It appears this will only affect installations with multiple networks, via the Networks+ plugin linked to above or perhaps otherwise, where the upgrade is first executed from within a network other than the main network.
I’m not sure it’s worth fixing; it’s barely a bug and this situation is rare. It will also, only be seen by people who’ve been warned a couple times along the lines of ‘if you don’t know what you’re doing, don’t get into this’, that is first hosting a network and second, a network of networks.
I saw the issue in one of them we admin, all sites were archived on upgrade. And I thought I had tested! but must have upgraded as expected in my test. C’est la vie. Depending on your situation the fix may not be as simple as setting archived to 0 all through the blogs table. In our case, several of the networks actually had archived sites, so I referred to the db backup grabbed pre-upgrade for correct archived values.
Yeah, @dd32 it was enum and is now tinyint and I suppose that was the fail.
