Should I have waited until everything was configured before doing this. If you need the details, I can provide them.
While testing, I’m assuming that once I’ve migrated everything, but not pressed the Finish Migration button, I can use NS8 as if it was fully configured system while leaving my NS7 as is.
If I want to start again, from scratch, what do I need to do at the NS7 side. Is killing the WG connection sufficient, or do I need to uninstall/reinstall the migration module. Anything else.
This is true for the “finish button” of the account provider: at that time NS7 is still running the account provider and NS8 apps use it by connecting through the Wireguard VPN.
But each NS8 app must not be touched until the “finish button” of its NS7 counterpart is pressed.
From your screenshot I suppose you started Mail configuration before you finish-ed it by NS7 side.
I tested the Mail migration with OpenLDAP provider: I found a new bug with OpenLDAP password policy in post-migration phase, but I didn’t hit your issue at all.
Press “connect to a different cluster”, then update the nethserver-ns8-migration RPM to version 1.0.5.
Ensure /etc/wireguard/wg0.conf does not exist any more.
On NS8 side you should either uninstall.sh + install.sh again, or remove all modules and LDAP providers. I’d go with the first one.
So for folks with a “production” NS7, we can’t test a migrated NS8, as it’s a one-shot deal, as the manual says that following the Finish Migration: disable the application inside NethServer 7
