We were having an issue on DB management: whenever the operators were making a checkpoint, they were calling GetAggregatedCheckpointMessages, which triggered a SELECT on the DB. Since we are continuously writing to the DB as well, if a write happened to be happening at the same time, the DB would be locked, as it's SQLite - not supposed to be concurrent.
An immediate fix to this is simply limiting the connections, as effectively we are not using more than one at once.
We were having an issue on DB management: whenever the operators were making a checkpoint, they were calling
GetAggregatedCheckpointMessages
, which triggered aSELECT
on the DB. Since we are continuously writing to the DB as well, if a write happened to be happening at the same time, the DB would be locked, as it's SQLite - not supposed to be concurrent. An immediate fix to this is simply limiting the connections, as effectively we are not using more than one at once.