Anton Gilgur

We may want to reconsider this later within the context of #10341, i.e. if pending and active time are split in two. The backoff `maxDuration` may make more sense for...

> We run ~500 workflows and ~500 pods concurrently So ~2500 concurrent Pods total? > It is expected that argo workflows do not flood into etcd or impact the stability...

> * Workflows behave different from business. Some of them finished in several seconds, while some run for a few hours. Yea Workflows in general have a lot of diverse...

Oh I forgot to mention earlier, there is also the [environment variable](https://argo-workflows.readthedocs.io/en/latest/environment-variables/) `ALWAYS_OFFLOAD_NODE_STATUS` that could help in this scenario as well

Merged `main` for new PR checks per #13027

Hmm this seems like a duplicate of #10157; @isubasinghe I thought you tested that this was working?

> I do not intend to implement tracing as part of this proposal. For reference, we do have a separate feature request for that: #12077 > The metric which are...

Potentially related (I haven't root caused it yet): #12206 had remaining issues with metrics as well. We've fixed the root cause of the issue, but the metrics are still confusing....

> I was wondering why there were two different endpoints; were these just the same data but differently formatted? > This would simplify a number of things for operators and...

@erikschul as mentioned [above](https://github.com/argoproj/argo-workflows/issues/11858#issuecomment-1730159573), we'll happily take a PR to update the examples to a similar image that is cross-architecture compatible