-
Notifications
You must be signed in to change notification settings - Fork 150
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
sync failure from 3.2.6-2+deb11u2 to 3.6 and 3.8.3 (last one + last patches) #4968
Comments
Other server, other mailbox :
We still have this "MBOXTYPE ei", MBTYPE_EMAIL and MBTYPE_INTERMEDIATE. Similarly, in the previous report, "FORMULAIRE FR .MOIS .JANV 24.22-01-24" doesn't exist yet but "FORMULAIRE FR .MOIS .JANV 24.22-01-24.FAIT " already exists. Sincerly, |
This isn't really surprising. The first replication works, because that's important when using replication as part of an upgrade process. But the newer replica will be storing different data than the older primary, and is likely to be confused when it receives incremental updates that it disagrees with. When this happens, the right thing to do is remove the data from the replica and do a fresh full replication. You generally shouldn't run mismatched primary/replica versions on an ongoing basis like this, just temporarily during an upgrade.
In this particular case, I think the problem will go away if you just create real mailboxes for the MBTYPE_INTERMEDIATE mailboxes on the primary. You could do this using IMAP CREATE or "createmailbox" in cyradm. In 3.6 and later, all mailboxes path components are real mailboxes (no more MBTYPE_INTERMEDIATE), so the replication is probably disagreeing about the MBTYPE_INTERMEDIATE on the primary vs the real mailbox on the replica. In a one-shot upgrade replication this difference wouldn't be a problem, but it is for ongoing incremental updates since the updates don't match. |
From time to time my 3.2.6-2+deb11u2 production servers can't synchronise to their 3.8.3 replicant, with segfault.
Getting a fresh new empty replicant make the synchro working again.
Moving the "bad" user on an other production server make the mailbox synchronise on the new replicant, and the initial server can finish its synchronisation list.
Here's an actual folder with the problem :
On the replicant :
How can I help you find the problem ?
The text was updated successfully, but these errors were encountered: