You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
What especially stands out here is that the client's /closes metric is more than twice that of /connects -- I assume that /closes counts failed connections and /connects does not?
We have disabled telemetry, usage reporting, etc, with no improvement. Why can't we keep connections established?
The text was updated successfully, but these errors were encountered:
Just to add some more information about the setup - k8s (v1.4.8) cluster was started through kops, 1 master + 1 node, default kubenet networking setup and the default Debian image.
I had similar 502s on 0.8.6 the day before. Here's a 502 example:
@aalimovs reports that he's seeing 502s when trying to do a simple test against a kubernetes l2l setup on AWS t2.large instances.
Using this configuration: https://github.com/BuoyantIO/linkerd-examples/blob/master/k8s-daemonset/k8s/linkerd-zipkin.yml
metrics.json snapshot
What especially stands out here is that the client's
/closes
metric is more than twice that of/connects
-- I assume that/closes
counts failed connections and/connects
does not?We have disabled telemetry, usage reporting, etc, with no improvement. Why can't we keep connections established?
The text was updated successfully, but these errors were encountered: