Migration to a new server

80 pts.
Server migration
We recently replaced and upgraded to a new server (Dell PowerEdge T410 Server). We began the migration on the evening of May 11th. Over one month later we are still not fully migrated and are running both the old and new servers in tandem. We have had continuous response and file access issues for the past month. I think this is unacceptable. Am I being unreasonable? In my past experience normally a server migration is done in one weekend completely. Granted I work for a medium size law firm with a lot of data. But, I still think it should have been done in one weekend. Am I being unreasonable?

Software/Hardware used:
Dell PowerEdge T410 Server - Microsoft Small Business Server Std 2011 with 15 CALs

Answer Wiki

Thanks. We'll let you know when a new response is added.

You are definitely not being unreasonable.

Better practitioners do lots of research before performing a migration (with the caveat that sometimes migrations are mandated by “corporate,” giving little time for local administrators to catch their breath before launching).

Once it has been decided to perform the migration, administrators should write up a migration plan of one and document it to the Nth degree. Not before the test case has been written, administrators should migrate one (1) test subject. This is a great learning opportunity during which the administrators should make a note of everything that deviates even a little from the test plan.

In most cases, the migration test case should be conducted on a dummy asset created specifically to fully stress the test plan.

It may prove quite beneficial to log times of completion. If it takes 20 minutes per mailbox, for example, you can establish up front that 200 mailboxes will take 4000 minutes (4200minutes * hour/60 minutes = 70 hours) You can report to your boss(es) that, “The migration will take approximately seventy hours.” Maybe you want to break the migration up into smaller increments.

After the pilot migration, the full migration plan should be updated with what has been learned during the test case. A second test case of one may be called for.

Success! Once the first test case has been completed, you are ready for piloting. This entails taking a small number of employees, typically consisting of mostly the IT department, and migrating them. The pilot migration should be conducted exactly in accordance with the piloting plan, which must be fully documented before commencing.

NOTE: Have a roll-back plan that has been tested during the test case.
NOTE: Have a roll-back plan that has been tested during the test case.

At the conclusion of a successful pilot group migration, evaluate again how their migration diverged from the pilot migration plan and re-write the migration plan to accommodate new data.

Only now are you ready for full migration. See how much competency your people have built doing the test case, roll-back, and pilot migration?

The final migration should have few hiccups. It should not need to be rolled back.

Finally, the migration should take not much longer than the prediction. So, the question for you is, what was the prediction provided to the working staff and is what you are observing consistent with the prediction?

If you are working without a prediction, someone either didn’t perform the above steps or they failed, neglected, or declined to inform the people affected by the migration, which is a sign of unprofessionalism.

Discuss This Question: 4  Replies

There was an error processing your information. Please try again later.
Thanks. We'll let you know when a new response is added.
Send me notifications when members answer or reply to this question.
  • carlosdl
    I think you are right. One month is way too much. If you found issues during the migration process, I think you should have cancelled it, and only try the migration again after all of them have been investigated and corrected. Sometimes it is better to go back to the old server, and re-format the new one, and perform a clean install of everything again, instead of trying to fix many individual issues on the fly.
    85,885 pointsBadges:
  • LeChat
    Thank you so much for the information from both responses. I will present this to the senior partners of our firm. The person doing the migration was doing the best she could. But, apparently we are suffering the consequences of a lack of well thought out plan.
    80 pointsBadges:
  • TomLiotta
    But note that we don't know the status of the previous server environment, the problems that have been encountered during migration, the restrictions placed upon resources, the allocation of appropriate resources to the task, the application requirements for which the server is responsible nor anything really useful. We have no idea what the "server" is serving. Was it a SQL Server server? Did it serve Active Directory? Was it an Exchange server? An application server? A domain controller? DNS? (All of the above? More than the above?) How did client systems rely on it? What effect does the new server have on client systems? What are the client systems? We can say that migration shouldn't take a month. We cannot say that a month is unreasonable in this case. Tom
    125,585 pointsBadges:
  • LeChat
    Tom, indeed you raise very valid questions regarding the server. I don't think it's too complex of a scenario. We do host our own email (not sure why) with Microsoft Exchange and have Active Directory. We do not host our website. Basically we consist of about six attorneys, paralegal, secretary and a bookkeeper. We specialize in contract law, which means we host volumes of agreements, contracts....lots of data. That is probably the most important thing to note. We do a lot of scanning of documents. The fact is that the support company that we use is fairly responsive. However, they obviously had no documented plan to do the migration. Shame on us.... here we are a bunch of lawyers that permitted this to happen. We are slowly resolving the issues that continue to haunt us. The person that did the migration went on vacation within a week of doing the migration. Her backup was clearly not interested in accepting responsibility for the fall-out. We are beginning to recover. Thanks for the information...
    80 pointsBadges:

Forgot Password

No problem! Submit your e-mail address below. We'll send you an e-mail containing your password.

Your password has been sent to:

To follow this tag...

There was an error processing your information. Please try again later.

Thanks! We'll email you when relevant content is added and updated.


Share this item with your network: