Solved

Is there a way to sync two mdaemon servers for redundancy?

Posted on 2008-10-09
16
1,579 Views
Last Modified: 2008-10-21
Hi,

I need to know if there is a way to sync users, mailboxes on two mdaemon servers.

My main site uses mdaemon for mailing, we just bought another licence and installed it on a leased server on a difrent geographical location for redundancy.

I imported all users to the server, but in time it has proven difficult to update all this data manualy.

Is there a way to replicate this automaticaly?

I too need my users to start cheking their mail on the secondary server automaticaly if the main one fails.

I tried doing this though DNS, but it is just to slow, it takes a lot of time for the dns to propagate.

0
Comment
Question by:lyanez30
[X]
Welcome to Experts Exchange

Add your voice to the tech community where 5M+ people just like you are talking about what matters.

  • Help others & share knowledge
  • Earn cash & points
  • Learn & ask questions
  • 9
  • 7
16 Comments
 
LVL 31

Expert Comment

by:moorhouselondon
ID: 22681441
If you are only using one copy of Mdaemon at a time (the primary copy), the other one is there simply for mirroring purposes, and you wish to slew across to the secondary copy at any time, then you need to ensure that the mailstore on the primary copy is regularly transferred across to the secondary copy - basically this will be any changes that have occurred since the last transfer - addition of new messages, deletion of old messages.  Note that any messages accepted between the last transfer and the line going down will not be replicated across, unless that is somehow done manually - the logs will need to be analysed to see which sent messages reached their destination when the mail link went down.  When the secondary mailserver becomes the primary mailserver then presumably this state of affairs will persist until another outage occurs - so transfers will be sent from the new primary to the new secondary mail server.

You will need to make sure that all sent and received messages are stored in the mailstore.

There's probably other ways to achieve this aim, but in my view this is the easiest to understand and implement.
0
 
LVL 31

Expert Comment

by:moorhouselondon
ID: 22681520
Making sure users interrogate the up to date data is more difficult to achieve.  Doing it by changing DNS on the web could be unreliable - the change cannot be guaranteed to occur immediately - taking the secondary mail server off-line is the best way to do this.  Users using WorldClient will have a note of two (dotted IP) addresses to use - unless you register two extra domains for internal use only  which don't change IP when an outage hits e.g., company1.com company2.com.  POP3 and IMAP client software will have two accounts setup - one for each mail server.  
0
 

Author Comment

by:lyanez30
ID: 22682794
ok, but is there a way for mdaemon to do this by itself?

I mean, I dont realy care much for the messages to be on both servers, that´s the second face, what i need is that whenever there is an outage (internet), my users can still send and recive using the backup server, with no knowledge there has been an outage.

I have over 70 domains on this server, more than 500 accounts, so it is not that I´m lazy ;)



0
Is Your DevOps Pipeline Leaking?

Is your CI/CD pipeline a hodge-podge of randomly connected tools? You’ve likely got a tool to fix one problem & then a different tool to fix another, resulting in a cluster of tools with overlapping functionality. Learn how to optimize your pipeline with Gartner's recommendations

 
LVL 31

Expert Comment

by:moorhouselondon
ID: 22682941
I don't believe Mdaemon has this facility.  Have you spoken to Altn about this?
0
 

Author Comment

by:lyanez30
ID: 22683008
I do not have support anymore :(
0
 
LVL 31

Expert Comment

by:moorhouselondon
ID: 22684672
Whatever method you choose, there is a problem with failure modes.  For instance, presumably your users are attached to the primary Mdaemon server through a LAN.  If they had to attach to the secondary server due to a failure in the comms line supplying the primary mail-server, how would they do that?  If the Internet to Primary line goes down, how are you envisaging users attached to the Primary line would attach to the Secondary line?
0
 

Author Comment

by:lyanez30
ID: 22689666
Both servers are on the internet, they are not phisicaly on my location,
0
 
LVL 31

Expert Comment

by:moorhouselondon
ID: 22689936
I would say that having a backup program back up "changed" messages then transmitting it to the secondary mailstore and restoring there is your best bet.  Thought needs to be put into what happens in various failure scenarios, and action to be taken to remedy any shortcomings, particularly as it looks as if you will be reliant on DNS switching.  Some manual intervention would be inevitable, I think, to make sure that the mailstores are made identical soon after switching from Primary to Secondary.  There are too many combinations of things that can go wrong, or time taken to propagate changes through, to totally automate it.
0
 

Author Comment

by:lyanez30
ID: 22768138
Ok, I managed to get the account trouble out of the way, It happens that MDaemon has a feature that allows you to have the users account on a remote ODBC database, so I went and mounted an SQL2000 SP4 on another server, so now, I have both MDaemon's consulting the same database.

With the users out of the way, I now have only one problem:

How do I make the transition the list noticeable to my users?

I thought of using cnames on the dns, instead of usin (A) pointers, but still, I have to manually change the cnames (which is faster than changing IP's)

Kind Regards
0
 
LVL 31

Expert Comment

by:moorhouselondon
ID: 22769021
I am taking a big guess here, but would changing the SOA parameters (SOA Refresh) help with this?  Certainly my domain provider has no visible means to allow the user to change these values.
0
 

Author Comment

by:lyanez30
ID: 22769151
I have control over my DNS servers, I use windows dns server, and changing the SOA refresh value could help for the A records to refresh sooner, but still the use of aliases cnames is faster to my understanding.

Still, the main problem remains which is human intervention.
0
 
LVL 31

Expert Comment

by:moorhouselondon
ID: 22769249
>human intervention

It's been a long day, not sure I'm with you?  Are you referring to the users who are clients of the Mdaemon servers knowing which server to select in order to send out an email?
0
 

Author Comment

by:lyanez30
ID: 22769389
Sorry, you are right about the long day.

What I mean is, I do not want the clients (users of mdaemon) to know which server they are using. That is, in the event that one of the mail servers should fail they would not know about it, they can still send and recibe mail as normal.

If I have to change DNS's there would be a gap in the service, that is what I want to avoid.
0
 
LVL 31

Accepted Solution

by:
moorhouselondon earned 500 total points
ID: 22769510
I would be careful about automating that mechanism for dealing with failure though, as you could end up with a situation where some kind of "oscillation" occurs due to the overall latency in the system.  At what point do you "give up" on the mail server responding to some kind of watchdog request?  If the records take time to propagate, at what time do you say that the other mail server is "open for business"?  These watchdog events need to be done in a way that is separate from the method used to observe them.  I have thought about this situation for a couple of my clients, and my thought is that it is best to do it manually.
0
 

Author Comment

by:lyanez30
ID: 22769658
I get your point, it is kind of a guess work to know if one service is out or not just by pinging it, but you just made me aware of the oscillation problem that could occur.

I´m just curious, how does hotmail, gmail, yahoo do it?, they have a range of servers doing it right now.

Anyway, Thank you very much for your patience and time!
0
 
LVL 31

Expert Comment

by:moorhouselondon
ID: 22770207
System resilience is in a league of its own - I used to work for a company where everything had to be hot-swapable in case of failure.  Two electricity supplies, two computers, two methods of control.  Unfortunately one day the switch that switched the system from A to B (because of a failure that necessitated it) failed and the system went down.

Thank you for the points.


0

Featured Post

Free NetCrunch network monitor licenses!

Only on Experts-Exchange: Sign-up for a free-trial and we'll send you your permanent license!

Here is what you get: 30 Nodes | Unlimited Sensors | No Time Restrictions | Absolutely FREE!

Act now. This offer ends July 14, 2017.

Question has a verified solution.

If you are experiencing a similar issue, please ask a related question

Moving your enterprise fax infrastructure from in-house fax machines and servers to the cloud makes sense — from both an efficiency and productivity standpoint. But does migrating to a cloud fax solution mean you will no longer be able to send or re…
Data center, now-a-days, is referred as the home of all the advanced technologies. In-fact, most of the businesses are now establishing their entire organizational structure around the IT capabilities.
Familiarize people with the process of retrieving data from SQL Server using an Access pass-thru query. Microsoft Access is a very powerful client/server development tool. One of the ways that you can retrieve data from a SQL Server is by using a pa…
Nobody understands Phishing better than an anti-spam company. That’s why we are providing Phishing Awareness Training to our customers. According to a report by Verizon, only 3% of targeted users report malicious emails to management. With compan…

688 members asked questions and received personalized solutions in the past 7 days.

Join the community of 500,000 technology professionals and ask your questions.

Join & Ask a Question