cancel
Showing results for 
Search instead for 
Did you mean: 
cancel
144
Views
0
Helpful
1
Replies

CUCM 12.5 to 15.x - Data Export/Import + moving DCs + New HW

Hi All,

I have been through all the discussions on this community wrt to the migration with Data Export/Import.

I have a unique situation we have to upgrade to new hardware as well as move to brand new data centres. 

My question is and maybe I have answered myself already, when you run the "utils system upgrade dataexport initiate' command, this is a pure export of the data to be imported, it won't cause any issues if I leave the system running for a couple of days with a change freeze in place for no changes and additions on the current system while we build the new cluster and get it transported and installed in the new data centres. The IP addresses will change thus we won't have IP clashes on the new work if that is the only real reason they say in the documentation to shut down the publisher after the export and then start the new publisher and install/import.

The thinking is to stage the brand new cluster with a import, ship it off, install it and then get a change windows to change option 150 on the network and shutdown the old cluster and bring the new cluster online. Moving to BCP mode for the duration the switch over will happen and then once all phones back online, testing done move out of BCP into operational mode.

To summarize the question, I won't break the current system if I do an "utils system upgrade dataexport initiate" and the system needs to run for a couple of days still after the export? The command won't alter anything in the database causing system problem?

I should be able to do an export, do a test run in the lab and when the time come, do a fresh export and do the real install without causing downtime or system issues?

 

Best Regards
1 Reply 1

mumbles202
Level 5
Level 5

I'm fairly certain this should work.  I did something similar last year where I stood up a new Pub/Sub in the same subnet w/ different names and ip addresses and was able to change option 150 and reboot all the phones after new cluster was fully up.  Allowed us to keep the maintenance window fairly short.  This was going from 10.5 to 15.

 

If you can test that would certainly be best, but I don't remember having an issue doing what you suggested. Hopefully others will chime in w/ their experience as well.