Category Archives: 885

LegacyDN to the rescue

Continuing with the swing migration from the last post, all continued to go exceptionally well with the process. We kicked off an Exchange backup remotly prior to heading to site so there was less waiting around. (I use RDP to manage servers just about everyday and I am still extremely impressed with how it has made my job so much easier).


The data transfer went a treat, in fact it was all going very well until it came time to mount the Exchange databases. The databases and log files were all in the right location but when mounting the databases we got error “ID No: c1041724” which I’d seen before.


Due to not seeing this error everyday I had to rummage through my notes and do some searching online, plus referred to Jeff’s swing migration notes. The usual checks against file system permissions, database integrity etc didn’t yield any change.


It was getting late and the client wanted to go home so I said I would keep working on this remotely for them to get it going (there’s that wonderful RDP again).


One of the things about this site is the DNS domain name is VERY long such that the NETBIOS name has been truncated. Add to this the fact the original IT support person/people hadn’t done things the SBS way and we have a rather interesting environment to move from. I figured there were some naming issues at play here and looked into using LegacyDN to resolve things.


Fortunately I’d brought their old server home with me just in case (Dell 400SC with 512Mb RAM and 2 IDE hard drives!) and so fired it up to use LegacyDN to check the Exchange organisation name details. Cross checking this against the new server showed there was a mismatch.


You should refer to http://support.microsoft.com/?id=324606 for details on how to use LegacyDN and the tool itself can be downloaded from http://www.microsoft.com/downloads/details.aspx?familyid=5ef7786b-a699-4aad-b104-bf9de3f473e5&displaylang=en.


Once it’s downloaded you need to run it from a command prompt as “legacydn /forcewrite” which runs it in edit mode. Be aware that this is a powerful tool and using it incorrectly can render your Exchange environment completely unusable – you’ve been warned.


I updated the organisation name and saved the settings….now for the big test.


I went to mount the private information store database and got another error message. Dang, what was it now? Ahhh yes – I’d not checked the box to say “this database can be overwritten by a restore”. Checked the box and tried mounting again – this time it was successful. Same for the public information store.


We’ve had migrations where the database just mounted seamlessly but there are the odd times where we have to resort to additional steps. Next week we swing from SBS2000 to SBS2003 so I’m sure there will be some other issues to work with but in the case it was simply the organisation name that required massaging.


So there are several lessons here.


Firstly – the server should have been setup using the wizards in the first place. Get with the program folks – don’t go playing with building servers if you don’t know what you’re doing!


Secondly – in future I’m going to run LegacyDN as a matter of course to at least document the organisation name settings. If nothing else this saves me from having to take away the old server for reference.


Thirdly – make sure you set the right perception for the client when performing work. I always make sure I let the client know there can be issues and problems that have to be worked out with anything involving server changes. Don’t over promise or set false expectations.


Fourth – make sure you adjust the mailbox quotas to what they were before (or as agreed with the client) before you start the SMTP service. I missed doing this and they had some mail bounce before I worked out what was happening [:O]


Fifth – always, always, always make sure when making a server changeover that you have a way of connecting to the Internet for searching, IMing for help, downloading additional tools etc. SBS can be the gateway for the whole network and if it’s not fully operational such that Internet access if not available then you can get very stuck. I fortunately had my 3G card with me so my notebook had a connection, plus I tend to carry a heap of resources with me but it’s worth noting this anyway.



Perhaps it’s time I starting writing a Tips & Tricks book?

Following the Feb SBS users group meeting…

We had a great meeting the other week and covered a lot of ground. The PowerPoint deck will shortly be available for download on the www.sbsusers.net site (go to the meetings page).


We had a discussion about some issues being experienced with Backup Exec on SBS. Andrew (one of my team) has covered this ground several times and came up with the goods – in particular how to get jobs that get to 99% to complete successfully rather than just sit in a pending state.


Turns out David from the Sydney SBS group has the same frustrations, so for all of you with Backup Exec woes, get ready to smile…


How to configure VERITAS Backup Exec ™ to automatically respond to media alerts, including when a backup hangs at 99%, until the user responds to a tape eject prompt


Identified by looking in Event Viewer – Applications: Event ID 58061 is shown


http://seer.support.veritas.com/docs/192055.htm


That should do it! And for some more little gems Andrew has found…


OFO: Initialization failure on: “Shadow?Copy?Components” Advanced Open File Option used: Microsoft Volume Shadow Copy Service (VSS). VSS Snapshot error. Unexpected provider error.


Identified by looking in Event Viewer – Applications: Event ID 34113 is shown. Often also associated as being a METADATA error


http://seer.support.veritas.com/docs/265927.htm


And be aware of ports in use when installing Backup Exec 10


After a reboot or restarting the Backup Exec services, the Backup Exec Server Service fails to start with the following error: “An internal error (-14) occurred in object 10.”


Identified by looking in Event Viewer – Applications: Event ID 57802 is shown.


http://seer.support.veritas.com/docs/275017.htm


You will need to use netstat –a to find what ports are open


Refer to “netstat /?“ for full usage details




Another topic discussed was clearing out badmail folder entries where there are HEAPS of them present. Opening the folder in Windows Explorer, or even a command prompt, can be unsuccessful when there are thousands of entries. So, Microsoft have a VBscript to deal with this (again thanks to Andrew for finding this one). Smply visit http://support.microsoft.com/?kbid=867642


Finally, if you find an Exchange server that’s being used as an open relay and want to get it cleaned up, visit http://support.microsoft.com/?kbid=324958


Of course you’re better of not being an open relay in the first place.


Right – time to get back to work. Oh, that reminds me, details for the March SBS users group meeting need to be posted here…stay tuned…