64 bit humour

32 bit computing is limited to 4GB of RAM, we all know that.

Here is 11 gig of ram...



Gig...stage...I'll get my coat.

Goodbye old friend

Michael, one of the IT team where I used to work, has left the building. Having got married last month, he boarded a plane bound for Australia on Saturday with a one way ticket.

 

I’ve barely seen him do anything but smile, I’ve never seen him badly dressed. Clever, friendly and articulate with a thoroughness that is rare. During our friendship I have seen him endure hardships that were greeted with a resolute will and he came through them well.

 

So go my friend and know that England’s loss is Australia’s gain.

Microsoft Songsmith

Go and download Songsmith from Microsoft Research at http://research.microsoft.com/en-us/um/redmond/projects/songsmith/index.html

 

It is fun and the demo version allows 6 hours of use but not 6 hours from time of install – 6 hours of actual use of the application.

 

OK, so the demo shows it being run on an Apple Mac and a number of people may not see the point or use of the software but I like the version numbering – Year.Month.Day.Build. Really simple and something everyone can understand.

 

Karaoke by email in a convenient format. Now where did I put my Elvis wig?

Exchange 2007 - Mail store considerations #2

How should you split your users inside the stores?

 

My preferred naming conventions are to call the storage groups:

·         First Storage Group

·         Second Storage Group

 

And so on.

 

The upside is that everyone can see what is going on, the folder hierarchy on disk has the same names and the documentation is easily comprehensible. Calling them Fred, Barney, Daphne and Wilma is entirely equally valid of course.

 

Consider though your recovery plan. When Exchange goes, you implement your recovery plan and assuming you don’t have the luxury of CCR with its few minutes of downtime, you need to recover from some form of backup. Backups work in alphabetical order! Therefore, in my preferred naming strategy, “Fourth Storage Group” would restore before “Second Storage Group” which caught me out first time.

 

You may (legitimately) think therefore that all key users should go into “First Storage Group” or at least the first one alphabetically. But before you implement that, think strategically for the firm. How about these group names?

 

“1 – Directors”

“2 – Senior Managers”

“3 – Middle Managers”

“4 – Everyone else”

 

In that instance, during the restore, the Directors will come back first and can get on with their work and mailboxes will arrive back in hierarchy order. Great! But if you only lose one storage group due to corruption on disk, you can bet your arse that it will be the one with all of the Directors. What happens when one Director loses his email? He goes to see his team. When they all lose email? They all come and see you.

 

Should you spread the risk? Well, that’s not one I can answer because as with so much in Exchange...”it depends”. Maybe that should be the title of my book.

Microsoft likes Apple again

Microsoft have announced that they will once again release a product for the Mac that will support Exchange Server.

According to http://tech.yahoo.com/news/macworld/20090106/tc_macworld/expomicrosofttoenhanceentourageexchangesharepointonthemac (and probably other places too), they'll be releasing a version of Entourage that uses Exchange Web Services.

Who cares? Well, Macintosh users account for about 5% of the user base although probably less when it comes to those with access to an Exchange server. Nonetheless, the media industry loves Macs and Mac users don’t like Outlook 2001 for Mac which was clunky and only ran under OS9 / classic mode.

With this announcement, Microsoft are showing that they support the Mac platform and Administrators will no longer need to use Entourage as an IMAP4 client and have to pick up the GAL over LDAP which was a pain unless you had someone incredibly knowledgeable and talented to configure it called Matt.

To support it, you’ll need Exchange 2007 SP1 or later. Eric Wilfred, the General Manager of the Macintosh Business Unit (MacBU) announced this in conjunction with the Document Collaboration Companion which integrates the Mac into Sharepoint, Document Library and Office Live Workspace.

At the moment, a Mac user has a choice of Entourage as an IMAP4 client to Exchange or Outlook Web Access Light (due to the lack of ActiveX controls on the Mac browsers). Not only will this go away with the EWS enhanced version, but there’s a public Beta this month and a full release later this year and its FREE.

Exchange 2007 - Mail store considerations #1

Well I may as well start putting #1 on these articles rather than assuming I will only do one (famous last words?).

 

“How big should my mail store be? Microsoft says 50GB for Exchange 2003 or 100GB  for Exchange 2007 and 200GB in particular circumstances.”

 

It isn’t a hard and fast rule and the real answer (unfortunately) is “It depends”.

 

Let’s go to extremes…

 

2 users with a 50GB mailbox each and you have the choice of 1 or 2 mail stores (obviously in the same storage group, see earlier article). With 1 store you are risking hitting a 100GB store size if there is no single instance effect. However, consider the process of emailing a 10MB document from one user to the other.

 

Single store. 10MB of transaction logs are generated for the email plus 10MB of store traffic / changes / increase in store size if there is no white space.

Two stores. 10MB of transaction logs are generated in EACH store plus 10MB of store traffic / changes / increase in store size if there is no white space in EACH store.

 

However, going to the other extreme, 1,000 employees with draconian 50MB mailboxes each would give you a mathematical 50GB store assuming no single instance effect. The performance hit of everyone hitting one store with one set of logs would be sizeable as would the impact of losing that store.

 

A happy and healthy compromise needs to be reached.

Eileen Brown, Google Reader and the IE RSS feed

I don’t normally pass comment on the Blogs I read but feel the need here:

 

Quote from Eileen Brown: “3: Google Reader.  I whizz through my feeds (over 500 of them) in no time. Star what i want to return to and blog it later.  Yes, i know that you’ll be thinking that i should ONLY use Microsoft products, but i like this reader MUCH more than the IE version of the RSS reader – and when we produce something that’s better – I’ll switch to that.”

OK, so Eileen is very well respected within Microsoft and is “allowed” to say that, but as one of the most read Blogs in Microsoft (last seen), that’s quite a statement and I hope the programmers take it seriously. I’ve not tried Google Reader but the IE RSS feed is a bit cumbersome.

The next paragraph doesn’t say too many complementary things about MSNBC either. http://blogs.technet.com/eileen_brown/archive/2008/12/31/final-thoughts-of-the-year.aspx

If it isn’t right, let’s say so and fix it – push people to produce great products not just adequate ones. Now, how about that UK economy Eileen, how can we fix that?

Exchange 2007 - CCR Part 3

So we know WHAT and we know WHY.

 

Let’s cover CLEVER.

 

Because the cluster is essentially running as a shipping server and a receiving server, there is no clever gadgetary in between. This not only allows the servers to be in separate data centres but also means that the performance hit of CCR is practically non existent on the hardware.

 

Obviously, because you are replaying every single transaction (and that therefore includes internal traffic and housekeeping as well as inbound and outbound traffic), you need to look at bandwidth between the servers.

 

If your existing Internet connectivity is running at 30% with your inbound and outbound email, you will double that usage figure just shipping that traffic over to your passive node. Add internal emails and mailbox level changes and pretty soon you will hit a bottleneck if you are shipping off site.

 

Failover

If you fail the nodes over, the active node becomes the passive node and vice versa. Yes, obviously! The transaction logs shipping completes and the process reverses AUTOMATICALLY so that the new active node ships its logs back to the old active node without admin intervention. As long as the firewalls in between allow the communication is both directions, it should all be seamless. Failover would typically take less than 5 minutes and so for testing purposes would be done as scheduled maintenance!

Exchange 2007 - CCR Part 2

I missed some points off the CCR article.

 

Because CCR is ASYNCHRONOUS, it means that the live server is AHEAD of the passive node.

 

As the log completes i.e. once it has completed, it is sent to the passive node. On receipt that is played forwards.

 

Upsides: Passive node doesn’t need high speed connectivity / instant response link to Active node.

Can be different hardware.

Can be lower spec hardware (terms and conditions apply!)

The Passive node does NOT need to have a copy of all of the logs.

 

Downsides: Well there aren’t any really other than needing a second server with RAID array, copy of Exchange Server and so on.

Exchange Storage Groups #3 - DR planning

Further to the last article, here is another compelling reason to use the “one store per storage group” rule.

 

It is called continuous clustered replication.

 

A lot of people break out into a cold sweat with the word “clustered” as they think of Windows clusters, complexity, Active/Active nodes and tend to think of it as black magic. Not any more.

 

Continuous clustered replication is clustering but in a totally clever way...the live email server sends a copy of its transaction log to your passive node which receives that log and plays it forward in its own copy of the mail store. Consequently it is only ever about 1MB behind the live server and is working – minimising the downtime by removing the time spent recovering.

 

That doesn’t sound like DR so why is this titled DR when it is more about continuity and high availability? Well because you are shipping logs which can be done over the Internet without the need for both servers connected to each other (quorum). So if one server burns to the ground, assuming you have prepared for it (!), your remote CCR node is not only your high availability node but is also your DR plan.

 

You *have* got a DR plan, right?

 

Exchange Storage Groups # 2 - Transaction Logs

Following on from yesterday’s article about storage groups, here is another tip.

 

Transaction logs.

 

Transaction logs are created on a “per storage group” basis. i.e. if you have 2 stores in one storage group, you will generate a very mixed bag of transaction logs that are an amalgam of both stores within the group. Therefore your TL volume is filling up at twice the rate and in the event of a database failure of one store, the play forward after a restore will not be as efficient.

 

This reinforces the reasoning for using one storage group per store. Keep it simple.

Company holidays

You are the administrator of your company Exchange Server. The office manager asks you about distributing holidays to employees to show local events such as company days out, holiday period office closures, away days and other company specific events.

 

How should you best distribute these?

 

1.    Meeting request.

2.    Outlook.hol file.

3.    As text in an email.

 

Let’s discount option 3 for a start – sending out a list of events and expecting staff to either remember them of transcribe them into email is a lot given the speed of the 21^st century.

 

Option 1 – meeting request. You create a new meeting request called “Office closed” and distribute to all staff (there should be a group for that). You set the date and time, put in your own notes about whether security will be there, who to contact in emergency, mark it as “Out of office”, set the category and hit send.

 

Option 2 – outlook.hol file. You close email and edit your outlook.hol file in notepad having spent a while looking for the file. You remember the format and remember to update the number of events in the section. You close and save the file then open Outlook to find that one event is missing because you used the UK date format of yyyy/dd/mm instead of yyyy/mm/dd. You correct this. The outlook.hol file gets sent to all staff, 50% of whom read it and ignore it, 30% of whom read it and overwrite their outlook.hol file but fail to update Outlook, 10% phone the Helpdesk to help them and the other 10% get all steps done themselves.

 

There are other ways to implement this, but a meeting request is fast and visible.

Stick of chocolate?

http://www.play.com/PC/PCs/4-/7728144/Maxell-Chocolate-AROMA-8GB-USB-Flash-Drive/Product.html

I give up. USB drives that smell of chocolate? Either desperation or alcohol must surely have been involved.

Exchange Storage Groups

This one cropped up a while ago actually but might benefit others moving to Exchange 2007 from earlier knowledge of Exchange.

 

In the olden days when the world was black and white and we weren’t in a recession, you put all of your Exchange database stores in one storage group. This was due to a memory leak when you used more than one storage group but it was addressed.

 

People who went on the training courses or read the books had it drummed into them that you filled the storage group with stores before creating the second storage group.

 

That was old school.

 

Now you fill the server with storage groups putting one store into each until you run out of storage groups before putting a second store in the first group.

 

This applies to Exchange 2003 and 2007.

 

Exchange Performance Guideline # 2

Following on from yesterday’s article about disks and volumes, here is a further gem that a minority of people are thinking.

 

Bigger is better – no, it isn’t

 

I can kind of see the thought process though. You are on the whizzy Dell site, configuring your new Exchange server and you see that the 146GB drive can be upgraded to a 300GB drive for a few extra pennies / dollars / yen.

 

If you don’t need the extra space, don’t order the drive. The analogy doesn’t work perfectly, but think about your garage – if you have a small garage you can probably find your hammer quite easily. Make the garage 10 times larger and you’ll take longer to find it.

 

Horses for courses. 10 terabytes of disk space for 100 staff isn’t always a cost effective future proofing plan.

 

Exchange 2007 Performance Guideline

This came up from a question on a forum that should be on about page 1 of the Exchange Server deployment guide but seemingly it isn’t known by everyone.

 

Disks – these are the physical drives within the server. The more you have the faster things will run – by and large.

Volumes – these are logical drives that sit on disks. You can cut the disks up as you wish into volumes.

 

Can you see a “BUT” here?

 

If you have one disk and put 3 volumes onto it – C, E and F (D is the DVD drive here) then accessing C for the operating system, E for the Exchange application install and F for the Exchange database means that the disk head has to do a huge amount of moving which will actually slow things down.

 

Rule of thumb – one volume per disk. Increase the disks not the volumes.

Exchange 2007 turbo

OK, maybe people don't know about this tool. Perhaps they think it only applies to older versions. Possibly they even feel that this tool is beneath them?

 

The fact remains that the Exchange Best Practice Analyser is one of the fundamental applications within the Exchange engineers toolbox / arsenal / portfolio.

 

It shows you very quickly and in a neat format how to cure many of the common issues that could be seriously affecting performance.

 

www.exbpa.com

Seeing in 3D

Divorced.

I was married but unhappy.

Divorce does not mean that you failed, it means that you had the courage to stop being unhappy.

Courage tempered with the strength and support of friends helped me during a time of upheaval.

 

Dad.

I reserve the right to feel the loss of not seeing my son each day because I know that I offer him greater quality time when we are together.

I reserve the right to play with him like I am his older brother, to teach him the things I learned as a child that I see were useful to me as a man.

I reserve the right to cuddle him when he cries and tuck him in at night when he is asleep.

I reserve the right to behave like a mischievous Uncle and allow him to get away with actions that I deem safe.

 

Dignified.

I will not feel wrong in moving on with my life, being happy, enjoying myself or otherwise being positive.

I will not worry over not seeing one friend when I am seeing another.

 

Divorced, Dad and Dignified. I am 3D.

 

My route to the shop

Between my office and my local supermarket...saw this and it tickled me.

Vigil in a wilderness of brain mist

Ingredients
One large adult
One small boy
One inflatable mattress
One tent

Preparation
Combine all ingredients except small boy then add small boy directly to mix every 20 minutes throughout the night.

In the morning you should have a tenderised adult and a refreshed child.

Now where's the massage area?