proxy: v2.145.0

[olix0r]

• Controller clients of components with more than one replica could fail
  to drive all connections to completion. This could result in timeouts
  showing up in logs, but would not have prevented proxies from
  communicating with controllers. Outbound proxy delays writing of TLS handshake message  #6146
• ingress: Require the l5d-dst-override header linkerd2-proxy#992 made the l5d-dst-override header required
  for ingress-mode proxies. This behavior has been reverted so that
  requests without this header are forwarded to their original
  destination.
• OpenCensus trace spans for HTTP requests no longer include query
  parameters.

---

• ci: Update/pin action dependencies (Update/pin action dependencies linkerd2-proxy#1012)
• control: Ensure endpoints are driven to readiness (control: Ensure endpoints are driven to readiness linkerd2-proxy#1014)
• Make span name without query string (Make span name without query string linkerd2-proxy#1013)
• ingress: Restore original dst address routing (ingress: Restore original dst address routing linkerd2-proxy#1016)
• ci: Restict permissions in Actions (ci: Restict permissions in Actions linkerd2-proxy#1019)
• Forbid unsafe code in most module (Forbid unsafe code in most module linkerd2-proxy#1018)

[olix0r]

The multicluster test keeps stalling in CI. It passes locally, though it's surprisingly slow:

Test script: [install_test.go] Params: [--multicluster --certs-path /tmp/l5dcerts.Gkr]
ok      command-line-arguments  151.824s
Test script: [target1] Params: [--multicluster]
ok      github.com/linkerd/linkerd2/test/integration/multicluster/target1       0.341s
Test script: [install_test.go] Params: [--multicluster --certs-path /tmp/l5dcerts.Gkr]
ok      command-line-arguments  146.376s
secret/cluster-credentials-target created
link.multicluster.linkerd.io/target created
clusterrole.rbac.authorization.k8s.io/linkerd-service-mirror-access-local-resources-target created
clusterrolebinding.rbac.authorization.k8s.io/linkerd-service-mirror-access-local-resources-target created
role.rbac.authorization.k8s.io/linkerd-service-mirror-read-remote-creds-target created
rolebinding.rbac.authorization.k8s.io/linkerd-service-mirror-read-remote-creds-target created
serviceaccount/linkerd-service-mirror-target created
deployment.apps/linkerd-service-mirror-target created
service/probe-gateway-target created
Test script: [source] Params: [--multicluster]
ok      github.com/linkerd/linkerd2/test/integration/multicluster/source        8.501s
Test script: [target2] Params: [--multicluster]
ok      github.com/linkerd/linkerd2/test/integration/multicluster/target2       15.226s
INFO[0000] Deleting cluster 'source'                    
INFO[0000] Deleted k3d-source-serverlb                  
INFO[0002] Deleted k3d-source-server-0                  
INFO[0002] Deleting image volume 'k3d-source-images'    
INFO[0002] Removing cluster details from default kubeconfig... 
INFO[0002] Removing standalone kubeconfig file (if there is one)... 
INFO[0002] Successfully deleted cluster source!         
INFO[0000] Deleting cluster 'target'                    
INFO[0000] Deleted k3d-target-serverlb                  
INFO[0002] Deleted k3d-target-server-0                  
INFO[0002] Deleting image volume 'k3d-target-images'    
INFO[0002] Removing cluster details from default kubeconfig... 
INFO[0002] Removing standalone kubeconfig file (if there is one)... 
INFO[0002] Successfully deleted cluster target!         
        Command being timed: "bin/tests /home/ver/b/l2/bin/linkerd --name multicluster"
        User time (seconds): 35.92
        System time (seconds): 7.74
        Percent of CPU this job got: 8%
        Elapsed (wall clock) time (h:mm:ss or m:ss): 8:14.67

It takes ~2.5 minutes for each install (which occur serially, so ~5 mins total) and there's about 3 minutes spent doing cluster setup, mostly loading images (most of which don't appear used by these tests -- tap, grafana, etc).

In CI a single install_test takes significantly longer:

ok  	command-line-arguments	454.913s

which makes it virtually impossible for this test to complete in under 10 minutes (which appears to be a timeout we're hitting):

Error: TestInstallOrUpgradeCli - failed to wait rollout of deploy/linkerd-proxy-injector
268
--- FAIL: TestInstallOrUpgradeCli (602.82s)
269
    kubernetes_helper.go:330: failed to wait for rollout of deploy/linkerd-proxy-injector: exit status 1: Waiting for deployment "linkerd-proxy-injector" rollout to finish: 0 of 1 updated replicas are available...
270
        error: deployment "linkerd-proxy-injector" exceeded its progress deadline
271
FAIL
272
FAIL	command-line-arguments	605.375s
273
FAIL
274
Error: Process completed with exit code 1.

While we probably need to increase these timeouts in the short-term, we probably need to invest in reducing unnecessary overhead in our tests...

[Pothulapati]

LGTM!

(looking forward for the tests to pass)

[alpeb]

LGTM 👍