There is also the option of an error “in between”, also known as PEBCAC:
“Problem exists between Console and Chair”, or simple human error, typos, not reading the Fine manual (RTFM)…
A defective LAN cable has also been guilty in countless migrations!
Note:
This is not intended as a sidekick to the excellent work of @Stephdl or @NLS in debugging NS8 migration!
PEBCAC is always a factor, speaking from my 40 year computing experience of which 25+ are professional, although I don’t love it if someone rubs it on my face @Andy_Wismer …
To be honest I find it weird that the migration fails in a scenario as simple as mine, I suspect way more complex and fully utilized NS7 than mine, that… worked? (or just not migrated yet)
I assure you the network connection between the two VMs on the same physical machine works fine and the server is on a 10Gbit switch anyway if that was a factor.
I did specifically write it’s NOT intended as a sidekick against either you or Stephan.
And I did write “excellent work”, and that’s meant that way. I am for anything that forewards NS8!
Besides which: As you’re not even using Roundcube, yet the migration barfs due to Roundcube, I can’t really blame you for this issue in any way! I see you as helping solve a bug you found, excellent work!
But if you find an offence in any of the above…
My actual intention was to remind Stephan that there are not only programmers at fault when something goes south. Often, there are “unexpected” issues in any migration.
Ever ran into a 4 hour long, City wide Blackout during. right in the middle of a migration? Longer than our UPS would support, and all due to a Squirel which jumped in the wrong hole in the Substation responsible for the City of Zurich. A few years back, on a Saturday before xMas in December! (True).
The squirel made national news - as a pieve of Grilled meat (Not anymore identifyable). And I had a fried RAID…
And no, I’m not blaming you for training that squirel!
But as you also know from experience in IT: Sh*t happens!
No worries man.
I was not offended as much as you think.
Anyway…
Backed up my NS7.
Started from clean NS8 (where by “clean” is an image already set up and just waiting for an NS7 to connect to it and migrate).
Installed new core update to NS8.
Installed the test migration version above (which btw is still 1.07 like the one that was installed) in my NS7. Took more time than expected but finished.
…starting migration…
Shows Mattermost, Email, Webtop, Ejabberd, POP3/IMAP Connector, Local Active Directory. I only care about email, webtop, connector, AD, so I “skip” the other two.
Press “start” on Email.
I see various notifications in NS8, one “failed” (remove external domain, validation error …which always failed before too).
I “sync data” once…
…takes quite some time, “migration sync” always shows 0% (never seen it move even in the previous attempts, even when it succeeded), I don’t see any new notification in NS8 (shouldn’t I?), but I see my VM disk image grow, show it must be doing something…
I press “finish”, seems to do one more “migration sync”…
…BOOM…
Error
Synchronization failed
The following command has failed:
nethserver-ns8-migration/migration/update
Unfortunately we couldn't catch the exact error. If you want to help, please click on the button below to copy the failed command to the clipboard, paste it into the Terminal and submit command output to the developers.
About your question in PM, I got that log from “/var/log/ns8-migration.log” (in my NS7).
I haven’t managed to somehow move my logs to somewhere I can actually pull them from NS8.
Maybe I should first enable cockpit properly in NS8 as it can help me with that most probably.