smtp farm, an incoming MX farm. Figure out how much storage space is to be provided per user, and then figure out how you want that storage space to be accessible. Probably your best bet is to have a round-robin DNS farm of imap/pop servers which proxy connections based on the users login to a backend farm of actual mailservers responsible for storage. Plan the ability to move users from server to server to rebalance as needed. Outgoing smtp is a lot easier since you're not really storing things long term. Plan a web farm for webmail. (And pick software) Don't forget to plan some sort of backup, and make sure your system is flexible as far as email retention; chances are the email retention policy will change at some point and your setup should be able to change with it.
. html
(2) Test. For each server, hammer it. Test it's load under as close to real world circumstances as you can. Then create unreal punishing loads and see how it handles it. Plan in advance for how your server farm handles something like virus-generated mass emails causing 1000% spikes in load.
(3) Using your testing results, spec out the actual hardware. RAID, cheap hardware, redundancy, etc. If you have control over the network choice, plan a location with multiple fiber trunks coming into the building and provider redundancy. Remember backhoes in concert? Don't get hit by that. Plan for server failures, drive failures, network failures, power failures, and security compromises.
(4) Deploy! If you did the rest right, this is the easy part. You'll have redundant network connections, HSRP, redundant switches, a proxy farm, an imap/pop farm the proxies connect to, an smtp farm for outgoing emails, and a web server farm for serving up webmail (depending on how you choose to architect the disk space, the web farm and the pop/imap farm may be one and the same; depends on how you set things up.)
Here's a starter link to a setup which is smaller but, in principle, fairly similar:
http://www.itd.umich.edu/umce/features/2004/cyrus
Finally, if you don't want to screw it up, ask someone who has done it before. Paying someone $300/hr for a 10-30 hour review of your plan is dirt cheap compared to horking the setup. Someone who has worked in huge email environments (a la, hotmail) could show you gotchas before they bite you. (If you need help figuring out who to ask, I could even point you to some of the appropriate people)
Gee whiz... I'm surprised that the groupware is getting tossed out. If as small as 20% of the user is accustom to Outlook Calendaring, they'll represent 95% of the complaints in a new system. An advance warning to all existing account should be mailed out (both paper and email) so that nothing falls through the cracks.
Now to the mega-infrastructure that I set up for an undisclosed company for under 50K (and also didn't want groupware).
1. Transport Sender (sendmail). That's right! Good ol' plain sendmail scales. It does require some pretty savvy tweaking so get Sendmail.Com consultant onboard just for this. Use SleepyCat DB for speed for all sendmail setups. For one million, I had about 23,000 transaction per minutes during the day. You'll require 10 servers for this for cushion (against some idiots sending an ISO attachment).
2. Payload receiver (sendmail). A second group of machine to handle the reception of SMTP payloads.
3. IMAP4S/POP3S - Hey what's with the "S"? Nothing like sending your user's password in the clear. Unless you enforce VLAN in your corporate environment and limit all IMAP4/POP3 to VLAN, the "S" is a mandatory security feature, inside and outside. Guess what "S" stands for?
4. Webmail - SquirrelMail - Yet another dedicated server (in which I had to add two more load-balanced server to handling the growing pain). Use https for login only.
5. AntiVirus (ClamAV) - It was the best back then, now its just running in the middle of the pack. sendmail has milter that allows extensibility such as MIMEDeFang, wilter, rureal (reverse-DNS check), spamassasin, and SPF.
6. Support - Half the effort is put into those webpages that would 'hand-hold' these newbies into reconfiguring their machine. Worth the effort if you have over 20 expert PC users that can do their boxens. Otherwise do it yourself at each PCs. These pages should cover Thunderbird, Evolution, as well as Outlook and Outlook Express.
7. Learn to spin 11 plates, one on each pole. Keep them spinning... If they start to drop and break, bring in some more Unix dudes.
Chances are you're not going to be just turning off those Exchange servers, you're going to need to migrate the data. That being the case your going to want something with good migration tools that can handle that much migration in a relatively speaking short amount a time. I just completed an Exchange to Groupwise migration and there are some really great migration tools out there for it. Groupwise also meets all your requirements out of the box. Not to mention by buying Novell you're (at least indirectly) supporting open source. I'm not as sure about Lotus Notes, but regardless if your going to have that many users, you want big name vendor support.
If you are intererested in commercial packages, either Sun's Java System Messaging Server or Openwave's Mx product will easily scale to a million accounts and beyond. Many of the larger ISPs are using these packages or have their own custom mail server. Other possibilities may be Mirapoint(who offers an appliance type solution) or Sendmail.com
If you are into benchmarks, the folks at SPEC have published results from several packages.
My job is building systems like this. Current mailserver system I designed and built is hosting 80,000 email accounts, and will scale out to a million quite cheaply by just adding more machines.
/maildirs/domain.com/user/Maildir - split the domains up with a 2 level deep hashing algorithm (if you're virtual hosting domains, which is what it sounds like to me), so make it something like /maildirs/xx/xx/domain.com/user/Maildir, where xx/xx might be something like 3f/6b (depending on the hash). Use MD4 for the hash because its more balanced than MD5.
/var/spool/exim the internal mirrored disks. DHCP them, then all you do is plug a machine in and set it to PXE boot. Pretty trivial to do.
:)
OpenLDAP
You need a central configuration repository to store the email accounts, their passwords, etc. OpenLDAP is perfect for this, and you can replicate it out for scalability. Be prepared to learn about LDAP schemas.
Exim
Use Exim because it has a simple process model (a single binary that does all the work, like sendmail) but has a human readable configuration file and has to be the most flexible MTA out there. You will have customers with weird requirements sometimes, and Exim will be able to meet those. Plus, it has Exiscan-ACL built-in these days, which allows you to do virus scanning and spam scanning at the DATA stage, before the mail is actually accepted by the MTA. It means you can make the sending MTA deal with the bounces if the mail is a virus or is obvious spam.
Courier-IMAP for POP3 and IMAP access.
Yeah its written by a sociopath, but nothing else works as good in the field. It works out of the box with sensible LDAP schemas and is fast, reliable and secure. Handles SSL, all the different authentication methods, what have you. Maildir compatible.
Maildir message store.
Store the mail in maildirs. Don't put them in
NFS mount the maildirs from a fast NFS device like a Netapp. Netapps are recommended because you can plug them in, and they just work, plus they are easy to scale by adding more trays.
Linux NFS servers set up with heartbeat and shared disk also make a nice HA NFS, and would be cost effective, but you'll have to buy an array anyway (probably fiber channel) so it might be better just get something thats completely integrated like the Netapp.
Spamassassin.
Can be configured to scan make at DATA time in the SMTP conversation. A LOT of configuration work here to make it play nice on a massively scaled platform, but it can be done. Mostly it needs to have things like the auto whitelisting and bayseasn filtering turned off, as the extra DB file work is a bit excessive.
Actually, I'm sure there is a way to make it work with a less resource intensive repository, but using the standard SA rules seems to work well for my environment. *shrug*
ClamAV.
Free antivirus, it works, and integrates well with Exiscan-ACL. Set it up to scan via the daemon, and configure it to update every couple of hours from cron, and bob's your uncle.
Scaling out
Make every box the same. Make every box an MTA, a POP3/IMAP server, etc. Use something like Kickstart to automate builds so that you can build a machine in 10 minutes, and all you have to do is configure the IP address and plug it in. If you want to be REALLY sexy, you could make the machines boot off the network, and mount / from a shared NFS area, and make
Load balancing
Hardware load balancers are pretty much a necessity. Don't touch cisco stuff. Its not very good. Go with Foundry Networks ServerIrons. The XLs can handle 1 billion requests/day if you configure them in Direct Server Return mode (also known as DSR/Foundry switchback). Use it. It makes all the return traffic go directly out to the net, meaning your ServerIrons have to switch less traffic and track less sessions. I would recommend however for a million users a pair of the ServerIron 450GTs, or bigger. Maybe one per VIP/Service.
Now, if this is all looking pretty daunting, you could always hire me to build it for you
First, you need to start by drafting real requirements. What do you need exactly? Antispam? Antivirus? Try to have it fill up at least a page.
Once you have that done, you can start looking at solutions. You will have two parts to your solution:
1) The DMZ email relays (possibly including other antispam/antivirus functions) You really need high availability here.
2) Your email storage and retrieval systems. These may be a little more tolerant to downtime on an individual basis. But if you need to have redundancy here, there are ways to do it.
I think Hotmail did fine with BSD and Qmail.* I am sure Postfix is equally capable.
* Although Qmail itself has never had a security vulnerability discovered, you should be careful. TCPRules (on which qmail relies) has a vulnerability that can lead to root access for local users. This is not a problem on systems with no local users, however. I am not aware of any patch for the TCPRules vulnerability.
LedgerSMB: Open source Accounting/ERP
quick 15 minute brainfart:
in order to increase reliability, you want to adopt a clustered design - if a machine or two fail, nothing should happen to the service.
in order for all the machines to be able to find the user preferences/passwords/etc, you'll want some sort of common storage for them. it could be on a shared filesystem, in ldap, mysql, etc. ldap is common and a good choice (it has very fast read/query performance) - make sure you use replication so an ldap server failure doesn't take you down (or better yet, a multi-master setup). if you use ldap or sql, make sure you are indexing correctly on the data you most commonly pull up.
in order for all the machines to access the user's mail, you'll want some sort of shared message storage. a shared filesystem is easiest, you could choose from nfs, redhat gfs, veritas cluster fs, etc. if you use nfs, make sure the nfs server can failover to a backup system if the nfs master dies (netapps are great for this).
rather than using round-robin dns, i'd invest in a load balancer. there are some free options for bsd and linux, but the commercial products are very nice and easy to use. f5 labs bigips are very nice, cisco CSSes are garbage.
other suggestions about breaking the services into different groups are spot on. personally, i'd have 3-4 inbound smtp servers inside a loadbalanced pool that handled inbound mail and passed the messages to virus and spam scanning services before delivering them to the shared message store (your load might dictate you need more servers, but if you design right you can just add more as time goes on). i'd probably put pop3 and imap services on those hosts as well, and possibly only allow pop3s and imaps (the ssl encrypted varients).
i'd also have a set of outbound mail servers that users would connect to to relay outbound mail. they would require smtp auth, and possibly only allow connections on smtps ports. spam/virus scanning would be performed before the message was accepted by the server, so users would get immediate feedback if their message didn't go through. the outbounds would not do any local delivery, so they would not mount the shared message store (you'll get proper bounces for all invalid mail addresses this way, instead of smtp rejections for invalid email addresses in local domains).
i'd have another set of servers that did virus and spam scanning for both the inbound and outbound smtp servers. you'd want these machines to have faster cpus than the rest, and virus and spam scanning are usually quite cpu intensive. again, if your load increased (or was more than you had anticipated), the system is easy to grow just by adding more machines.
another set of servers would handle the shared filesystem (if nfs, or gfs exported via gnbd), and possibly also the shared preferences store (ldap).
the final set of servers would handle webmail.
each set of servers should be firewalled from the others (especially the webmail servers, which are probably the most vulnerable to attack), with only the neccessary allowed traffic going through.
qmail and postfix can easily read ldap, i'm sure sendmail can also (as can commercial solutions). anything will work for the smtp daemon.
since you are supporting pop3 users, maildir is a better choice over mailbox for your message stores. courier or cyrus would be a good choice, and come with pop3, imap, and MDA (message delivery agent) components.
i'd have the inbounds accept mail from remote sources immediately (assuming the user being delivered to was valid) and have them hand off the message to an MDA, which would perform spam scanning, virus checking, and any user filtering configured before delivering the message to the user's mailstore. (scanning after the message is accepted uses more resources, but grants you more flexibility - users can have their own spamassassin settings, or you can add any number of filtration steps).
for virus scanning, check out ClamAV. for spam scanning, look at spamassassin (
We have account data stored in an LDAP store, mirrorred to a second (read-only) store for redundancy/scaling when busy. LDAP scales wonderfully for read-heavy tasks such as this one.
As has been mentioned separately, separating recipient (edge), storage, and outbound mail servers is really important. Our edge servers perform RBL checks, greylisting (on some domains that want it), SPF (ditto), reject various attachment types, perform a reverse-MX check to try to accept from valid addresses only, and perform a recipient address check to quickly reject incorrectly addressed messages. That cuts down 80% of incoming mail (with very few false positives). Mail is then forwarded to a second set of edge servers that run SpamAssassin (set to flag spam, not stop it) and ClamAV on attachments. Finally, it goes into the storage servers. POP3/IMAP/Webmail points at the mail directories on these servers. Our outgoing servers are quite a simple setup, with SMTP Auth (also hooked to LDAP). We also have a few listservs setup, but they are a side issue.
Qmail is a bear to setup, and asking the author for advice is a good way to get flamed. Other than that, it works very well, we haven't had any security issues, and it's adequately fast - especially if you apply the "silly qmail todo" patch, fixing concurrency problems under high load. It's part of the Qmail-LDAP distribution (as is almost everything else I listed).
For servers, we use FreeBSD. I'm sure other OSes would do a fine job, but FreeBSD has been rock solid for us.
Lead developer, http://wisptools.net
Hi Cliff;
:)
;)
Sounds like a fantastic design opportunity here. The 5% of the project that is Enterprise architecture is what I enjoy the most as well. I'm assuming money probably isn't an object in terms of how much gear and bandwidth you may have to feed to this.
I'm happy to let my fingers type away below, I'd love to keep in touch and see how you end up shaping this system. my email is allowmx at hotm...
Before I ask, are there actually a million accounts? Or is that just a ceiling that you have to show proof of concept with?
I've only implemented up until about 250,000 accounts of any kind, as I'm sure you're probably aware, the base transactional resource costing is essentially the same..
For me, I would look at this for sure from at least these two angles:
1) knowing your transactional costs (how much of your hard resources, bandwidth, cpu and disk space) will each type of transaction in your system take?) I mostly use this approach to get not an exact number, but an idea of magnitude, and detail where it happens on it's own to make sure the proper attention is applied to them.
2) Failsafe intelligence & capacity in the infrastructure, as well as the failsafe intelligence & capacity in at the application layer. You have to know that your hardware, software, os, business logic and applications are all monitorable internally, externally for availabilty and actual "can I use it". Transactional logs, etc, of having information available when the inevitable problems come up.
Also, having a capacity for as many of these layers to be self-healing, and fungible to the point that your service delivery is homogenous in as many ways possible. If your network finds something doesnt work or route, with mail, you can find another way to route it. Having a transactional manager of some kind, direct or not, could be useful in this case depending on what the client wants.
99.9% uptime equates to about 526 minutes, or 87.6 hours you _could_ be down each year. Thats about 7.3 hours a month, or one day a month.
Based on that, having flexible, redundant tools setup in a high-availabily arrangement at their respective operating capacities is key. I'm not sure if your current exchange problems are being aided by not enough equipment, bandwidth, or other stability issues, so I'll just assume that it's all of them
I apologize if anyone else has already mentioned some of this, but here's some of what I've found to help me where email has become as crucial to a business as their cell phone.
On the hardware level:
- STORAGE: Everything goes on a SAN, if not more than one. Don't waste your time with anything less.
- SERVERS: All servers have redundant hot swappable parts in the very least, power and hard drives. I'd even suggest making the servers Iscsi bootable so they can boot off the backbone. Beyond this, I like to buy my servers in piles of identical ones. Have 1-2 spare serevrs of each kind sitting there, ready to throw hot swap drives into from a failed server. That way if a server dies, you can address the power supplies, or get the HD's in that machine into another identical server and get it up and running while you diagnose the hardware problem independantly. My approach to any kind of problem is FIX, DETECT and REPAIR. Get it up and running, find out what was wrong, make sure it's fixed for good. Too many of us stop at the first too
The idea I have in mind is a smaller scale of a google beige box army. linux/bsd offer so much more transcations for each piece of hardware, so that works very much in your favor. Obviously something enterprise grade to satisfy the client such as the Compaq/HP Proliants, etc. I feel these Servers ahve the best overall support, manageability and information tools, and their openlinux drivers interface wonderfully with open source operating systems)
Networking/Communication level:
- Entire mail processing architechture communi
Your point about putting more effort up-front into design is well taken, but thhat advice applies to any platform...
WIth that said, and without turning this thread into an Exchange bitchfest...
Why in the hell can't you restore a mailbox from backup using only the tools you already have if the user is no longer present in Active Directory? You can't even export the mailbox with EXMERGE... Your choices are 1) 3rd party recovery tool (like Quest Recovery for Exchange) or 2) Build an ENTIRE OTHER SERVER and do a normal, full restore of the entire mail store so you can extract one measly mailbox.
OBviously, the "Recovery Storage Group" feature is a VAST improvement over the old Exchange 5.5 way of bringing back just one mailbox (that being setup another server) but this is a MAJOR duh situation on Microsoft's part. They seem to think that since their "best practice" is to never ever erase any user account ever ever ever, that its okay to leave this gaping flaw in their enterprise groupware product. Sorry, but I think that sucks. We paid out the ass for "Enterprise" edition (to avoid the arbitrary 16gb limit on the mail store) and goddammit, I should be able to bring back a mailbox without its corresponding AD account without wasting a whole day setting up another server... I've only had to do it once (today) but the whole time I Was thinking how much esaier a mailbox restore on my OS X Server at home would be... Just restore the frickin' files and move on with your life.
Who did what now?
One million email accounts is quite a lot. You getting into the big league ISP category with something like this. It's not a one person operation to put something like this together. You're going to need a substantial number of well trained people to do this. There's only a couple players in the field at this level. Sun's JES Messaging system owns a sizeable chunk of the market, followed by OpenWave and a small gaggle of fly-by-nights with unproven track records.
Some of the larger email systems however are homegrown using open source parts. Yahoo and Google immediately come to mind, and they do work quite well. But you probably don't have the resources that they do to engineer & test something like this. Yahoo is rumored to have more than 200 people working on email alone.
Sun has a deployment like this canned, sitting on a shelf in Santa Clara. Tell them what you need, write a check, and they'll show up with the kit. 99.999% uptime if you write a big enough check. Make them to throw in the Waveset stuff.
Of course, everyone should note that recommendation is coming from an IBM employee.
Sorry, but Lotus Notes sucks; it's an abomination in almost every way. It's bloated, slow, buggy and has what is arguably the worst user interface ever (The User Interface Hall Of Shame said they could have based their entire site on this one app!) Sure, it does group meeting notes and can let you check other people's calendars but it falls flat as an email system. If it can't do the basics, who cares about the "advanced" features.
Doubt me? Okay. Let's try a little experiement.
First, sort your inbox by subject. Oh, I forgot. YOU CAN'T. Well, let me take that back. You can if you simply follow these simple instructions...
First, you need to have Domino Designer installed. In Designer, open Folders in left pane, then open folder $Inbox, highligh the Subject column. In the window with Columns properties in second tab you can check-in the "Click on column header to sort..." checkbox. Close $Inbox folder window. To prevent design refresh, in Folders view, right-click on $Inbox folder, choose Design properties and on third tab check-in "Prohibit design refresh or replace to change".
[blinks eyes in disbelief]
Un. Fucking. Believable.
Oh, and the feature I like the best is the pop-up dialog that tells you you have new mail. So you click to make that go away, switch over to LN to read the new mail and it's not there... Oh, yes, that's right, you have to press F9 to actually download the email to your client, even after being notified by an obnoxious popup that you have new mail.
Want to know another neat little feature related to that F9 key? According to our LN System Admin, get a few dozen people to all press and hold the F9 key for a few seconds at the same time and you can crash the Domino server backend requiring the server to reboot. Nice.
I could go on but I think I've made my point. I have never, ever, encountered anyone who has switched from Notes and been pleased with the change.
I've had GMail go down once or twice in the year I've had my account. Problem might be on your end. Holy frick, I've had my GMail account for a year. And three days. O.O
Information wants to be anthropomorphized!
Cyrus IMAP is designed for this size of installation. You can split the backends up with Murder on the front-ends to distribute load; divide mailboxes on each host between filesystems (which, you'd presumably spread over multiple disks); use a SAN and GFS or other shared-storage cluster filesystems and share the spool among servers; use the new pre-release 2.3 code with mailbox replication and use more discrete, commodity components. Lots of other features that are designed for large-scale implementations.
For authentication, of course you have choices among LDAP, Kerberos (both of which are usable even if you're stuck with a Windows domain for authentication), PAM and other things. Very flexible; too flexible for some and it can be a bit confusing.
I've been working on rewriting the HOWTO, although I haven't made a ton of progress, it may still be useful to you: http://nakedape.cc/info/Cyrus-IMAP-HOWTO and here's a presentation I put together for Linuxfest Northwest: http://nakedape.cc/info/Cyrus-IMAP-Intro.
You mention a million mailboxes, but that doesn't really mean much--that is just an estimate of storage requirements. What is more important to determine is how many concurrent users you will have and how much actual traffic--storage is cheap, memory not so much.
Wil
wiki
While not quite a million users, HEC Montréal switched from Netscape Messaging Server running on AIX to Postfix/Cyrus/SquirrelMail running on Linux. Linux Journal ran a really nice article and a follow-up about their transition.
One of the first things the school did was figure out how exactly their current system was failing them. Their old AIX boxes were being stressed just by the volume of mail coming through the system, they had little power left over to do any sort of filtering. This led to users getting drowned in unwanted e-mail which only exacerbated the existing load issues. This is one of the first things you need to do, figure out why your current system isn't working properly. You'll be better equipped to fix the problems when they've actually been identified.
HEC Montréal also went for heavy redundancy and specialization. Instead of a handful of servers sharing all of the tasks equally each node in the cluster has its own job with every class of job having a backup server. Every job is going to take a beating with so many users, even if only a fraction of them are using the system at any given time.
I'd say the most important part of what you're doing will be modeling your current use. Are you getting a ton of traffic from viruses and worms spreading over your internal network? Do you get huge amounts of spam traffic to users? In such cases filtering at your SMTP servers will relieve the rest of the system from extraneous traffic. While you might need really beefy external SMTP servers you won't need nearly as much storage space on a SAN or NAS.
I'm a loner Dottie, a Rebel.
Not defending Microsoft here, but I have to take care of an Exchange 2003 Enterprise server, and I wouldn't think of trying to do it without Symantec (formerly Veritas) Backup Exec with the additional Exchange agent. Yes, you can back up and restore individual mailboxes, and even individual messages. Backup Exec has its quirks, but it's the best thing going if you have to take care of Outlook users. Over the years, starting with Exchange 5.5, Backup Exec has saved my rear when information stores got corrupted, log files were deleted accidently, and so on. Combined with a nice, fast AIT tape library, it's a great data preseration product for the small- to medium-size enterprise.
It's only funny until someone gets hurt. Then, it's hilarious.
AKO (www.us.army.mil) is the Army's official intranet portal. We provide email for over 1.72M users, and we move almost 3 million messages a day. We do it all with Sun Messaging Server ver5.2 (soon to be Jes3) and we have exactly 2 (count 'em) two mail administrators. Sun mail is rock solid and scales great. We offer POP, SMTP, enterprise SPAM and Virus filtering as well as personal address books besides. We don't get the rich Outlook fat client, but then we want to be all web-based anyway. Can't say enough about Sun mail. If we had to do this with Exchange, I'd have to hire prolly 50 admins and deploy order of magnitude more machines.
No. 0.1% != 0.1
365 days * 24 hrs/day = 8760 hours per year
0.1% downtime = 0.001 downtime
8760 * 0.001 = 8.76 hrs
You're off by two orders of magnitude.
8.76 hrs / 12 months = 0.73 hrs/month = 43.8 minutes/month
One 45 minute scheduled downtime (assuming its scheduled) per month isnt terrible. It's not great, but costs really start to go up as you add nines beyond those 3.
Want a high quality FOSS RTS game? Try Warzone 2100!
Check your match before telling him he doesn't know what he's talking about.
It's 8.76 hours of outage a year.
Just an idea... if you want to go with open sources products in your company.
g fs/
First, the most important is the backend storage.
- I would try using a SAN for storage, like a small Clarion for example. I would carve the storage for the mail there on a volume.
- I would create a set of export servers that would connect directly to the SAN and re-export the volumes to a set of front end servers using a combination of gndb, gfs, etc...
See this document:
- http://www.redhat.com/magazine/008jun05/features/
- configure a set of servers that would act now as the mail servers themselves (frontends). I would strongly suggest using maildir. CourrierIMAP for the pop3/imap accounts is great. Install this on all the machines. For the SMTP agent you could use courrier but I usually prefer Exim.
- run both the IMAP/POP/SMTP servers on all the servers, using maildir only.
- use a mysql database to store the users information (passwords, email addresses, etc...). You might want to configure 2 mysql servers. One as the Master slave that will receive only the writes, and the other that would be accessed for read and balanced with the first one as reads to access user information and accounts will probably be 99% of the database activities.
- use a load balancer to put in front of all the frontend servers, do a load balancing for all the services (POP3/IMAP/SMTP) with sticky session that will try to keep the same users on the same machines when they try to download their mail.
When you are running out of capacity, simply adds new frontends, put them behind the load balancers and voila...
of course I would advise going right away with powerfull 2x3.6GHZ P4 servers and like 4GB of memory. That is powerfull and can certainely serve a LOT of users already per server.
my 2c, written quickly. I apologies if not complete but I am pretty sure the general idea is there and sound.
open to comments
Exchange 2003 - any edition. You can scavange the restored database and bind it to any account that doesn't have any exchange.. I.E. a new temporary account... RTFM!!!!!
(1st sig) If this were a snappy sig, you'd be reading it right now. (2nd sig) I'm a karma whore. >Insert FUD here
Not to be a smart ass, but it's not SLA agreement. It's an SLA. SLA stands for service level agreement. SLA agreement would be service level agreement agreement.
Simple. Its cross platform. The entire product is cross platform. Yeah, like java. Only they did it before java was a pipe dream. Late 80's.
.NET.
It has this thing called a seperation layer. All the code except the ui is the same on all the platforms. Clients used to be for os/2, mac, win16, win32, and solaris. Client side that got scalled back because nobody paid for the others -- client is win32 and mac now -- soon with code under linux as part of the next generation client. Lots of people are using on Wine.
Now, the server is still cross platform. Win32, Linux, Aix, iseries (as/400), zseries.
The problem with making something cross platform is, you don't use all the nifty little Windows specific integration and custom pretty things. You don't get something for nothing -- you have to make all those bits.
Oh, the other thing? Outlook feels integrated because everything automatically does the windows automatica launch active-x thing. Just highlight a message subjet, bingo! Embedded code launches! that's why viruses and worms.
If stuff wants to run in Notes, it has to be have a signature. OHHH, public/private key signatures and encryption. When? 1991. Hunh? Yeah, since 1991.
If something wants to run in Notes -- It need PERMISSION to run. Thus, no viruses or worms unless you're stupid enough to tell them "OK, sure, go screw up my machine".
Yes -- the development environment is weird and pretty unsophisticated. It takes a lot of time to learn because its not like other things. BUT -- I can make it do cool, secure, reliable things at a tenth of the cost you can in J2EE or MS
Excited about JSR170? Ah, me too. The Notes database internals match it almost perfectly. Domino will make a great JSR170 back end. Hell, its almost that already.
Meantime, you trolls are whining about a product that runs in Linux as a server and (using Wine) as a client. Runs on Mac. Has a fully functional JAVA environment for development and a remote API through CORBA and DIIOP. No no, instead you'll use a proprietary only -- Windows Only, Active Directory Only, Virus Distribution Engine from Microsoft.
ahahahahahaha. Enjoy it!
The problem with quotes on the internet, is that nobody bothers to check their veracity. -- Abraham Lincoln
With such a strongly worded title you'd think you'd actually have some experience to back up your claims. Memory access is faster than disk access, period. I don't care what file system you're on or what kind of caches the OS implements, fact is it's going to go to the disk almost immediately to store the change. And we're not talking about one user checking every minute. We're talking about tens of thousands of users checking every minute or few minutes. That's a continuous load on the disk - not a desirable situation for a server. Also remember that access logs are also written to disk as well.
I'm not arguing that relational db's are the way to store everything; I'm totally about the right tool for the job. But file systems are good for storing files, they're not intended for the level of data updates (new files / deleted files) that a high use email server generates. Databases are. Also disk writes from databases are also optimized if your database is well designed and resistant to paging. If you don't want a RELATIONAL database, fine. There are other types of databases you know. Mail servers don't have anything in common with file servers in terms of resource usage.
Mo it doesn't. Grep searches horribly slow. If you're sorting through 2 gigabytes of email (a fairly common amount per user in corporations), you're going to be heavily limited by the disk speed and processor time. i.e. Searches could take on the order of minutes. Not good when you want to show a list of emails and the user attempts to sort by something, or search for that email from three years ago.
Javascript + Nintendo DSi = DSiCade
Hypothetically (since nobody is dumb enough to believe this is a real life case of a million users being defined by someone betting his career on slashdot trolls)
If it were me starting from scratch -- the model for a million uses is the internet itself. SMTP, DNS, and mabe a big LDAP directory tool. For calendaring, you're SOL, but nobody calendars with a million poeple. That's meaningless. Calendaring is only useful at the workgroup level anyway. Look to any good workgroup calendaring tool and let users define thir own working groups.
Now, backing off the big million user stupid number. In the real corporate world, you have two real players and a ton of also-rans. The two real players are IBM/Lotus with Notes and Microsoft with Exchange.
The market is split roughly evenly. In the US Microsoft leads a bit, in Europe and EMEA IBM/Lotus leads. How much and actual numbers are hard as hell to track down. IBM doesn't release them and Microsoft likes to count every copy of Office as an Outlook seat. Suffice it to say both companies own about a hundred million actual users.
The basic trade off between the two - With Exchange you get tighter integration with Active Directory and smooth look and feel integration on windows. It feels like all part of the operating system. On purpose. On the other hand, you're forced to use Active Directory, forced to use Win32, and all that integration without any real security means viruses are unstopable. With Notes you get a bulky client that many users find hard to understand. You also get almost 100% prevention of virus spread (it has built in security) and other goodies. Its also a development platform and its cross platform. The client is Win32 and Mac, and users have writen howto docs for WINE. The server is linux, win32, AIX, ZSeries, and iSeries (as/400).
You may not know this, but BOTH can use the Outlook client. Yes, the outlook client is supported with a Domino mail infrastructure. Who'd have thunk it?
Oh, and Domino supports other mail clients too. Pop3, IMAP, and a very good Web Browser -- all at once for the same person if you like. Its got native SMTP support, as well.
What Notes isn't, its pretty. Most people say Outlook is prettier. Ok. Easy to do if you own the OS and make software that only runs in one environment.
So, I hear rants about Notes. I hear trolls whining about a product that runs in Linux as a server and (using Wine) as a client. Runs on Mac. Has a fully functional JAVA environment for development and a remote API through CORBA and DIIOP.
No no, instead they'll use a proprietary only -- Windows Only, Active Directory Only, Virus Distribution Engine from Microsoft.
You gotta love that. Why? Well, its pretty.
The problem with quotes on the internet, is that nobody bothers to check their veracity. -- Abraham Lincoln
I searched on Google for "email system" for "1 million users" ...
this page came up:
@Mail with large user bases
-> it even gives you a case stude of Hotmail!!!
the company is called @Mail
it is the exact same solution that seeqmail.com uses and they have over a million users.
Read it... Find out more... and Google some more
Don't pay over-priced consultants unless it is something you have absolute no expertize in. It is your job to figgure out how to get it done.
Who did what now?
GE runs Exchange. I don't know of any company that has more employees likely to use email then GE (Walmart has more employees but a LOT of them are minimum wage drones who are unlikely to need email access). If they can make it work, and work well, I don't think anyone can deny that it's enterprise ready =)
There are 4 boxes to use in the defense of liberty: soap, ballot, jury, ammo. Use in that order. Starting now.
I would disagree with the filesystem approach as high availability and disaster recovery aren't as good. If you think the relational database approach won't work, you're not using Oracle 10g RAC.
Billion row tables are common. Deletes on these tables aren't an issue as you're deleting by primary key and you can store that in an index-organized table partitioned however you want. I manage TB size database daily, I know.
Using a real relational database (not lame ass mySql) you can take advantage of all of the high availability & disaster recovery features of Oracle too, like backups, RAC and standby databases.
Incidentally, this is how the Cyrus mail server implements its single instance store.
Yes, there is. Try some kerneltrap articles to learn more about Linux (and OS internals in general
2^32 = 4 "GiB". That's all you can address with 32 bits.
On ia32, Linux allocates 1GiB of virtual address space to the kernel, and the remaining 3GiB to user space.
Thus, the maximum amount of physical memory that can be mapped to a stock ia32 kernel is 1GiB.
There are awkward things you can do at kernel compile-time to get more than 1GiB accessible to the kernel on ia32, but it's not as pretty as you seem to be thinking.
Please point to a mail system that actually uses the file system to hard link 1000 emails like the grandparent proposed.
i ew.html#singleinstanceN 824
http://asg.web.cmu.edu/cyrus/download/imapd/overv
http://doc.powerdns.com/powermail/indepth.html#AE
There are awkward things you can do at kernel compile-time to get more than 1GiB accessible to the kernel on ia32, but it's not as pretty as you seem to be thinking.
Well, if setting CONFIG_HIGHMEM=y counts as "awkward". The kernel docs say that's the correct setting for machines with between 1 and 4 GiB of RAM.
From my laptop (with 1.5GB RAM and a Linux 2.6.13 kernel, with CONFIG_HIGHMEM=y, which is actually the default setting on most distros these days):
1228596 KiB == 1199.8 MiB == 1.1717 GiB
So my kernel is currently using more than 1GiB for caching disk storage. In fact, my kernel can address up to 4GiB of RAM. I have another 1GiB DIMM on the way (which will push my laptop to 2 GiB RAM), so in a few days I'll be able to show my machine caching around 1.7GiB. (Yes, there is a reason I need 2 GiB RAM in my laptop, and it's actually not file caching).
For machines with more than 4GiB of RAM you have to use PAE. That will allow the system to use up to 64GiB of RAM, but each process (including the kernel, even though it's not really a process) can only access 4GiB. So, your argument holds some water in the case that:
In that case, the database can cache more than the kernel. I'm not aware of any database engine that has such cache daemon processes. IMO, if you're putting more than 4 GiB in the box, you should probably go ahead and buy an Opteron for it also, avoiding the whole issue.
Note to ACs: I usually delete AC replies without reading them. If you want to talk to me, log in.
Yes, passwords are transmitted in plain text. So is IMAP, and so is SMTP. You do make your users authenticate for SMTP, right? Picking another protocol will not help in this regard.
What you need to do is support STARTTLS for these protocols. That lets the client connect then negotiate an encrypted connection with the server before sending passwords. It's easy to configure the server to refuse to authenticate the client unless an SSL session has been set up if that's what your security policy dictates. It's also possible to have the server demand a client certificate from the client before setting up the SSL connection, adding an extra layer of authentication.
You'll probably also have to support the old IMAPs, POP3s, and SMTPs standards, but they should be considered deprecated and only in place for crap clients that don't know about STARTTLS.
Backups.
With POP3, the client downloads mail and deletes it off the server. Without a significantly butchered POP3 server there's no way to hold copies of that mail for a period of time (say, to ensure it goes on to your archival tapes, or to make sure you can recover files the user deleted accidentally). It's one less thing to worry about if their workstation / laptop dies, too - just give 'em another one. If more mail clients supported LDAP address books and WebDAV calendars this would be even nicer; as it is I still have to keep their mail folders in their network home dir so I can back up their address book.
You can back up POP3 boxes if you're on a corporate network, by forcing the client to keep its spools on the user's homedir. That tends to be slow and inefficient, though, and it doesn't let you do things like transparently split out attachments and store only one copy of an identical attachment for everybody.
It's also easy to lose mail with POP3 if your client does something silly. Most clients seem pretty decent now, but I remember old Eudora versions used to DELE mail off the server then crash, corrupting their mailboxes. Woohoo.
IMAP gives admins much more control over user mail. You can back up their mail folders, including their outbox and filed mail. You can enforce mail lifetime limits if your information retention policy requires it. You can store single copies of duplicate messages and attachments. You can give users access to shared mailboxes, and to each other's mailboxes where necessary. You can manage their mail folders remotely ("I can't delete $message, help!"). You can set up filters that deliver mail into sub-mailboxes automatically. Good clients automatically sync the IMAP mailbox so it can be used when the client is offline, like POP3. You can have your anti-spam software learn from their mail client's Junk folder. It's just much saner for business environments, in much the same way that network home directories and thin clients are much saner than a bunch of desktops with local storage are.
IMAP also permits you to give the user a single view of their mailboxes from their desktop and when they're on the road, or accessing their mail from home. Don't even talk about "leave mail on server" for POP3 - users WILL misconfigure it and suck all their mail down onto one of their machines, then come to you looking for help cleaning up the resulting awful mess.
Now, for an ISP, things are the opposite. You want to get the users' mail through your system and get rid of it. Most ISPs only offer POP3 and have small mailbox caps, so the user can't set their client to never delete mail off the server. They don't want to be responsible for user mail, they want it off their hands ASAP. An ISP can just tell a user who deleted a message then wants it back "well, that was silly then wasn't it?". An ISP doesn't want to back up 5 years worth of mail for 500,000 users.
My point is that for corporate environments IMAP is so superior that it's almost nuts to offer anything else, but for an ISP POP3 is a much more viable option. So what's so bad about POP3 depends entirely on what your needs are.
> > > FWIW, I've experimented with 750k mailboxes on a single system with 8GB RAM and we
> > > plan to put that number in production in a couple of months here.
> >
> > Ouch, 750k? How many concurrent accesses?
> >
>
> We currently have 1.6M, 1.2M and 940k mailboxes in 3 boxes with fiber to a single emc storage, all boxes
> dual Xeon 3.4Ghz EMT64T with 4G.
We tend to have quite large mailbox lists, but not as large as this. The biggest issues we've found with large mailbox lists are:
1. Number of concurrent connections.
If you support/encourage IMAP usage, then you tend to end up with quite a few more connections than POP.
Although technically IMAP can be very long lived, we find there are lots of short connections (mostly due to things like Outlook Express which when doing a "sync" pass does a logout and login for each *folder* in a users account!) and some long ones. With about 650,000 folders on one machine (about 130,000 users) and at peak times we see about 3500 imapd processes. We use linux 2.6, and find that this is a good number of maximum processes to have. Although the kernel is just about O(1) for everything these days, we find that there does seem to be a bit of an elbow point around the 5000 process mark where things just seem to start showing higher latency and average loads on the server
2. Size of mailboxes.db file
With a large mailbox file, you probably want to use the skiplist format. Part of the implementation of the skiplist db however is that the entire file is mmap'ed into memory. While this is generally fine since each process shares the same mmap file backing, with really large mailboxes.db files you can end up with just huge page tables.
For instance, the above 650,000 folders mailboxes.db is about 100M is size. With pages being 4k each, that means each process needs 25,600 pages just to mmap that file into it's process space. If you have > 4GB of RAM, you have to use x86_64 or PAE mode in linux. Both of these mean that each page requires a 64bit page table entry (8 bytes). If you have 3500 process then...
3500 * 25600 * 8 = 716800000 = 683M
Yes, that's 700M of memory just to hold the memory map of all your processes, no actual real data at all!!
This also means that you MUST use the high-PTE option in linux, or else you'll have lots of low memory pressure.
3. IO
CPU isn't an issue. IO definitely is. Cyrus uses minimal CPU on todays hardware, but it still is an IO hog.
That's part of the reason we sponsored the meta-data split patches that have gone into 2.3 so that you can separate out the email store part and the cyrus.* files onto separate partitions/spindles to improve overall performance. Where possible, split out:
user.seen state files
quota files
cyrus.* files
email spool files
Onto separate spindles/partitions. At least that way you'll be able to use something like "iostat -p ALL 120" to see which parts of your system are generating the largest IO.
The main challenge when I was doing it 5 years ago (I designed and wrote most of the prototype of a free webmail system, and managed the development team that completed it) was lack of good open source webmail solutions and lack of scalable mail storage systems, and hardware limitations.
Today there's a huge number of GOOD IMAP based webmail packages, such as IMP, and mail storage isn't much of a problem anymore - you can get a couple of TB of storage relatively cheaply.
Today, if I was going to do this in a corporate setting, I'd buy 3-4 small cheap servers to process inbound/outbound mail, 2-3 reasonably high powered machines with good IO capacity and RAID5 to split the users mail storage, POP/IMAP access over (IO is more or less the ONLY thing that really matters - whenever you need to make a choice, always choose higher IO capacity over almost anything else), 2 machines for an LDAP directory of which server the user is on, 2-3 cheap servers to run the web frontend on.
All in all for that kind of scale, if your total cost pans out to more than 20-30 cents per user in hardware these days you're doing something very,very wrong (and you can manage for MUCH less depending on usage patterns of your users and how much time you're willing to spend on tweaking the software).
My university also uses the Sun Messaging server. But we're only about 15000 students, so it's not a huge deal. But it works really well, at least compared to the old system with NFS-mounted mailboxes; there were constant problems with that, and it was overloaded and slow too.
It's unfortunate you got so many junk answers to your query (e.g. "resign", gmail, .mac, etc). I had a server running ~15,000 accounts on a Pentium 133 with IMail 7 a while back. It wasn't pretty, but mail got sent and received as it should.
Hula claims to scale pretty well, integrate with ClamAV and SpamAssassin, and have lots of other cool gimicks for calendars and such. For 1 million accounts, I'd get some sort of dedicated spam/virus filter, though.
Sun has the Sun Outlook Connector that allows MS Outlook to behave normally while there are Sun Messaging, Calendar, Addressbook and Directory serves instead of MS Exchange. In addition Sun has a SAFE methodology and toolkit to migrate out of MS Exchange.
If you want to guarantee anything beyond 99.0% availability, you must have complete redundancy at least 200 miles apart. This distance make all the MPLS links unusable, signifiantly increasing system complexity.
You never mentioned what your RTO and RPO were. If you can lose 24 hours worth of data, there are fairly standard methods. 12 hours is doable. Less than that and you need to spend a ton more money. SRDF/RA is interesting when you get down to the 5 minute area and don't want to write across the WAN for both locations.
Probably the easiest solution is to get 4 mainframes, 2 per site, create linux partitions on them and use some commercially supported MTA. Use all the mainframe replication facilities to do the remote replication daily.
Or you could use email like it was meant with federation and each dept or location having their own local server.
Don't forget about spam filtering, SOX compliance, and automatic encryption of external communications. IronMail merged with a PGP product can do this. The free PGP implementations make the data the individual's, not the companies. I'll just say that commercial PGP has "other solutions available" so the company still can get access to encrypted information.
When it works at all it's slow. Sometimes you can hit the Send button and just sit there and wait a while.
When we have to work on a Navy project we had to start bringing our own equipment and hubs. Even their developer machines come loaded with 10 year old software and you can't get your email and be logged in as a developer at the same time. To check mail you have to log out, log back in under a different account, then log back in as a developer. The NMCI machines are boat anchors.
NMCI is the worst defeat the US Navy has ever suffered.
That's our life, the big wheel of shit. - The Fat Man, Blue Tango Salvage
I would seriously look at Cyrus (http://asg.web.cmu.edu/cyrus/), which is designed to be scalable for huge numbers of email accounts. And the email users don't have to have accounts on the Unix boxes. It stores the messages in the file system but sets up index databases so that accessing the mailboxes is fast. It can also handle single-instance storage of the messages sent to multiple mailboxes.
Just do the math:
1 yr = 24 * 365 h = 8760 h
99.9% reliability => 8760h * 0.999 uptime = 8751.24 hours uptime or 8.67 hours downtime similarly 99.99% leads to 0.867 hours downtime = 52.56 minutes
you're off by one magnitude!