feat: Handle heartbeat events #1200
Conversation
fixed: 908cfa1 Later we probably want to store offset progress for each slice from the heartbeats and that will also require special case in the offset store. |
| TimestampOffset(startTime.plus(evictInterval).plus(JDuration.ofSeconds(1)), Map(p4 -> 1L)), | ||
| p4, | ||
| 1L)) | ||
| .futureValue |
There was a problem hiding this comment.
These are now no longer evicted because of the keep latest change?
There was a problem hiding this comment.
I'm not sure what the original intention with this one was, but it was never saved because it was a duplicate. Same p4 and seqNr 1 as above. I changed isDuplicate to also look at timestamp when same seqNr and then this test failed. By removing this the test is kind of as it was before.
| TimestampOffset(startTime.plus(evictInterval).plus(JDuration.ofSeconds(1)), Map(p4 -> 1L)), | ||
| p4, | ||
| 1L)) | ||
| .futureValue |
There was a problem hiding this comment.
I'm not sure what the original intention with this one was, but it was never saved because it was a duplicate. Same p4 and seqNr 1 as above. I changed isDuplicate to also look at timestamp when same seqNr and then this test failed. By removing this the test is kind of as it was before.
| TimestampOffset(startTime.plus(evictInterval).plus(JDuration.ofSeconds(1)), Map(p4 -> 1L)), | ||
| p4, | ||
| 1L)) | ||
| .futureValue |
patriknw
force-pushed
the
wip-heartbeat-patriknw
branch
from
908cfa1 to
32db079
Compare
|
We decided to not emit the heartbeats after the query subscriber, so they will not reach the offset store. |
See akka/akka-persistence-r2dbc#599