I think istio really missed a chance to display observability across multicluster - the way the traffic can move from 1 cluster to another and quickly visualize in kiali. No other service mesh has this direct "viewability" - Kuma's MeshTrace needs lot of config to correctly send data to Jaeger and still cannot differentiate b/w boundaries of one cluster to another. Linkerd Viz is great but again restricted to 1 cluster and shows lot of compoennts that are not needed. Hence Istio's Kiali wins the "observability" goal IMHO
I think it would be better if the guests answer that one instead of me. They might not be watching the comments in the video so it might be better to ask it in Slack (where the vote it).
I'm asking because the Gamma implemention page only lists three: Istio, Kuma, and Linkerd; and of those only Istio and Kuma are listed as GA. (And I've commented in slack.)
I wasn't aware that Kuma is one of those. More importantly, I'm not sure what project-specific plans are for the future, hence better if those directly involved comment on it.
Let me disagree, I think he said just what he wanted, he talked about the DNA of the project with its visions and goals. I don’t use any service mesh, btw.
Just gonna be honest here, this is excruciating to watch. Can you really not try to give these people a little bit of respect for 5 minutes? Constantly downplaying, constantly rejecting (even though your own input is simply wrong), typing louder than the person that's introducing themselves can speak. Man, if you didn't want to do this, you didn't have to. It's really hard watching Whitney try to pull it together while you are doing everything you can to signal that you're completely disinterested. I hope you don't act like this towards your coworkers.
The Istio demo was great!
I think they should really start showing the multicluster view in kiali - this is a clear winner in observabilit IMHO
I think istio really missed a chance to display observability across multicluster - the way the traffic can move from 1 cluster to another and quickly visualize in kiali. No other service mesh has this direct "viewability" - Kuma's MeshTrace needs lot of config to correctly send data to Jaeger and still cannot differentiate b/w boundaries of one cluster to another. Linkerd Viz is great but again restricted to 1 cluster and shows lot of compoennts that are not needed. Hence Istio's Kiali wins the "observability" goal IMHO
Starts at 5:09.
54:16 How are these projects positioned for the k8s Gateway API Gamma project that standardizes service mesh? Is everyone on board?
I think it would be better if the guests answer that one instead of me. They might not be watching the comments in the video so it might be better to ask it in Slack (where the vote it).
In the meantime, the short answer is that some are prepared, while others are not. Still, everyone will have to be onboard or risk loosing users.
I'm asking because the Gamma implemention page only lists three: Istio, Kuma, and Linkerd; and of those only Istio and Kuma are listed as GA. (And I've commented in slack.)
I wasn't aware that Kuma is one of those. More importantly, I'm not sure what project-specific plans are for the future, hence better if those directly involved comment on it.
It’s impressive how the Linkerd guy talked for five minutes without actually saying anything of substance.
Let me disagree, I think he said just what he wanted, he talked about the DNA of the project with its visions and goals.
I don’t use any service mesh, btw.
Just gonna be honest here, this is excruciating to watch. Can you really not try to give these people a little bit of respect for 5 minutes? Constantly downplaying, constantly rejecting (even though your own input is simply wrong), typing louder than the person that's introducing themselves can speak. Man, if you didn't want to do this, you didn't have to. It's really hard watching Whitney try to pull it together while you are doing everything you can to signal that you're completely disinterested. I hope you don't act like this towards your coworkers.