istio: Increase in source_app being reported as "unknown" for HTTP 200 requests in 1.8.2

Bug description

After deploying Istio 1.8.2 on our cluster, there has been a noticeable jump in the source app being “unknown” on the istio_requests_total metric. This is in stark contrast to before, where no requests were being labeled as such.

The graph below shows the effect - before the upgrade, no HTTP 200 codes. After, they occur consistently.

image

Note that it doesn’t seem to be re-classifying requests from the ingress gateway as “unknown”, either - HTTP 200 from the source_app of istio-ingressgateway is pretty much the same before and after the deployment:

image

Is there any explanation for this behavior?

[ ] Docs [ ] Installation [ ] Networking [ ] Performance and Scalability [ x ] Extensions and Telemetry [ ] Security [ ] Test and Release [ ] User Experience [ ] Developer Infrastructure [ ] Upgrade

Expected behavior

No change in source_app reporting in telemetry between versions

Steps to reproduce the bug

Version (include the output of istioctl version --remote and kubectl version --short and helm version --short if you used Helm)

Istio 1.8.2, Helm v3.5.0

How was Istio installed?

Using the Istio Operator and Istio Profile, standalone via Helm with our own supplied values.

Environment where the bug was observed (cloud vendor, OS, etc)

Azure Kubernetes Service

About this issue

  • Original URL
  • State: closed
  • Created 3 years ago
  • Reactions: 1
  • Comments: 39 (20 by maintainers)

Most upvoted comments

@AlexDHoffer You’d need to not only remove the ServiceMonitor, but also update prometheus to scrape based on standard annotations (i.e. scraping 15020 as prometheus.io annotation indicates, instead of http-envoy-prom port).

@Starefossen Could you please create a separated issue? Based on your metrics, it looks like tcp exchange is not happening, which we need to troubleshoot separately.