luistilingue
luistilingue
I'll check it if PVCs won't be full filled over time.
@joe-elliott I guess it was fixed by upgrading Prometheus depedency. It is a very odd issue. Thank you very much.
@mapno I'm getting an issue related to that, but in Tempo 2.4.0: after enable the FeatureToggle in Grafana 10.3.3 `traceQLStreaming` and `stream_over_http_enabled` in Tempo: ``` Tempo Query-Frontend: level=warn ts=2024-03-04T19:11:50.349751175Z caller=server.go:1480...
My problem is on Grafana https://github.com/grafana/grafana/issues/81998
I'm with the same issue related to the serviceGraph connector :(
``` apiVersion: apisix.apache.org/v2 kind: ApisixUpstream metadata: name: httpbin-upstream namespace: default spec: loadbalancer: type: ewma discovery: type: kubernetes serviceName: default/httpbin:http ``` that way works for me.
It is still very useful and relevant be able to limit ingress access per-namespace or per-ingress, instead in per-cluster only. @networkop is there anything related to that in Cilium roadmap...
It'll be great!
> This bug is introduced with latest release of Loki which is version 3.5.0. Yes, I'm facing it. It'll be great if 3.5.1 be released soon with that fix!
I'm working for conversational platform company and it uses (like many others) WebSockets and raw-TCP with a XMPP-based protocol for their apps due to mainly full-duplex communication, called [LIME](https://limeprotocol.org/) ....