I've just finished adding an option 3 to the proposal that balances off tradeoffs between the previous two options: https://docs.google.com/document/d/1DcaHoWZnA-N5UlQ7sJ0IlPKe4ul2nimWlzWaZSxPkNU/edit?ts=602cd53c#heading=h.f1ppnnbncb4l
On Tuesday, February 16, 2021 at 1:43:25 PM UTC-5 Robert Fratto wrote: > Hi! > > I've recently observed that the tolerance to a Remote Write outage is > variable. The amount of lost data depends on the volume of existing data in > WAL segments, which WAL segment is currently being tailed by Remote Write, > and how much time remains before the next WAL checkpoint is created. > > I've written a proposal to try to address this and make tolerance to an > outage more predictable. Feedback > would be appreciated: > https://docs.google.com/document/d/1DcaHoWZnA-N5UlQ7sJ0IlPKe4ul2nimWlzWaZSxPkNU/edit# > > I plan to contribute the changes for this myself. > > Best, > Robert > -- You received this message because you are subscribed to the Google Groups "Prometheus Developers" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. To view this discussion on the web visit https://groups.google.com/d/msgid/prometheus-developers/438ddc39-d24e-4cc3-a967-60a0752fe133n%40googlegroups.com.
