So I just wanted to migrate my mail (with smart host) + webtop to NS8.
I connected from my NS7 VM to my NS8 clean instance.
Detected the services that I can migrate.
I chose to only migrate the above.
I clicked to start migration. OK.
I clicked to sync. OK.
I re-clicked to sync. OK but took way more time (!?)…
I clicked to finalize migration…
BOOM.
I got this that it failed and could not detect the error.
Gave me the following command to paste to console and got this:
I’ll recover my VM from backup now.
echo '{"app":"nethserver-mail","action":"finish","migrationConfig":{"roundCubeVirtualHost":""}}' | /usr/bin/setsid /usr/bin/sudo /usr/libexec/nethserver/api/nethserver-ns8-migration/migration/update | jq
{
"progress": "0.00",
"time": "0.0",
"exit": 0,
"event": "migration-sync",
"state": "running",
"step": 0,
"pid": 0,
"action": ""
}
rsync: failed to connect to 10.5.4.1 (10.5.4.1): Connection refused (111)
rsync error: error in socket IO (code 10) at clientserver.c(126) [sender=3.1.2]
rsync: failed to connect to 10.5.4.1 (10.5.4.1): Connection refused (111)
rsync error: error in socket IO (code 10) at clientserver.c(126) [sender=3.1.2]
rsync: failed to connect to 10.5.4.1 (10.5.4.1): Connection refused (111)
rsync error: error in socket IO (code 10) at clientserver.c(126) [sender=3.1.2]
rsync: failed to connect to 10.5.4.1 (10.5.4.1): Connection refused (111)
rsync error: error in socket IO (code 10) at clientserver.c(126) [sender=3.1.2]
rsync: failed to connect to 10.5.4.1 (10.5.4.1): Connection refused (111)
rsync error: error in socket IO (code 10) at clientserver.c(126) [sender=3.1.2]
rsync: failed to connect to 10.5.4.1 (10.5.4.1): Connection refused (111)
rsync error: error in socket IO (code 10) at clientserver.c(126) [sender=3.1.2]
rsync: failed to connect to 10.5.4.1 (10.5.4.1): Connection refused (111)
rsync error: error in socket IO (code 10) at clientserver.c(126) [sender=3.1.2]
rsync: failed to connect to 10.5.4.1 (10.5.4.1): Connection refused (111)
rsync error: error in socket IO (code 10) at clientserver.c(126) [sender=3.1.2]
rsync: failed to connect to 10.5.4.1 (10.5.4.1): Connection refused (111)
rsync error: error in socket IO (code 10) at clientserver.c(126) [sender=3.1.2]
rsync: failed to connect to 10.5.4.1 (10.5.4.1): Connection refused (111)
rsync error: error in socket IO (code 10) at clientserver.c(126) [sender=3.1.2]
rsync: failed to connect to 10.5.4.1 (10.5.4.1): Connection refused (111)
rsync error: error in socket IO (code 10) at clientserver.c(126) [sende
Also my NS8 is reporting this:
module/mail1/import-module
Task module/mail1/import-module failed validation
Copy to clipboard
Copy task trace
module/mail1/import-module
Task ID: Copy to clipboard
5608c325-759f-49b5-bcea-85a279ca851b
OK it failed again.
NS7 working fine (from BU), NS8 clean.
Migration module correctly detected NS8 fast and reported correctly the modules that could be migrated.
I disabled mattermost (and one other experimental) and only left mail that also includes imap connectors and webtop.
The initial “start migration” was rather fast.
The one time I pressed “sync”, took like 10 minutes. I thought was stuck as I never got a progress (stayed 0%). But eventually it returned me to the migration page.
So I pressed “finish”… Which started a new sync (!?)…
And this went BOOM after a few minutes:
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.
At this point, cockpit works fine in NS7 but mail sever is dead (or webtop …oh wait postfix is disabled, so mail is fully dead)…
In my NS8 I do not see any fails (at least after I pressed the “finish”), so I cannot do what you asked.
Last notifications (that are after I pressed finish) are “cluster/list-user-domains” twice, completed.
I will restore my NS7 from backup once more. (I will keep the “broken” VM in case we need to get something from it)
I will keep NS8 as is for now, pause the VM, in case there is a log that can help us.
I cannot give up, I need to eventually migrate.
I will get back to you with files.
I don’t remember anything about Roundcube as it was an experimental build and never used, this is why I didn’t choose to migrate.
Which is already a bug, as something that you don’t choose to migrate (or touch) shouldn’t bother the migration whatsoever.
In the end I might just remove/delete unneeded services from NS7 before I migrate - but we better make the migration process more resilient first, right?
It should not be running out of resources.
It is a KVM based VM (running on UNRAID).
The machine is a 6 core/12 thread CPU, 32GB RAM and the VM itself is on M.2.
The VM configuration is 4 threads and 2GB RAM for NS8 and 4 threads 4GB RAM for NS7.
hello mate, something new to test, we do not want to migrate nethserver-roundcubemail-next (from Nethforge, no worry, manual dump can been done) so we have changed the way how to detect nethserver-roundcubemail from nethserver-roundcubemail-next
@stephdl can you please tell me… Do I still need to manually install that update? Because there was an official update yesterday and migration is already version 1.07.
Also again: I don’t use roundcube, I use webtop… will this also help me?
Should I just retry migration from scratch? Anything else? Please let me know how to proceed.
