Replicating existing DB, error on startup: Unable to install change data capture pipeline

[hmhackmaster]

I am attempting to set up a secondary copy of Jellyfin (a network media player) to offload some specific components of the app. Among other things, Jellyfin uses two sqlite DBs (jellyfin.db and library.db).

For this use case, I have a 'primary' server and (currently) one secondary. If the primary isn't online I would not expect the secondary to work, so my goal was to use marmot to replicate each db between the two.
I haven't dug into the snapshot stuff yet as I kinda expected the secondary to 'catch up' from the primary db if the secondary was offline. (Each DB is currently less than 10MB). From what I understand, the snapshot would help make that faster, so I would probably do that down the line a bit (unless I am advised otherwise!).

After giving up on running the example (due to a port binding issue with NATS?), I decided to just try giving it a go.

I have two issues so far; I am fully aware these are likely not issues with marmot itself, but I am hopeful that I might be able to gain some insight here.

1. for the jellyfin.db, when I start Marmot I get the following error:
   ERR Unable to install change data capture pipeline error="near \"Order\": syntax error" node_id=1
   To test, I completely nuked the Jellyfin install and started fresh with a brand new DB and got the same error, so I assume something wonky is happening inside their DB schema. Unfortunately, I am a sysadmin so I don't know where to look next.

2. for library.db, I was able to get marmot running on my primary and secondary server. I ended up changing the Journal mode to WAL (to try and fix this issue) and saw that change appear on the secondary, so replication seems to work.
   Unfortunately, when I then try to start Jellyfin, it complains that Busy: SQLitePCL.pretty.SQLiteException: database is locked.
   If I start marmot second, marmot complains about a lock as well:

DBG Opening database node_id=1 path=/var/lib/jellyfin/data/library.db
ERR Unable to open database error="database is locked" node_id=1

To be honest, I didn't think db locking was going to be an issue based on my understanding of what marmot does, but perhaps I am going about this incorrectly.

Any thoughts are appreciated!

[maxpert]

I will try to set it up today and see if I can get it reproduced. Seems like bug in table creation or something where table name is order and I am not getting name quoted. Would you mind dumping the DB file here?