I am trying to configure SymmetricDS to run a scheduled replication from one postgres database to another.
I've seen people suggesting doing that by setting these fields in node config:
job.routing.period.time.ms
job.push.period.time.ms
job.pull.period.time.ms
Or by creating time windows for replication in this table:
NODE_GROUP_CHANNEL_WINDOW
The first options seems to be better as I don't want to specify end time of replication. The issue is that I want to only replicate the database from one point in time, when the replication starts, so the data integrity is preserved (eg. if the replication takes 2h and there are changes made to the source DB after the replication had started I want them to be replicated in the next cycle) Should I set all of these values to 24h, or does the routing job need to run more frequently?
While trying to test this I also ran into the following issue - not all the data has been replicated in single scheduled sync (the database is quite large):
Reached the total byte threshold after 3 of 5 batches were extracted for node '001'. The remaining batches will be extracted on a subsequent sync
How to disable the limitation? I know of 3 properties of each channel but I'm not sure how to configure them.
MAX_BATCH_SIZE
MAX_BATCH_TO_SEND
MAX_DATA_TO_ROUTE
The other thing I am not certain about is how to assign tables to routers and channels in order to achieve what I specified above.