Jaeger Collector Istio Sidecar Exclusion

[timcosta]

Hey all! I'm curious about the reasoning for jaeger-collector explicitly disabling the istio sidecar injection with no option to re-enable it. I check the code for comments, looked at the original PR, and search for issues but I don't see any discussion for why it was done. I'm trying to understand because it results in a lot of unknown destinations for workloads in our cluster that talk to it and have the sidecar enabled, and I can't come up with a reason why the injection would be hard disabled like it is. Anyone have any thoughts or know the reason?

[sereneshikari]

Ditto. I went through the CHANGELOG.md and could only find two PRs (#67 and #1860) related to the Istio sidecar annotations and neither of them provide any insight into why it's being done. No dice with the operator documentation either. Would really appreciate it if someone could explain this seemingly arbitrary decision! 😅

[sereneshikari]

Almost been three weeks since I asked this 😅 @jpkrohling, I see that you approved #67. Could you please provide some insight into this decision? I apologise if there's a simple reason for this but I'm not able to understand it. One thing that I did read in the documentation is

[Defining sampling strategies] is not relevant if a trace was started by the Istio proxy as the sampling decision is made there. And the Jaeger sampling decisions are only relevant when you are using the Jaeger tracer (client). [reference]

But it's still not clear to me if this is the reason why it's done