Throwing caution to the wind, I finally ran the full reverse migration with the builds. It took about two hours to successfully copy and unzip the “super_stack” build from the BigTest machine, Very nice, but then after the migration started I got the following error
Not enough space on the backup disk. Required: 569.0 GB, available: 192.0 GB
DOH! In this test environment the backup drive is only 500 GB. So, it looks like I need to expand that to match the production backup drives at about 1000 GB (948). But, even if that works, it will mean that On the production servers, I may need to delete most of the local backups to have enough room.
This leads me back to breaking up the giant build into smaller subsets. (probably the better idea).
#-----slightly off topic, read with grumpy voice-----#
Now, If I could only get the FMDeveloperTool --rename feature to work on these files I would be delighted, Although off-topic, it’s another step that I am being asked to do in these migrations (and thus I’m really looking forward to being able to do it through OttoDeploy/FMS). Alas even If I run the command on just a couple files I keep getting the unexplained Error 13.
If any one has a clue to what that might mean, I would love to know, because if it won’t work with the FMDeveloperTool directly, it won’t work when the feature is added to Otto.
This is a pretty new tool so there isn’t much community knowledge about it, yet. But if I had to guess on something that might be actionable, I’d look focus on “Outanding references”.
Maybe, just maybe, that is referring to some external file references that are still in the files that are hidden or no longer un use or just broken, and try to handle them is throwing a seg fault.
This is just a guess, but I thought it was worth mentioning.
This is what I was thinking too. As a simple test, I took my four related test files: Albert, Benny, Carrol, and Donna, and I removed Donna, thus leaving a broken reference. Unfortunately the three remaining files still got renamed. This was a very rudimentary test, so I will be investigating more.
Well, I’m continuing with my migration tests. So far I’ve had two failures, the first due to not enough room, the second because ONE file had a different encryption password. DOH!
But this is the challenge. A lot of files, and a lot of room.
The key on space seems to be to clear out Almost everything on the Backup drive. I’m not sure if our developers will be pleased with this in the production environment Fortunately, there is much off site backup too.