Hi, @MaBeuLux88_xxx and @Jascha_Brinkmann !
I was also looking for a way to resolve this issue of Change Stream when the resume point is no longer in the oplog.
I understood your reply that increasing oplog size is necessary but I believe it is not that ultimate solution, because Change Stream must not lost resume point in production level. Even if I increase the size, there may be situations that much more data than I expect updates happen sometimes. What I mean is, the size I have already increased may not always be enough.
Is there any way to definitely prevent such a situation in advance?
+)
I have one more question.
If Mongo Source Connector is already stopped by the exception, how can I restart it to resume consuming Change Stream? It’s very hard to find a way to recover it with Mongo Source Connector in googling. T.T
Can you please give me some advice?
Thank you in advance.