Björn Rabenstein
Björn Rabenstein
It is _very_ vaguely planned. Hidden in a dev summit note from ages ago… https://docs.google.com/document/d/11LC3wJcVk00l8w5P3oLQ-m3Y37iom6INAMEu2ZAGIIE/edit?pli=1#bookmark=id.m8eq3rv3yopu
Maybe this is worth a top-level flag like `--agent`? See also my ramblings in #14744 .
This should go into the v3 release branch, shouldn't it?
Clearly, the intention was to have the metric field filled. @zenador would you like to investigate?
Ideally, we would have both names, or even be more specific about the exact label match that leads to the collision. Or maybe that's way too much noise and we...
Thanks for investigating. Yeah, so at the very least, we would need a more generic wording in the current annotation (or more flexible wording if the metric name is empty)....
Yes, we should have a better fitting annotation. Not sure how detailed we want the message to be. If we enumerate all label matches that lead to a mix, we...
@NeerajGartia21 since you added annotation for other bin ops, I'll assign this to you for now.
> We can use the existing info annotation like- incompatible sample types encountered for binary operator: exponential bucket histogram $OP custom bucket histogram. It seems good but it is an...
Hello from the bug scrub! We close this based on @pracucci 's response. If anyone is missing anything, please follow up.