If we have the node-name then we can save work requesting the list of
local pods from kubelet. And the best way to get the node name is via
the "downwards api", i.e. put this in the Daemonset Yaml:
- name: KUBERNETES_NODENAME
valueFrom:
fieldRef:
apiVersion: v1
fieldPath: spec.nodeName
* Add topology.ReplaceNode() for efficiency
In some places AddNode() was called after adding to an existing node,
in which case the Merge() is just a waste of time.
The link is disabled if the `monitor` flag is not true.
If Weave Cloud runs without Monitor, the link is dead. The flag allows to pass Weave Cloud's Monitor state down to Scope.
The link returned by service metrics is now built by looking up the deployment name in `report.Report` instead of assuming the service and the deployment have the same name.
The previous link to 'Monitor' contained a recording rule metric, and since not all Weave Clound instances have recording rules, the link would redirect to a 'Monitor' page without data points.
The recording rule metric has been substituted with the underlying query.
The instructions for installing via Weave Cloud were rather out of
date, and in any case are embedded in Weave Cloud itself, so there is
no point repeating them here.
Reflectors are created and run within the same function, asynchronously from the main thread.
Creating reflectors may require calls to the kubernetes api, which can return errors.
API errors are not handled in the main thread, but are handled asynchronously by retries.
- It should only be visible on the kubernetes contoller node-details
panels.
- This fixes a bug where we were showing it on all node-details if you
were on the controller topology.
Moving `deletePipe()` in `componentWillUnmount()` ensures the pipe is deleted when:
1. the terminal is closed
2. the terminal is closed when closing the pod panel
The pipe still does not get deleted when popping out the terminal.
Upgraded from 99c19923, branch release-3.0.
This required fetching or upgrading the following:
* k8s.io/api to kubernetes-1.9.1
* k8s.io/apimachinery to kubernetes-1.9.1
* github.com/juju/ratelimit to 1.0.1
* github.com/spf13/pflag to 4c012f6d
Also, update Scope's imports/function calls to be compatible with the new client.
This includes:
* an eBPF object built with a newer kernel (4.14)
* an increased threshold for guessing offsets, which is necessary for
the eBPF tracer to work on Google's Container-Optimized OS (used by
GKE)
