There was an issue where GoldenGate trail files consumed 95% of 197G from filesystem.
PURGEOLDEXTRACTS was changed from MINKEEPHOURS 48 to MINKEEPHOURS 20
The above configuration is relative and with increase transactions can still fill up the filesystem.
Why not size the trail file accordingly and keep n trail files?
https://docs.oracle.com/en/middleware/goldengate/core/19.1/reference/purgeoldextracts-manager.html
PURGEOLDEXTRACTS for Manager
MINKEEPFILES n
Keeps at least n trail files, including the active file. The minimum is 1 and the maximum is 100. The default is 1.
File system will never get full because manager will do the clean up.
This is how an extract trail file can be sized when adding.
ADD EXTTRAIL /u09/dirdat/aa EXTRACT e_hawk, MEGABYTES 500
https://docs.oracle.com/en/cloud/paas/goldengate-cloud/gwuad/changing-size-trail-files.html
ALTER EXTTRAIL trail, EXTRACT group, MEGABYTES n
ALTER RMTTRAIL trail, EXTRACT group, MEGABYTES n
When implementing the above, it would be best if trail files reside on a separate mount from GoldenGate Home.
Would you rather have 20 redo logs at 100MB or 10 at 200MB?
There’s no right or wrong. It’s whether the shoe fits or not.