We are receiving ORA-04031 errors. We have done some testing and it seems that the apply process is dying -- this is causing the Oracle instance on the target to also die. Architecture - SUNOS 5.9 sun4u-sparc SUNW, Ultra-Enterprise, Oracle 188.8.131.52 + patch 3573467. All of the above is on source and target shared pool 110 MB. We use default size for first_spare_paramater (I believe this is 10%).
The real question is how do you easily recover from these problems? When Streams has a problem it seems to bring the whole instance down with it. Is this normal?
You might be getting this error because your shared pool is too small. Query V$SGASTAT regularly and look for how much 'free memory' your shared pool has. If this number is very small (less than 1 MB), then consider increasing your shared pool.