We Lost Unicode Characters after Confluence Upgrade

[Xin Li]

We just upgraded our confluence from 5.5.1 to 5.10.9.

after the upgrade, we found out our Unicode strings are not visible, they became question marks instead. I checked the database (SQL Server), in the table BODYCONTENT, the Unicode characters became question masks!

as there is no changes for the table BODYCONTENT, I don't know why the data in the table are changed.

during the upgrade, I am aware that it takes almost one hour working the table BODYCONTENT, but don't know what they are doing:

2020-06-16 16:33:15,995 INFO [localhost-startStop-1] [confluence.upgrade.impl.DefaultUpgradeManager] beforeUpgrade Generating pre-upgrade recovery file...
2020-06-16 16:33:16,054 INFO [localhost-startStop-1] [confluence.upgrade.recovery.ConfluenceProgressMonitor] begin []
......
2020-06-16 16:33:29,110 INFO [localhost-startStop-1] [confluence.upgrade.recovery.ConfluenceProgressMonitor] begin TABLE_DATA: [BODYCONTENT]
2020-06-16 17:24:34,055 INFO [localhost-startStop-1] [confluence.upgrade.recovery.ConfluenceProgressMonitor] end TABLE_DATA: [BODYCONTENT]
......
2020-06-16 17:36:02,716 INFO [localhost-startStop-1] [confluence.upgrade.impl.DefaultUpgradeManager] entireUpgradeFinished Generating post-upgrade recovery file...
2020-06-16 17:36:02,831 INFO [localhost-startStop-1] [confluence.upgrade.recovery.ConfluenceProgressMonitor] begin []
......
2020-06-16 17:36:12,887 INFO [localhost-startStop-1] [confluence.upgrade.recovery.ConfluenceProgressMonitor] begin TABLE_DATA: [BODYCONTENT]
2020-06-16 18:14:56,127 INFO [localhost-startStop-1] [confluence.upgrade.recovery.ConfluenceProgressMonitor] end TABLE_DATA: [BODYCONTENT]
......

I tried to find out the schema upgrade SQL scripts, but no luck.

anyone see the same or similar issue?

thanks in advance.