'Integrity' is doing the right thing even when no one else is looking.
Quoted from the writings of the 'Goompahs', in "Omega" by Jack McDevit.
Monthly Archive for August, 2006
IE7 has some strange and quite annoying behaviour. After installing it and rebooting your computer, you may notice all of your fonts look fuzzy. This is because [this beta of] IE7 includes new anti-aliasing software called ClearType, which is applied not just on Internet Explorer's fonts, but on any software that uses IE to render HTML, like Outlook and Outlook Express, Windows Explorer, MSN Messenger ... etc.
Here's the annoying thing - on my monitor (and countless others - read these comments) ClearType fonts look absolutely awful. It makes me squint and actually hurts my eyes after 10 minutes of exposure to it. And, the IE installer does not ask you if you want ClearType enabled, it just installs it and you have to lump it. It's basically a system-wide change to your fonts and display options, without any notificication to the user, and with no means to configure it or switch it off in Control Panel.
But before you uninstall IE7 in a flash, you might like to know there is actually a control panel applet for tuning Cleartype, but it's a seperate download from Microsoft called the ClearType Tuner Powertoy. It allows you to optimise ClearType for your monitor, or if it really doesn't look good at all, you can just select a box to turn ClearType off.
Microsoft, if you're reading this, please make this powertoy part of the IE7 installer and run it as a wizard the first time IE7 is launched? It will save users a lot of headaches - literally.
Technorati Tags: IE7, Internet Explorer, ClearType
In Exchange 2003 it is impossible to delete a public folder store if it contains replicas. However, this can cause a problem - even if your public store has just been created and is completely empty, you can sometimes receive this error message:
You cannot delete this public folder store because it contains folder replicas. To avoid data loss, right click the public folder store and use Move All Replicas to move the replicas to a different server. It may take several hours until the content is replicated to the new server and the local replicas are removed.
Of course using the 'move all replicas' option does nothing to relieve you from this message, if your public store is empty.
To cleanly remove the store (i.e. without using adsiedit) you first need to remove 2 system generated public folders inside the store. If you expand your public folder store and click on Public Folder Instances, you will see the folders you need to remove in there - globalevents and OWAScratchPad, but, you can't remove them from this part of the interface. You will need to expand the Folder tree at the bottom of your Administrative Group and View System Folders. Right-click on the top level of the tree and "Connect to..." the server with the public store you want to delete. Find and examine the globalevents folder.
Now, you need to be extremely careful that you have the correct system folders here, because each public store contains system folders with these names. To be sure you have the right one, check the path and it should be something like:
/NON_IPM_SUBTREE/StoreEvents{BHH7CC57-934F-4CD9-ACC5-A8BC080A1423}/globalevents/
Go back to your 'public folder instances' view on the public store from earlier and check the folder properties. If the path matches this name EXACTLY, it is the right system folder and therefore OK to delete. Follow the same procedure for the OWAScratchPad folder and you should see the remaining system folders vanish, and then you can remove the store. Voila!
Technorati Tags: Exchange 2003, Public Folders, Exchange Admin
I recently added a new server to my company's org. After installing Exchange 2003 and SP2 for Exchange, I ran Microsoft update and brought everything up to date and started moving mailboxes across to it on that night. The next day I started getting complaints that OWA was not working for some users. I quickly realised the affected users were all on the new server. I checked and all of the IIS websites were configured in the same way, and the HTTP server in ESM was set up identically across all servers. After rebooting the server out of hours and resetting the IIS websites, I made a small amount of progress and via the front-end I could get it to start loading 'basic version' before giving up half way through. This is how it looked:
Not very pretty eh ... Going directly through the back-end though, it worked perfectly.
After some more investigation and following a hunch, I moved my mailbox to a test server (where my mailbox worked through OWA) and started looking at Windows updates that were available for the test server, that were already installed on the new server. Through some testing I eventually found that if KB911829 is installed on the back-end but not the front-end, it will break OWA for users on that server. The fix is simple - remove the patch, or update all the Exchange servers with the same patches.
Technorati Tags: Exchange 2003, KB911829, Exchange Admin, OWA
I use the superb free services at freedns.afraid.org to control the DNS for ukstokes.com. This is so I can point www.ukstokes.com and my blog to my webhost, and mail.ukstokes.com and other subdomains to my broadband connection at home, for hosting e-mail for my domain, and other smaller websites that don't need to be reliable.
This weekend, a botnet started attacking afraid.org and bought it and all 4 of it's name servers to their knees, bring mine and hundreds of other websites down. The attack is still going on now. I guess I can't moan too much because my blog and server at home are just for fun, but I feel sorry for the other paying, premium customers who will also have been affected by this, and who's websites downtimes can be directly translated to lost sales and revenue.
I managed to get on to the afraid.org forum for long enough to view the thread describing the attack. Josh Anderson who owns and maintains the services, posted this in the thread:
"I don't want to get too specific on the forums publicly with the fine attack details, but the attack was difficult to filter against at least by the night crew, and was about 300 mbps so the ISP would not allow it in for me to ipfw it off, because it would impact other customers, more then the network its self can handle, and they were not able to build a filter against it to scrub it out for some reason, normally they're reasonably good about it, but this has been going on for like 8 days off and on. Could be the time that it happened, since it was like in the early AM on a Sunday, maybe during business hours things would be different with a different crew, I honestly don't know.
As for retalliation, no one is worth the time it would take to bother attacking back, and other innocent networks that would probably be harmed in the process. I'm simply more exposed to attack vs them, since I actually have a human driven communitiy here. I have to accept it and try to build around it. I need to focus on the beneficial-to-me things I can do to help avoid total downtime vs detremental-to-them things I could do that just wouldn't be worth the effort.
I appreciate the kind words and helpful suggestions guys
Josh"
There's not much I can do apart from wait for the attack to finish and watch for updates on the forum. I want to keep using afraid.org, and it doesn't seem fair to move my custom elsewhere as it's not their fault that this is happening.
No doubt the source of all this is a teenage grunt on a power trip controlling an army of compromised XP computers. I suspect that if the penalties for computer crime were more severe and this sort of stuff was actually policed, he would probably be outside the reach of the law anyway due to his young age. But, somehow I can't see stuff like this EVER being policed, so something else needs to change; Windows needs to become more secure. But it's not only Microsoft's fault, the public at large need to keep their computers updated and understand what happens when they don't. We need to make things less easy for the hackers.
I dread to think how many XP computers there are connected on fast broadband connections, that are always switched on and don't have any anti-virus installed. If all of those computers were suddenly patched, had anti-virus installed and then got rebooted tonight, what would become of the hackers and their software zombies then? Would the Internet stay clean and safe for long? I doubt it. But maybe if one third of the world used Apple Mac's, one third used Linux, and the remainder used Windows, that would begin to make things tougher for them. (yeah, keep on dreaming... )
Technorati Tags: Rant, Opinion, Hackers, BotNet, FreeDNS, afraid.org
I've been asked to install Sharepoint Server 2007 beta for evaluation by part of the business. I'm also pretty interested myself to see how it integrates into Office 2007, and I'm pretty sure we have the beta of that lying around somewhere. Official documentation for Sharepoint 2007 seems to be pretty sparse (or just tricky to find) but I have found some good blog posts that describe the installation and important pre-requisite steps.
- Sharepoint Cafe - Installing a New Microsoft Office SharePoint Server 2007 Portal: Step-by-Step Instructions
- A Marvellous Point - How to: Install Sharepoint 2007 Beta1TR on a single machine
Hopefully I should have a demo of it ready for later on this afternoon, I'll post my feedback and comments as I go forward with this.
Technorati Tags: Sharepoint 2007
There seem to be 2 schools of thought on offline defrags. Some say they should be performed at regular intervals to ensure the integrity of your exchange databases, whereas others say they are a risky procedure that should be undertaken carefully and only when absolutely necessary. So, what is the recommended practice?
Event 1221 in the application log will show how much white space exists in your exchange databases. This appears once per mailbox store at the end of the online maintenance period:
Event Type: Information
Event Source: MSExchangeIS Mailbox Store
Event Category: General
Event ID: 1221
Date: 08/08/2006
Time: 06:00:02
User: N/A
Computer: S05010025
Description:
The database "1st Storage Group\Mailbox Store (S05010025)" has 39368 megabytes of free space after online defragmentation has terminated.
Personally I would not worry about a few gigs of white space as it will be reclaimed fairly quickly. As WAN and Broadband links become stronger, Exchange is more abused as a method of sending and storing large files. 40Gb of free space (as above) is worth reclaiming though. Bringing the database size down by 40Gb will significantly reduce the backup, and more importantly, restore times, and reduce the number of tapes used from week to week (and therefore the overall cost of Exchange to your company). There are a few other instances where you might carry out an offline defrag:
- When your mailbox indexing is completely hosed
- If your disks are extremely fragmented, and your server is too busy during the online defrag period to make much difference.
In my last company, users were using public folder trees as a ‘check in, check out’ document management system. This caused very heavy fragmentation of the database until we moved the folder trees to their own server which had more spindles.
Before launching an offline defrag, make sure you have a good backup of your store. Also ensure you have enough free space on your drive (110% of the size of your store). Then dismount it (there is no need to stop the information store service), and open a command prompt window to issue the commands:
eseutil /d e:\path\to\priv1.edb /s e:\path\to\priv1.stm
Note: by default it will create temp files on your C: drive and these could potentially get very large. If you don't have a large C: drive, use the /t and /f switches to specify the locations for your temporary files.
eseutil /d e:\path\to\priv1.edb /s e:\path\to\priv1.stm /t F:\temp\temp.edb /f F:\temp\temp.stm
You should see something like this:
Microsoft(R) Exchange Server Database Utilities
Version 6.5
Copyright (C) Microsoft Corporation. All Rights Reserved.Initiating DEFRAGMENTATION mode...
Database: priv1.edb
Streaming File: priv1.stm
Temp. Database: TEMPDFRG3940.EDB
Temp. Streaming File: TEMPDFRG3940.STMDefragmentation Status (% complete)
0 10 20 30 40 50 60 70 80 90 100
|----|----|----|----|----|----|----|----|----|----|
.................
Remember to mount your store again once the defrag has finished. If it has worked you should see a message like this one:
Moving 'TEMPDFRG3940.EDB' to 'priv1.edb'... DONE!
Moving 'TEMPDFRG3940.STM' to 'priv1.stm'... DONE!
Note:
It is recommended that you immediately perform a full backup
of this database. If you restore a backup made before the
defragmentation, the database will be rolled back to the state
it was in at the time of that backup.Operation completed successfully in 5210.484 seconds.
In this instance it took 1 hour and 26 minutes to defrag a 23Gb database file.
Technorati Tags: Exchange 2003, Exchange Admin
Sometimes things can go wrong during a mailbox migration which will cause a mailbox to be orphaned. Things like a network link going down or the migration hanging on "Saving changes to the directory". If you have to forcibly terminate the migration task, you may find yourself left with an orphaned (disconnected) mailbox on the destination server. Then, when you try to purge the mailbox, you may receive this message:
ID no: c1034ad6
The operation cannot be performed because this mailbox was already reconnected to an existing user.
Here are some of your options:
Try migrating the mailbox again to the same destination. Exchange should know the orphan exists from a failed mailbox move and will automatically purge it, and attempt the move again. This does not always not work though.
Another alternative, is use ESM to have a look at the mailbox store properties on the store with the disconnected mailbox. On the Limits tab, do you have a check mark next to "Do not permanently delete mailboxes and items until the store has been backed up"? If you do, try backing up the store and purging the mailbox again.
Here's the final option (which worked for me) if all else fails, and you should ensure you have a recent backup of your stores before doing this. Using ADUC, find the other mailbox in your organisation that has the same display name as your disconnected mailbox. Right click on the user and use the Exchange tasks wizard to delete the mailbox from that user. Now go back into ESM and run the cleanup agent on your stores, and both mailboxes will be in a disconnected state. You will then be able to purge the orphan you want to get rid of. Next you should simply be able to right-click on the remaining orphan mailbox in ESM and use the 'reconnect' option, and specify the user account the orphan belongs to.
Edit: One more method as suggested below is setting the "Keep deleted mailboxes" option to 0 days and then running the cleanup agent on that mailbox store. Judging by more recent comments this seems to do the trick. Thanks Mike!
Technorati Tags: Exchange 2003, Exchange Admin
Recent Comments