These two commands achieve somewhat similar goals but with some subtle differences. This article attempts to address them.
Here are some of the differences between “abort config” and “ora_cleansp”:
1. The abort config will not clean up the log directory and state directory but will clean up the contents of the rim directory to get rid of SharePlex queues and will also clean up the SharePlex internal tables in the database. This allows the next activation to succeed without any glitches. The ora_cleansp, on the other hand will not only clean up the rim directory and the SharePlex internal tables in the database but will also contents of log and state directories. This also allows subsequent replication to succeed without any glitches that can occur due to residual values in the rim directory or the SharePlex internal tables.
2. The abort config need only be run on the source and the cleanup occurs on both source and the associated target(s). The ora_cleansp, on the other hand need to be run separately on source and the target(s).
3. The abort config can be run while SharePlex is up on source and target. In fact, it is a requirement that SharePlex and its processes be running on both source/target when it is issued otherwise the command will be stuck and will only do the cleanup downstream once a stopped process or SharePlex instance is restarted. The ora_cleansp, on the other hand, should only be run when the SharePlex instance in question is shutdown.
4. For a consolidated target which receives data from multiple sources, the abort config will only get rid of a particular data stream. However, the ora_cleansp will not work with such granularity and will clean up the replication on target for all data streams for that port #.
These differences also apply to deactivate config and purge config.