Björn Rabenstein
Björn Rabenstein
The problem with the current metadata handling is that it is (a) ephemeral (just buffering the last known state) and (b) per metric family. It's perfectly possible that a counter...
I'm still not convinced that storing the created timestamp in the current metadata store makes any sense. What is the use case, indeed? Maybe propagating the created timestamp via remote...
To update the summary here: The remaining features are blocked on having a text format representation of native histograms. On the other hand, we probably don't need every single aspect...
s/Nodes/Node/ is a clear win, I'd say. _USE_ could be argued to be well known in the monitoring community. If we want to spell it out, we run into the...
I think it's the latter.
Three tests are left to convert, see TODO list at the top. @NeerajGartia21 are you planning to convert them, too?
@NeerajGartia21 are you still planning to work on this?
With no message from @NeerajGartia21 , assigning this to @fatsheep9146 .
> So I think the actual results should also be compacted with Compact(0) before comparing. Sounds like it. Feel free to change the code of the testing framework accordingly. About...
In different news: @NeerajGartia21 is also back in action, so maybe you could divide the work between the two of you or help each other with code reviews and such.