Mining GoldenGate Error Logs for Consumption Rates of Trail Files

After resolving Oracle GoldenGate issues, a question typically asked is, “When will the process catch up?” The usual method to find out is to use the lag command; however, I was more interested to see the rate of the trail file being consumed.

One method to do this is to mine ggserr.log for the GoldenGate process, e.g. replicat.prm.

From the data below, replicat is consuming trail at rate of ~7m per trail.

Hence, if there are 10 more trails to process then it will take ~70m to complete.

grep "replicat.prm:  Switching to next trail file" ggserr.log| tail
2019-11-14 14:50:20  replicat.prm:  Switching to next trail file dirdat/aa000026334
2019-11-14 14:56:42  replicat.prm:  Switching to next trail file dirdat/aa000026335
2019-11-14 15:02:43  replicat.prm:  Switching to next trail file dirdat/aa000026336
2019-11-14 15:08:19  replicat.prm:  Switching to next trail file dirdat/aa000026337
2019-11-14 15:17:42  replicat.prm:  Switching to next trail file dirdat/aa000026338
2019-11-14 15:23:46  replicat.prm:  Switching to next trail file dirdat/aa000026339

FYI: I have removed some non-relavant columns to make the format cleaner.

I didn’t capture lag information for comparison. You can do this as a comparison for future issues. Hopefully, the information is useful for your next troubleshooting efforts.