This job view page is being replaced by Spyglass soon. Check out the new job view.
PRliggitt: WIP - [v1.16] Update to go1.13.4
ResultFAILURE
Tests 1 failed / 14 succeeded
Started2019-11-08 22:35
Elapsed23m44s
Revision
Buildergke-prow-ssd-pool-1a225945-ztzz
Refs release-1.16:c7dab3b9
85019:5aae53ae
links{u'resultstore': {u'url': u'https://source.cloud.google.com/results/invocations/38ce4457-1b56-4a81-8256-48242dabae0c/targets/test'}}
pod0c0f35bd-0278-11ea-af0d-f65e2caa0e15
resultstorehttps://source.cloud.google.com/results/invocations/38ce4457-1b56-4a81-8256-48242dabae0c/targets/test
infra-commitf2291aeef
job-versionv1.16.3-beta.0.60+b6f0e56473ef1b
pod0c0f35bd-0278-11ea-af0d-f65e2caa0e15
repok8s.io/kubernetes
repo-commitb6f0e56473ef1b2bf904f4ebe53181a7c9c2a3c5
repos{u'k8s.io/kubernetes': u'release-1.16:c7dab3b96220499c9cba75285b3db61a30f79dd9,85019:5aae53aed58fa325f14736575f1a4330c3b5c20f', u'k8s.io/release': u'master'}
revisionv1.16.3-beta.0.60+b6f0e56473ef1b

Test Failures


Test 13s

error during ./hack/ginkgo-e2e.sh --ginkgo.focus=\[Feature:GPUDevicePlugin\] --minStartupPods=8 --report-dir=/workspace/_artifacts --disable-log-dump=true --cluster-ip-range=10.64.0.0/14: exit status 1
				from junit_runner.xml

Filter through log files | View test history on testgrid


Show 14 Passed Tests

Error lines from build-log.txt

... skipping 1635 lines ...
W1108 22:43:04.463] INFO: Elapsed time: 356.217s, Critical Path: 168.44s
W1108 22:43:04.464] INFO: 4356 processes: 2346 remote cache hit, 2010 processwrapper-sandbox.
W1108 22:43:04.470] INFO: Build completed successfully, 4423 total actions
W1108 22:43:04.475] INFO: Build completed successfully, 4423 total actions
W1108 22:43:04.484] 2019/11/08 22:43:04 process.go:155: Step 'make -C /go/src/k8s.io/kubernetes bazel-release' finished in 5m56.326644661s
W1108 22:43:04.485] 2019/11/08 22:43:04 util.go:265: Flushing memory.
W1108 22:43:04.524] 2019/11/08 22:43:04 util.go:275: flushMem error (page cache): exit status 1
W1108 22:43:04.525] 2019/11/08 22:43:04 process.go:153: Running: /go/src/k8s.io/release/push-build.sh --nomock --verbose --noupdatelatest --bucket=kubernetes-release-pull --ci --gcs-suffix=/pull-kubernetes-e2e-gce-device-plugin-gpu --allow-dup
W1108 22:43:04.602] $TEST_TMPDIR defined: output root default is '/bazel-scratch/.cache/bazel' and max_idle_secs default is '15'.
W1108 22:43:04.648] WARNING: Option 'experimental_remote_spawn_cache' is deprecated
W1108 22:43:04.649] INFO: Invocation ID: 7a6e52cb-e2fa-40e5-8775-6142db1a6d47
I1108 22:43:04.749] make: Leaving directory '/go/src/k8s.io/kubernetes'
I1108 22:43:04.750] push-build.sh: BEGIN main on 0c0f35bd-0278-11ea-af0d-f65e2caa0e15 Fri Nov  8 22:43:04 UTC 2019
... skipping 850 lines ...
W1108 22:49:15.446] NODE_NAMES=e2e-e04178beac-8baae-minion-group-62l5 e2e-e04178beac-8baae-minion-group-df8l e2e-e04178beac-8baae-minion-group-jbvx e2e-e04178beac-8baae-minion-group-tz9p
W1108 22:49:15.446] Trying to find master named 'e2e-e04178beac-8baae-master'
W1108 22:49:15.446] Looking for address 'e2e-e04178beac-8baae-master-ip'
I1108 22:49:16.243] Waiting up to 300 seconds for cluster initialization.
I1108 22:49:16.243] 
I1108 22:49:16.243]   This will continually check to see if the API for kubernetes is reachable.
I1108 22:49:16.244]   This may time out if there was some uncaught error during start up.
I1108 22:49:16.244] 
W1108 22:49:16.345] Using master: e2e-e04178beac-8baae-master (external IP: 35.197.108.151; internal IP: (not set))
I1108 22:49:37.484] ...Kubernetes cluster created.
I1108 22:49:37.626] Cluster "k8s-jkns-pr-gce-gpus_e2e-e04178beac-8baae" set.
I1108 22:49:37.765] User "k8s-jkns-pr-gce-gpus_e2e-e04178beac-8baae" set.
I1108 22:49:37.908] Context "k8s-jkns-pr-gce-gpus_e2e-e04178beac-8baae" created.
... skipping 127 lines ...
I1108 22:50:50.974] [1]     	If true, adds the file directory to the header
I1108 22:50:50.974] [1]   -allow-gathering-profiles
I1108 22:50:50.975] [1]     	If set to true framework will allow to gather CPU/memory allocation pprof profiles from the master. (default true)
I1108 22:50:50.975] [1]   -allowed-not-ready-nodes int
I1108 22:50:50.975] [1]     	If non-zero, framework will allow for that many non-ready nodes when checking for all ready nodes.
I1108 22:50:50.976] [1]   -alsologtostderr
I1108 22:50:50.976] [1]     	log to standard error as well as files
I1108 22:50:50.976] [1]   -application_metrics_count_limit int
I1108 22:50:50.976] [1]     	Max number of application metrics to store (per container) (default 100)
I1108 22:50:50.976] [1]   -boot_id_file string
I1108 22:50:50.977] [1]     	Comma-separated list of files to check for boot-id. Use the first one that exists. (default "/proc/sys/kernel/random/boot_id")
I1108 22:50:50.977] [1]   -cert-dir string
I1108 22:50:50.977] [1]     	Path to the directory containing the certs. Default is empty, which doesn't use certs.
I1108 22:50:50.977] [1]   -clean-start
I1108 22:50:50.978] [1]     	If true, purge all namespaces except default and system before running tests. This serves to Cleanup test namespaces from failed/interrupted e2e runs in a long-lived cluster.
I1108 22:50:50.978] [1]   -cloud-config-file string
I1108 22:50:50.978] [1]     	Cloud config file.  Only required if provider is azure.
I1108 22:50:50.979] [1]   -cloud-provider-gce-lb-src-cidrs value
I1108 22:50:50.979] [1]     	CIDRs opened in GCE firewall for LB traffic proxy & health checks (default 130.211.0.0/22,209.85.152.0/22,209.85.204.0/22,35.191.0.0/16)
I1108 22:50:50.979] [1]   -cluster-ip-range string
I1108 22:50:50.979] [1]     	A CIDR notation IP range from which to assign IPs in the cluster. (default "10.64.0.0/14")
... skipping 91 lines ...
I1108 22:50:50.996] [1]     	If set, ginkgo will emit node output to files when running in parallel.
I1108 22:50:50.996] [1]   -ginkgo.dryRun
I1108 22:50:50.996] [1]     	If set, ginkgo will walk the test hierarchy without actually running anything.  Best paired with -v.
I1108 22:50:50.996] [1]   -ginkgo.failFast
I1108 22:50:50.996] [1]     	If set, ginkgo will stop running a test suite after a failure occurs.
I1108 22:50:50.996] [1]   -ginkgo.failOnPending
I1108 22:50:50.997] [1]     	If set, ginkgo will mark the test suite as failed if any specs are pending.
I1108 22:50:50.997] [1]   -ginkgo.flakeAttempts int
I1108 22:50:50.997] [1]     	Make up to this many attempts to run each spec. Please note that if any of the attempts succeed, the suite will not be failed. But any failures will still be recorded. (default 1)
I1108 22:50:50.997] [1]   -ginkgo.focus string
I1108 22:50:50.997] [1]     	If set, ginkgo will only run specs that match this regular expression.
I1108 22:50:50.997] [1]   -ginkgo.noColor
I1108 22:50:50.998] [1]     	If set, suppress color output in default reporter.
I1108 22:50:50.998] [1]   -ginkgo.noisyPendings
I1108 22:50:50.998] [1]     	If set, default reporter will shout about pending tests. (default true)
... skipping 71 lines ...
I1108 22:50:51.009] [1]     	If non-empty, use this log file
I1108 22:50:51.009] [1]   -log_file_max_size uint
I1108 22:50:51.010] [1]     	Defines the maximum size a log file can grow to. Unit is megabytes. If the value is 0, the maximum file size is unlimited. (default 1800)
I1108 22:50:51.010] [1]   -logexporter-gcs-path string
I1108 22:50:51.010] [1]     	Path to the GCS artifacts directory to dump logs from nodes. Logexporter gets enabled if this is non-empty.
I1108 22:50:51.010] [1]   -logtostderr
I1108 22:50:51.010] [1]     	log to standard error instead of files (default true)
I1108 22:50:51.010] [1]   -machine_id_file string
I1108 22:50:51.011] [1]     	Comma-separated list of files to check for machine-id. Use the first one that exists. (default "/etc/machine-id,/var/lib/dbus/machine-id")
I1108 22:50:51.011] [1]   -master-os-distro string
I1108 22:50:51.011] [1]     	The OS distribution of cluster master (debian, ubuntu, gci, coreos, or custom). (default "debian")
I1108 22:50:51.011] [1]   -master-tag string
I1108 22:50:51.011] [1]     	Network tags used on master instances. Valid only for gce, gke
... skipping 93 lines ...
I1108 22:50:51.027] [2]     	If true, adds the file directory to the header
I1108 22:50:51.027] [2]   -allow-gathering-profiles
I1108 22:50:51.027] [2]     	If set to true framework will allow to gather CPU/memory allocation pprof profiles from the master. (default true)
I1108 22:50:51.027] [2]   -allowed-not-ready-nodes int
I1108 22:50:51.027] [2]     	If non-zero, framework will allow for that many non-ready nodes when checking for all ready nodes.
I1108 22:50:51.028] [2]   -alsologtostderr
I1108 22:50:51.028] [2]     	log to standard error as well as files
I1108 22:50:51.028] [2]   -application_metrics_count_limit int
I1108 22:50:51.028] [2]     	Max number of application metrics to store (per container) (default 100)
I1108 22:50:51.028] [2]   -boot_id_file string
I1108 22:50:51.028] [2]     	Comma-separated list of files to check for boot-id. Use the first one that exists. (default "/proc/sys/kernel/random/boot_id")
I1108 22:50:51.029] [2]   -cert-dir string
I1108 22:50:51.029] [2]     	Path to the directory containing the certs. Default is empty, which doesn't use certs.
I1108 22:50:51.029] [2]   -clean-start
I1108 22:50:51.029] [2]     	If true, purge all namespaces except default and system before running tests. This serves to Cleanup test namespaces from failed/interrupted e2e runs in a long-lived cluster.
I1108 22:50:51.029] [2]   -cloud-config-file string
I1108 22:50:51.029] [2]     	Cloud config file.  Only required if provider is azure.
I1108 22:50:51.030] [2]   -cloud-provider-gce-lb-src-cidrs value
I1108 22:50:51.030] [2]     	CIDRs opened in GCE firewall for LB traffic proxy & health checks (default 130.211.0.0/22,209.85.152.0/22,209.85.204.0/22,35.191.0.0/16)
I1108 22:50:51.030] [2]   -cluster-ip-range string
I1108 22:50:51.030] [2]     	A CIDR notation IP range from which to assign IPs in the cluster. (default "10.64.0.0/14")
... skipping 91 lines ...
I1108 22:50:51.046] [2]     	If set, ginkgo will emit node output to files when running in parallel.
I1108 22:50:51.046] [2]   -ginkgo.dryRun
I1108 22:50:51.046] [2]     	If set, ginkgo will walk the test hierarchy without actually running anything.  Best paired with -v.
I1108 22:50:51.047] [2]   -ginkgo.failFast
I1108 22:50:51.047] [2]     	If set, ginkgo will stop running a test suite after a failure occurs.
I1108 22:50:51.047] [2]   -ginkgo.failOnPending
I1108 22:50:51.047] [2]     	If set, ginkgo will mark the test suite as failed if any specs are pending.
I1108 22:50:51.047] [2]   -ginkgo.flakeAttempts int
I1108 22:50:51.047] [2]     	Make up to this many attempts to run each spec. Please note that if any of the attempts succeed, the suite will not be failed. But any failures will still be recorded. (default 1)
I1108 22:50:51.048] [2]   -ginkgo.focus string
I1108 22:50:51.048] [2]     	If set, ginkgo will only run specs that match this regular expression.
I1108 22:50:51.048] [2]   -ginkgo.noColor
I1108 22:50:51.048] [2]     	If set, suppress color output in default reporter.
I1108 22:50:51.048] [2]   -ginkgo.noisyPendings
I1108 22:50:51.048] [2]     	If set, default reporter will shout about pending tests. (default true)
... skipping 71 lines ...
I1108 22:50:51.061] [2]     	If non-empty, use this log file
I1108 22:50:51.061] [2]   -log_file_max_size uint
I1108 22:50:51.061] [2]     	Defines the maximum size a log file can grow to. Unit is megabytes. If the value is 0, the maximum file size is unlimited. (default 1800)
I1108 22:50:51.061] [2]   -logexporter-gcs-path string
I1108 22:50:51.061] [2]     	Path to the GCS artifacts directory to dump logs from nodes. Logexporter gets enabled if this is non-empty.
I1108 22:50:51.062] [2]   -logtostderr
I1108 22:50:51.062] [2]     	log to standard error instead of files (default true)
I1108 22:50:51.062] [2]   -machine_id_file string
I1108 22:50:51.062] [2]     	Comma-separated list of files to check for machine-id. Use the first one that exists. (default "/etc/machine-id,/var/lib/dbus/machine-id")
I1108 22:50:51.062] [2]   -master-os-distro string
I1108 22:50:51.063] [2]     	The OS distribution of cluster master (debian, ubuntu, gci, coreos, or custom). (default "debian")
I1108 22:50:51.063] [2]   -master-tag string
I1108 22:50:51.063] [2]     	Network tags used on master instances. Valid only for gce, gke
... skipping 93 lines ...
I1108 22:50:51.079] [3]     	If true, adds the file directory to the header
I1108 22:50:51.079] [3]   -allow-gathering-profiles
I1108 22:50:51.079] [3]     	If set to true framework will allow to gather CPU/memory allocation pprof profiles from the master. (default true)
I1108 22:50:51.079] [3]   -allowed-not-ready-nodes int
I1108 22:50:51.079] [3]     	If non-zero, framework will allow for that many non-ready nodes when checking for all ready nodes.
I1108 22:50:51.079] [3]   -alsologtostderr
I1108 22:50:51.080] [3]     	log to standard error as well as files
I1108 22:50:51.080] [3]   -application_metrics_count_limit int
I1108 22:50:51.080] [3]     	Max number of application metrics to store (per container) (default 100)
I1108 22:50:51.080] [3]   -boot_id_file string
I1108 22:50:51.080] [3]     	Comma-separated list of files to check for boot-id. Use the first one that exists. (default "/proc/sys/kernel/random/boot_id")
I1108 22:50:51.080] [3]   -cert-dir string
I1108 22:50:51.081] [3]     	Path to the directory containing the certs. Default is empty, which doesn't use certs.
I1108 22:50:51.081] [3]   -clean-start
I1108 22:50:51.081] [3]     	If true, purge all namespaces except default and system before running tests. This serves to Cleanup test namespaces from failed/interrupted e2e runs in a long-lived cluster.
I1108 22:50:51.081] [3]   -cloud-config-file string
I1108 22:50:51.081] [3]     	Cloud config file.  Only required if provider is azure.
I1108 22:50:51.081] [3]   -cloud-provider-gce-lb-src-cidrs value
I1108 22:50:51.082] [3]     	CIDRs opened in GCE firewall for LB traffic proxy & health checks (default 130.211.0.0/22,209.85.152.0/22,209.85.204.0/22,35.191.0.0/16)
I1108 22:50:51.082] [3]   -cluster-ip-range string
I1108 22:50:51.082] [3]     	A CIDR notation IP range from which to assign IPs in the cluster. (default "10.64.0.0/14")
... skipping 91 lines ...
I1108 22:50:51.096] [3]     	If set, ginkgo will emit node output to files when running in parallel.
I1108 22:50:51.096] [3]   -ginkgo.dryRun
I1108 22:50:51.096] [3]     	If set, ginkgo will walk the test hierarchy without actually running anything.  Best paired with -v.
I1108 22:50:51.096] [3]   -ginkgo.failFast
I1108 22:50:51.096] [3]     	If set, ginkgo will stop running a test suite after a failure occurs.
I1108 22:50:51.096] [3]   -ginkgo.failOnPending
I1108 22:50:51.096] [3]     	If set, ginkgo will mark the test suite as failed if any specs are pending.
I1108 22:50:51.097] [3]   -ginkgo.flakeAttempts int
I1108 22:50:51.097] [3]     	Make up to this many attempts to run each spec. Please note that if any of the attempts succeed, the suite will not be failed. But any failures will still be recorded. (default 1)
I1108 22:50:51.097] [3]   -ginkgo.focus string
I1108 22:50:51.097] [3]     	If set, ginkgo will only run specs that match this regular expression.
I1108 22:50:51.097] [3]   -ginkgo.noColor
I1108 22:50:51.097] [3]     	If set, suppress color output in default reporter.
I1108 22:50:51.097] [3]   -ginkgo.noisyPendings
I1108 22:50:51.098] [3]     	If set, default reporter will shout about pending tests. (default true)
... skipping 71 lines ...
I1108 22:50:51.108] [3]     	If non-empty, use this log file
I1108 22:50:51.108] [3]   -log_file_max_size uint
I1108 22:50:51.108] [3]     	Defines the maximum size a log file can grow to. Unit is megabytes. If the value is 0, the maximum file size is unlimited. (default 1800)
I1108 22:50:51.108] [3]   -logexporter-gcs-path string
I1108 22:50:51.109] [3]     	Path to the GCS artifacts directory to dump logs from nodes. Logexporter gets enabled if this is non-empty.
I1108 22:50:51.109] [3]   -logtostderr
I1108 22:50:51.109] [3]     	log to standard error instead of files (default true)
I1108 22:50:51.109] [3]   -machine_id_file string
I1108 22:50:51.109] [3]     	Comma-separated list of files to check for machine-id. Use the first one that exists. (default "/etc/machine-id,/var/lib/dbus/machine-id")
I1108 22:50:51.109] [3]   -master-os-distro string
I1108 22:50:51.110] [3]     	The OS distribution of cluster master (debian, ubuntu, gci, coreos, or custom). (default "debian")
I1108 22:50:51.110] [3]   -master-tag string
I1108 22:50:51.110] [3]     	Network tags used on master instances. Valid only for gce, gke
... skipping 93 lines ...
I1108 22:50:51.125] [4]     	If true, adds the file directory to the header
I1108 22:50:51.125] [4]   -allow-gathering-profiles
I1108 22:50:51.125] [4]     	If set to true framework will allow to gather CPU/memory allocation pprof profiles from the master. (default true)
I1108 22:50:51.125] [4]   -allowed-not-ready-nodes int
I1108 22:50:51.125] [4]     	If non-zero, framework will allow for that many non-ready nodes when checking for all ready nodes.
I1108 22:50:51.125] [4]   -alsologtostderr
I1108 22:50:51.126] [4]     	log to standard error as well as files
I1108 22:50:51.126] [4]   -application_metrics_count_limit int
I1108 22:50:51.126] [4]     	Max number of application metrics to store (per container) (default 100)
I1108 22:50:51.126] [4]   -boot_id_file string
I1108 22:50:51.126] [4]     	Comma-separated list of files to check for boot-id. Use the first one that exists. (default "/proc/sys/kernel/random/boot_id")
I1108 22:50:51.126] [4]   -cert-dir string
I1108 22:50:51.127] [4]     	Path to the directory containing the certs. Default is empty, which doesn't use certs.
I1108 22:50:51.127] [4]   -clean-start
I1108 22:50:51.127] [4]     	If true, purge all namespaces except default and system before running tests. This serves to Cleanup test namespaces from failed/interrupted e2e runs in a long-lived cluster.
I1108 22:50:51.127] [4]   -cloud-config-file string
I1108 22:50:51.127] [4]     	Cloud config file.  Only required if provider is azure.
I1108 22:50:51.128] [4]   -cloud-provider-gce-lb-src-cidrs value
I1108 22:50:51.128] [4]     	CIDRs opened in GCE firewall for LB traffic proxy & health checks (default 130.211.0.0/22,209.85.152.0/22,209.85.204.0/22,35.191.0.0/16)
I1108 22:50:51.128] [4]   -cluster-ip-range string
I1108 22:50:51.128] [4]     	A CIDR notation IP range from which to assign IPs in the cluster. (default "10.64.0.0/14")
... skipping 91 lines ...
I1108 22:50:51.142] [4]     	If set, ginkgo will emit node output to files when running in parallel.
I1108 22:50:51.142] [4]   -ginkgo.dryRun
I1108 22:50:51.142] [4]     	If set, ginkgo will walk the test hierarchy without actually running anything.  Best paired with -v.
I1108 22:50:51.142] [4]   -ginkgo.failFast
I1108 22:50:51.142] [4]     	If set, ginkgo will stop running a test suite after a failure occurs.
I1108 22:50:51.142] [4]   -ginkgo.failOnPending
I1108 22:50:51.143] [4]     	If set, ginkgo will mark the test suite as failed if any specs are pending.
I1108 22:50:51.143] [4]   -ginkgo.flakeAttempts int
I1108 22:50:51.143] [4]     	Make up to this many attempts to run each spec. Please note that if any of the attempts succeed, the suite will not be failed. But any failures will still be recorded. (default 1)
I1108 22:50:51.143] [4]   -ginkgo.focus string
I1108 22:50:51.143] [4]     	If set, ginkgo will only run specs that match this regular expression.
I1108 22:50:51.143] [4]   -ginkgo.noColor
I1108 22:50:51.144] [4]     	If set, suppress color output in default reporter.
I1108 22:50:51.144] [4]   -ginkgo.noisyPendings
I1108 22:50:51.144] [4]     	If set, default reporter will shout about pending tests. (default true)
... skipping 71 lines ...
I1108 22:50:51.154] [4]     	If non-empty, use this log file
I1108 22:50:51.154] [4]   -log_file_max_size uint
I1108 22:50:51.154] [4]     	Defines the maximum size a log file can grow to. Unit is megabytes. If the value is 0, the maximum file size is unlimited. (default 1800)
I1108 22:50:51.154] [4]   -logexporter-gcs-path string
I1108 22:50:51.155] [4]     	Path to the GCS artifacts directory to dump logs from nodes. Logexporter gets enabled if this is non-empty.
I1108 22:50:51.155] [4]   -logtostderr
I1108 22:50:51.155] [4]     	log to standard error instead of files (default true)
I1108 22:50:51.155] [4]   -machine_id_file string
I1108 22:50:51.155] [4]     	Comma-separated list of files to check for machine-id. Use the first one that exists. (default "/etc/machine-id,/var/lib/dbus/machine-id")
I1108 22:50:51.155] [4]   -master-os-distro string
I1108 22:50:51.156] [4]     	The OS distribution of cluster master (debian, ubuntu, gci, coreos, or custom). (default "debian")
I1108 22:50:51.156] [4]   -master-tag string
I1108 22:50:51.156] [4]     	Network tags used on master instances. Valid only for gce, gke
... skipping 93 lines ...
I1108 22:50:51.170] [5]     	If true, adds the file directory to the header
I1108 22:50:51.170] [5]   -allow-gathering-profiles
I1108 22:50:51.170] [5]     	If set to true framework will allow to gather CPU/memory allocation pprof profiles from the master. (default true)
I1108 22:50:51.170] [5]   -allowed-not-ready-nodes int
I1108 22:50:51.170] [5]     	If non-zero, framework will allow for that many non-ready nodes when checking for all ready nodes.
I1108 22:50:51.171] [5]   -alsologtostderr
I1108 22:50:51.171] [5]     	log to standard error as well as files
I1108 22:50:51.171] [5]   -application_metrics_count_limit int
I1108 22:50:51.171] [5]     	Max number of application metrics to store (per container) (default 100)
I1108 22:50:51.171] [5]   -boot_id_file string
I1108 22:50:51.171] [5]     	Comma-separated list of files to check for boot-id. Use the first one that exists. (default "/proc/sys/kernel/random/boot_id")
I1108 22:50:51.171] [5]   -cert-dir string
I1108 22:50:51.172] [5]     	Path to the directory containing the certs. Default is empty, which doesn't use certs.
I1108 22:50:51.172] [5]   -clean-start
I1108 22:50:51.172] [5]     	If true, purge all namespaces except default and system before running tests. This serves to Cleanup test namespaces from failed/interrupted e2e runs in a long-lived cluster.
I1108 22:50:51.172] [5]   -cloud-config-file string
I1108 22:50:51.172] [5]     	Cloud config file.  Only required if provider is azure.
I1108 22:50:51.172] [5]   -cloud-provider-gce-lb-src-cidrs value
I1108 22:50:51.173] [5]     	CIDRs opened in GCE firewall for LB traffic proxy & health checks (default 130.211.0.0/22,209.85.152.0/22,209.85.204.0/22,35.191.0.0/16)
I1108 22:50:51.173] [5]   -cluster-ip-range string
I1108 22:50:51.173] [5]     	A CIDR notation IP range from which to assign IPs in the cluster. (default "10.64.0.0/14")
... skipping 91 lines ...
I1108 22:50:51.187] [5]     	If set, ginkgo will emit node output to files when running in parallel.
I1108 22:50:51.187] [5]   -ginkgo.dryRun
I1108 22:50:51.187] [5]     	If set, ginkgo will walk the test hierarchy without actually running anything.  Best paired with -v.
I1108 22:50:51.187] [5]   -ginkgo.failFast
I1108 22:50:51.187] [5]     	If set, ginkgo will stop running a test suite after a failure occurs.
I1108 22:50:51.188] [5]   -ginkgo.failOnPending
I1108 22:50:51.188] [5]     	If set, ginkgo will mark the test suite as failed if any specs are pending.
I1108 22:50:51.188] [5]   -ginkgo.flakeAttempts int
I1108 22:50:51.188] [5]     	Make up to this many attempts to run each spec. Please note that if any of the attempts succeed, the suite will not be failed. But any failures will still be recorded. (default 1)
I1108 22:50:51.188] [5]   -ginkgo.focus string
I1108 22:50:51.188] [5]     	If set, ginkgo will only run specs that match this regular expression.
I1108 22:50:51.188] [5]   -ginkgo.noColor
I1108 22:50:51.189] [5]     	If set, suppress color output in default reporter.
I1108 22:50:51.189] [5]   -ginkgo.noisyPendings
I1108 22:50:51.189] [5]     	If set, default reporter will shout about pending tests. (default true)
... skipping 71 lines ...
I1108 22:50:51.199] [5]     	If non-empty, use this log file
I1108 22:50:51.199] [5]   -log_file_max_size uint
I1108 22:50:51.199] [5]     	Defines the maximum size a log file can grow to. Unit is megabytes. If the value is 0, the maximum file size is unlimited. (default 1800)
I1108 22:50:51.199] [5]   -logexporter-gcs-path string
I1108 22:50:51.199] [5]     	Path to the GCS artifacts directory to dump logs from nodes. Logexporter gets enabled if this is non-empty.
I1108 22:50:51.200] [5]   -logtostderr
I1108 22:50:51.200] [5]     	log to standard error instead of files (default true)
I1108 22:50:51.200] [5]   -machine_id_file string
I1108 22:50:51.200] [5]     	Comma-separated list of files to check for machine-id. Use the first one that exists. (default "/etc/machine-id,/var/lib/dbus/machine-id")
I1108 22:50:51.200] [5]   -master-os-distro string
I1108 22:50:51.200] [5]     	The OS distribution of cluster master (debian, ubuntu, gci, coreos, or custom). (default "debian")
I1108 22:50:51.200] [5]   -master-tag string
I1108 22:50:51.201] [5]     	Network tags used on master instances. Valid only for gce, gke
... skipping 93 lines ...
I1108 22:50:51.214] [6]     	If true, adds the file directory to the header
I1108 22:50:51.214] [6]   -allow-gathering-profiles
I1108 22:50:51.214] [6]     	If set to true framework will allow to gather CPU/memory allocation pprof profiles from the master. (default true)
I1108 22:50:51.214] [6]   -allowed-not-ready-nodes int
I1108 22:50:51.214] [6]     	If non-zero, framework will allow for that many non-ready nodes when checking for all ready nodes.
I1108 22:50:51.214] [6]   -alsologtostderr
I1108 22:50:51.215] [6]     	log to standard error as well as files
I1108 22:50:51.215] [6]   -application_metrics_count_limit int
I1108 22:50:51.215] [6]     	Max number of application metrics to store (per container) (default 100)
I1108 22:50:51.215] [6]   -boot_id_file string
I1108 22:50:51.215] [6]     	Comma-separated list of files to check for boot-id. Use the first one that exists. (default "/proc/sys/kernel/random/boot_id")
I1108 22:50:51.215] [6]   -cert-dir string
I1108 22:50:51.215] [6]     	Path to the directory containing the certs. Default is empty, which doesn't use certs.
I1108 22:50:51.216] [6]   -clean-start
I1108 22:50:51.216] [6]     	If true, purge all namespaces except default and system before running tests. This serves to Cleanup test namespaces from failed/interrupted e2e runs in a long-lived cluster.
I1108 22:50:51.216] [6]   -cloud-config-file string
I1108 22:50:51.216] [6]     	Cloud config file.  Only required if provider is azure.
I1108 22:50:51.216] [6]   -cloud-provider-gce-lb-src-cidrs value
I1108 22:50:51.216] [6]     	CIDRs opened in GCE firewall for LB traffic proxy & health checks (default 130.211.0.0/22,209.85.152.0/22,209.85.204.0/22,35.191.0.0/16)
I1108 22:50:51.216] [6]   -cluster-ip-range string
I1108 22:50:51.217] [6]     	A CIDR notation IP range from which to assign IPs in the cluster. (default "10.64.0.0/14")
... skipping 91 lines ...
I1108 22:50:51.230] [6]     	If set, ginkgo will emit node output to files when running in parallel.
I1108 22:50:51.230] [6]   -ginkgo.dryRun
I1108 22:50:51.230] [6]     	If set, ginkgo will walk the test hierarchy without actually running anything.  Best paired with -v.
I1108 22:50:51.230] [6]   -ginkgo.failFast
I1108 22:50:51.230] [6]     	If set, ginkgo will stop running a test suite after a failure occurs.
I1108 22:50:51.231] [6]   -ginkgo.failOnPending
I1108 22:50:51.231] [6]     	If set, ginkgo will mark the test suite as failed if any specs are pending.
I1108 22:50:51.231] [6]   -ginkgo.flakeAttempts int
I1108 22:50:51.231] [6]     	Make up to this many attempts to run each spec. Please note that if any of the attempts succeed, the suite will not be failed. But any failures will still be recorded. (default 1)
I1108 22:50:51.231] [6]   -ginkgo.focus string
I1108 22:50:51.231] [6]     	If set, ginkgo will only run specs that match this regular expression.
I1108 22:50:51.231] [6]   -ginkgo.noColor
I1108 22:50:51.232] [6]     	If set, suppress color output in default reporter.
I1108 22:50:51.232] [6]   -ginkgo.noisyPendings
I1108 22:50:51.232] [6]     	If set, default reporter will shout about pending tests. (default true)
... skipping 71 lines ...
I1108 22:50:51.242] [6]     	If non-empty, use this log file
I1108 22:50:51.242] [6]   -log_file_max_size uint
I1108 22:50:51.242] [6]     	Defines the maximum size a log file can grow to. Unit is megabytes. If the value is 0, the maximum file size is unlimited. (default 1800)
I1108 22:50:51.242] [6]   -logexporter-gcs-path string
I1108 22:50:51.242] [6]     	Path to the GCS artifacts directory to dump logs from nodes. Logexporter gets enabled if this is non-empty.
I1108 22:50:51.242] [6]   -logtostderr
I1108 22:50:51.242] [6]     	log to standard error instead of files (default true)
I1108 22:50:51.243] [6]   -machine_id_file string
I1108 22:50:51.243] [6]     	Comma-separated list of files to check for machine-id. Use the first one that exists. (default "/etc/machine-id,/var/lib/dbus/machine-id")
I1108 22:50:51.243] [6]   -master-os-distro string
I1108 22:50:51.243] [6]     	The OS distribution of cluster master (debian, ubuntu, gci, coreos, or custom). (default "debian")
I1108 22:50:51.243] [6]   -master-tag string
I1108 22:50:51.243] [6]     	Network tags used on master instances. Valid only for gce, gke
... skipping 93 lines ...
I1108 22:50:51.257] [7]     	If true, adds the file directory to the header
I1108 22:50:51.257] [7]   -allow-gathering-profiles
I1108 22:50:51.257] [7]     	If set to true framework will allow to gather CPU/memory allocation pprof profiles from the master. (default true)
I1108 22:50:51.257] [7]   -allowed-not-ready-nodes int
I1108 22:50:51.257] [7]     	If non-zero, framework will allow for that many non-ready nodes when checking for all ready nodes.
I1108 22:50:51.257] [7]   -alsologtostderr
I1108 22:50:51.257] [7]     	log to standard error as well as files
I1108 22:50:51.257] [7]   -application_metrics_count_limit int
I1108 22:50:51.258] [7]     	Max number of application metrics to store (per container) (default 100)
I1108 22:50:51.258] [7]   -boot_id_file string
I1108 22:50:51.258] [7]     	Comma-separated list of files to check for boot-id. Use the first one that exists. (default "/proc/sys/kernel/random/boot_id")
I1108 22:50:51.258] [7]   -cert-dir string
I1108 22:50:51.258] [7]     	Path to the directory containing the certs. Default is empty, which doesn't use certs.
I1108 22:50:51.258] [7]   -clean-start
I1108 22:50:51.259] [7]     	If true, purge all namespaces except default and system before running tests. This serves to Cleanup test namespaces from failed/interrupted e2e runs in a long-lived cluster.
I1108 22:50:51.259] [7]   -cloud-config-file string
I1108 22:50:51.259] [7]     	Cloud config file.  Only required if provider is azure.
I1108 22:50:51.259] [7]   -cloud-provider-gce-lb-src-cidrs value
I1108 22:50:51.259] [7]     	CIDRs opened in GCE firewall for LB traffic proxy & health checks (default 130.211.0.0/22,209.85.152.0/22,209.85.204.0/22,35.191.0.0/16)
I1108 22:50:51.259] [7]   -cluster-ip-range string
I1108 22:50:51.259] [7]     	A CIDR notation IP range from which to assign IPs in the cluster. (default "10.64.0.0/14")
... skipping 91 lines ...
I1108 22:50:51.272] [7]     	If set, ginkgo will emit node output to files when running in parallel.
I1108 22:50:51.272] [7]   -ginkgo.dryRun
I1108 22:50:51.273] [7]     	If set, ginkgo will walk the test hierarchy without actually running anything.  Best paired with -v.
I1108 22:50:51.273] [7]   -ginkgo.failFast
I1108 22:50:51.273] [7]     	If set, ginkgo will stop running a test suite after a failure occurs.
I1108 22:50:51.273] [7]   -ginkgo.failOnPending
I1108 22:50:51.273] [7]     	If set, ginkgo will mark the test suite as failed if any specs are pending.
I1108 22:50:51.273] [7]   -ginkgo.flakeAttempts int
I1108 22:50:51.273] [7]     	Make up to this many attempts to run each spec. Please note that if any of the attempts succeed, the suite will not be failed. But any failures will still be recorded. (default 1)
I1108 22:50:51.274] [7]   -ginkgo.focus string
I1108 22:50:51.274] [7]     	If set, ginkgo will only run specs that match this regular expression.
I1108 22:50:51.274] [7]   -ginkgo.noColor
I1108 22:50:51.274] [7]     	If set, suppress color output in default reporter.
I1108 22:50:51.274] [7]   -ginkgo.noisyPendings
I1108 22:50:51.274] [7]     	If set, default reporter will shout about pending tests. (default true)
... skipping 71 lines ...
I1108 22:50:51.284] [7]     	If non-empty, use this log file
I1108 22:50:51.284] [7]   -log_file_max_size uint
I1108 22:50:51.284] [7]     	Defines the maximum size a log file can grow to. Unit is megabytes. If the value is 0, the maximum file size is unlimited. (default 1800)
I1108 22:50:51.284] [7]   -logexporter-gcs-path string
I1108 22:50:51.285] [7]     	Path to the GCS artifacts directory to dump logs from nodes. Logexporter gets enabled if this is non-empty.
I1108 22:50:51.285] [7]   -logtostderr
I1108 22:50:51.285] [7]     	log to standard error instead of files (default true)
I1108 22:50:51.285] [7]   -machine_id_file string
I1108 22:50:51.285] [7]     	Comma-separated list of files to check for machine-id. Use the first one that exists. (default "/etc/machine-id,/var/lib/dbus/machine-id")
I1108 22:50:51.285] [7]   -master-os-distro string
I1108 22:50:51.286] [7]     	The OS distribution of cluster master (debian, ubuntu, gci, coreos, or custom). (default "debian")
I1108 22:50:51.286] [7]   -master-tag string
I1108 22:50:51.286] [7]     	Network tags used on master instances. Valid only for gce, gke
... skipping 93 lines ...
I1108 22:50:51.299] [8]     	If true, adds the file directory to the header
I1108 22:50:51.299] [8]   -allow-gathering-profiles
I1108 22:50:51.299] [8]     	If set to true framework will allow to gather CPU/memory allocation pprof profiles from the master. (default true)
I1108 22:50:51.300] [8]   -allowed-not-ready-nodes int
I1108 22:50:51.300] [8]     	If non-zero, framework will allow for that many non-ready nodes when checking for all ready nodes.
I1108 22:50:51.300] [8]   -alsologtostderr
I1108 22:50:51.300] [8]     	log to standard error as well as files
I1108 22:50:51.300] [8]   -application_metrics_count_limit int
I1108 22:50:51.301] [8]     	Max number of application metrics to store (per container) (default 100)
I1108 22:50:51.301] [8]   -boot_id_file string
I1108 22:50:51.301] [8]     	Comma-separated list of files to check for boot-id. Use the first one that exists. (default "/proc/sys/kernel/random/boot_id")
I1108 22:50:51.301] [8]   -cert-dir string
I1108 22:50:51.301] [8]     	Path to the directory containing the certs. Default is empty, which doesn't use certs.
I1108 22:50:51.301] [8]   -clean-start
I1108 22:50:51.302] [8]     	If true, purge all namespaces except default and system before running tests. This serves to Cleanup test namespaces from failed/interrupted e2e runs in a long-lived cluster.
I1108 22:50:51.302] [8]   -cloud-config-file string
I1108 22:50:51.302] [8]     	Cloud config file.  Only required if provider is azure.
I1108 22:50:51.302] [8]   -cloud-provider-gce-lb-src-cidrs value
I1108 22:50:51.302] [8]     	CIDRs opened in GCE firewall for LB traffic proxy & health checks (default 130.211.0.0/22,209.85.152.0/22,209.85.204.0/22,35.191.0.0/16)
I1108 22:50:51.302] [8]   -cluster-ip-range string
I1108 22:50:51.303] [8]     	A CIDR notation IP range from which to assign IPs in the cluster. (default "10.64.0.0/14")
... skipping 91 lines ...
I1108 22:50:51.316] [8]     	If set, ginkgo will emit node output to files when running in parallel.
I1108 22:50:51.316] [8]   -ginkgo.dryRun
I1108 22:50:51.316] [8]     	If set, ginkgo will walk the test hierarchy without actually running anything.  Best paired with -v.
I1108 22:50:51.316] [8]   -ginkgo.failFast
I1108 22:50:51.316] [8]     	If set, ginkgo will stop running a test suite after a failure occurs.
I1108 22:50:51.316] [8]   -ginkgo.failOnPending
I1108 22:50:51.317] [8]     	If set, ginkgo will mark the test suite as failed if any specs are pending.
I1108 22:50:51.317] [8]   -ginkgo.flakeAttempts int
I1108 22:50:51.317] [8]     	Make up to this many attempts to run each spec. Please note that if any of the attempts succeed, the suite will not be failed. But any failures will still be recorded. (default 1)
I1108 22:50:51.317] [8]   -ginkgo.focus string
I1108 22:50:51.317] [8]     	If set, ginkgo will only run specs that match this regular expression.
I1108 22:50:51.317] [8]   -ginkgo.noColor
I1108 22:50:51.318] [8]     	If set, suppress color output in default reporter.
I1108 22:50:51.318] [8]   -ginkgo.noisyPendings
I1108 22:50:51.318] [8]     	If set, default reporter will shout about pending tests. (default true)
... skipping 71 lines ...
I1108 22:50:51.328] [8]     	If non-empty, use this log file
I1108 22:50:51.328] [8]   -log_file_max_size uint
I1108 22:50:51.328] [8]     	Defines the maximum size a log file can grow to. Unit is megabytes. If the value is 0, the maximum file size is unlimited. (default 1800)
I1108 22:50:51.328] [8]   -logexporter-gcs-path string
I1108 22:50:51.328] [8]     	Path to the GCS artifacts directory to dump logs from nodes. Logexporter gets enabled if this is non-empty.
I1108 22:50:51.329] [8]   -logtostderr
I1108 22:50:51.329] [8]     	log to standard error instead of files (default true)
I1108 22:50:51.329] [8]   -machine_id_file string
I1108 22:50:51.329] [8]     	Comma-separated list of files to check for machine-id. Use the first one that exists. (default "/etc/machine-id,/var/lib/dbus/machine-id")
I1108 22:50:51.329] [8]   -master-os-distro string
I1108 22:50:51.329] [8]     	The OS distribution of cluster master (debian, ubuntu, gci, coreos, or custom). (default "debian")
I1108 22:50:51.329] [8]   -master-tag string
I1108 22:50:51.330] [8]     	Network tags used on master instances. Valid only for gce, gke
... skipping 93 lines ...
I1108 22:50:51.343] [9]     	If true, adds the file directory to the header
I1108 22:50:51.343] [9]   -allow-gathering-profiles
I1108 22:50:51.344] [9]     	If set to true framework will allow to gather CPU/memory allocation pprof profiles from the master. (default true)
I1108 22:50:51.344] [9]   -allowed-not-ready-nodes int
I1108 22:50:51.344] [9]     	If non-zero, framework will allow for that many non-ready nodes when checking for all ready nodes.
I1108 22:50:51.344] [9]   -alsologtostderr
I1108 22:50:51.344] [9]     	log to standard error as well as files
I1108 22:50:51.344] [9]   -application_metrics_count_limit int
I1108 22:50:51.344] [9]     	Max number of application metrics to store (per container) (default 100)
I1108 22:50:51.344] [9]   -boot_id_file string
I1108 22:50:51.345] [9]     	Comma-separated list of files to check for boot-id. Use the first one that exists. (default "/proc/sys/kernel/random/boot_id")
I1108 22:50:51.345] [9]   -cert-dir string
I1108 22:50:51.345] [9]     	Path to the directory containing the certs. Default is empty, which doesn't use certs.
I1108 22:50:51.345] [9]   -clean-start
I1108 22:50:51.345] [9]     	If true, purge all namespaces except default and system before running tests. This serves to Cleanup test namespaces from failed/interrupted e2e runs in a long-lived cluster.
I1108 22:50:51.345] [9]   -cloud-config-file string
I1108 22:50:51.346] [9]     	Cloud config file.  Only required if provider is azure.
I1108 22:50:51.346] [9]   -cloud-provider-gce-lb-src-cidrs value
I1108 22:50:51.346] [9]     	CIDRs opened in GCE firewall for LB traffic proxy & health checks (default 130.211.0.0/22,209.85.152.0/22,209.85.204.0/22,35.191.0.0/16)
I1108 22:50:51.346] [9]   -cluster-ip-range string
I1108 22:50:51.346] [9]     	A CIDR notation IP range from which to assign IPs in the cluster. (default "10.64.0.0/14")
... skipping 91 lines ...
I1108 22:50:51.360] [9]     	If set, ginkgo will emit node output to files when running in parallel.
I1108 22:50:51.360] [9]   -ginkgo.dryRun
I1108 22:50:51.360] [9]     	If set, ginkgo will walk the test hierarchy without actually running anything.  Best paired with -v.
I1108 22:50:51.360] [9]   -ginkgo.failFast
I1108 22:50:51.360] [9]     	If set, ginkgo will stop running a test suite after a failure occurs.
I1108 22:50:51.361] [9]   -ginkgo.failOnPending
I1108 22:50:51.361] [9]     	If set, ginkgo will mark the test suite as failed if any specs are pending.
I1108 22:50:51.361] [9]   -ginkgo.flakeAttempts int
I1108 22:50:51.361] [9]     	Make up to this many attempts to run each spec. Please note that if any of the attempts succeed, the suite will not be failed. But any failures will still be recorded. (default 1)
I1108 22:50:51.361] [9]   -ginkgo.focus string
I1108 22:50:51.361] [9]     	If set, ginkgo will only run specs that match this regular expression.
I1108 22:50:51.361] [9]   -ginkgo.noColor
I1108 22:50:51.362] [9]     	If set, suppress color output in default reporter.
I1108 22:50:51.362] [9]   -ginkgo.noisyPendings
I1108 22:50:51.362] [9]     	If set, default reporter will shout about pending tests. (default true)
... skipping 71 lines ...
I1108 22:50:51.372] [9]     	If non-empty, use this log file
I1108 22:50:51.372] [9]   -log_file_max_size uint
I1108 22:50:51.372] [9]     	Defines the maximum size a log file can grow to. Unit is megabytes. If the value is 0, the maximum file size is unlimited. (default 1800)
I1108 22:50:51.372] [9]   -logexporter-gcs-path string
I1108 22:50:51.372] [9]     	Path to the GCS artifacts directory to dump logs from nodes. Logexporter gets enabled if this is non-empty.
I1108 22:50:51.372] [9]   -logtostderr
I1108 22:50:51.372] [9]     	log to standard error instead of files (default true)
I1108 22:50:51.373] [9]   -machine_id_file string
I1108 22:50:51.373] [9]     	Comma-separated list of files to check for machine-id. Use the first one that exists. (default "/etc/machine-id,/var/lib/dbus/machine-id")
I1108 22:50:51.373] [9]   -master-os-distro string
I1108 22:50:51.373] [9]     	The OS distribution of cluster master (debian, ubuntu, gci, coreos, or custom). (default "debian")
I1108 22:50:51.373] [9]   -master-tag string
I1108 22:50:51.373] [9]     	Network tags used on master instances. Valid only for gce, gke
... skipping 93 lines ...
I1108 22:50:51.386] [10]     	If true, adds the file directory to the header
I1108 22:50:51.387] [10]   -allow-gathering-profiles
I1108 22:50:51.387] [10]     	If set to true framework will allow to gather CPU/memory allocation pprof profiles from the master. (default true)
I1108 22:50:51.387] [10]   -allowed-not-ready-nodes int
I1108 22:50:51.387] [10]     	If non-zero, framework will allow for that many non-ready nodes when checking for all ready nodes.
I1108 22:50:51.387] [10]   -alsologtostderr
I1108 22:50:51.387] [10]     	log to standard error as well as files
I1108 22:50:51.387] [10]   -application_metrics_count_limit int
I1108 22:50:51.388] [10]     	Max number of application metrics to store (per container) (default 100)
I1108 22:50:51.388] [10]   -boot_id_file string
I1108 22:50:51.388] [10]     	Comma-separated list of files to check for boot-id. Use the first one that exists. (default "/proc/sys/kernel/random/boot_id")
I1108 22:50:51.388] [10]   -cert-dir string
I1108 22:50:51.388] [10]     	Path to the directory containing the certs. Default is empty, which doesn't use certs.
I1108 22:50:51.388] [10]   -clean-start
I1108 22:50:51.389] [10]     	If true, purge all namespaces except default and system before running tests. This serves to Cleanup test namespaces from failed/interrupted e2e runs in a long-lived cluster.
I1108 22:50:51.389] [10]   -cloud-config-file string
I1108 22:50:51.389] [10]     	Cloud config file.  Only required if provider is azure.
I1108 22:50:51.389] [10]   -cloud-provider-gce-lb-src-cidrs value
I1108 22:50:51.389] [10]     	CIDRs opened in GCE firewall for LB traffic proxy & health checks (default 130.211.0.0/22,209.85.152.0/22,209.85.204.0/22,35.191.0.0/16)
I1108 22:50:51.389] [10]   -cluster-ip-range string
I1108 22:50:51.389] [10]     	A CIDR notation IP range from which to assign IPs in the cluster. (default "10.64.0.0/14")
... skipping 91 lines ...
I1108 22:50:51.403] [10]     	If set, ginkgo will emit node output to files when running in parallel.
I1108 22:50:51.403] [10]   -ginkgo.dryRun
I1108 22:50:51.403] [10]     	If set, ginkgo will walk the test hierarchy without actually running anything.  Best paired with -v.
I1108 22:50:51.403] [10]   -ginkgo.failFast
I1108 22:50:51.404] [10]     	If set, ginkgo will stop running a test suite after a failure occurs.
I1108 22:50:51.404] [10]   -ginkgo.failOnPending
I1108 22:50:51.404] [10]     	If set, ginkgo will mark the test suite as failed if any specs are pending.
I1108 22:50:51.404] [10]   -ginkgo.flakeAttempts int
I1108 22:50:51.404] [10]     	Make up to this many attempts to run each spec. Please note that if any of the attempts succeed, the suite will not be failed. But any failures will still be recorded. (default 1)
I1108 22:50:51.404] [10]   -ginkgo.focus string
I1108 22:50:51.404] [10]     	If set, ginkgo will only run specs that match this regular expression.
I1108 22:50:51.405] [10]   -ginkgo.noColor
I1108 22:50:51.405] [10]     	If set, suppress color output in default reporter.
I1108 22:50:51.405] [10]   -ginkgo.noisyPendings
I1108 22:50:51.405] [10]     	If set, default reporter will shout about pending tests. (default true)
... skipping 71 lines ...
I1108 22:50:51.415] [10]     	If non-empty, use this log file
I1108 22:50:51.415] [10]   -log_file_max_size uint
I1108 22:50:51.415] [10]     	Defines the maximum size a log file can grow to. Unit is megabytes. If the value is 0, the maximum file size is unlimited. (default 1800)
I1108 22:50:51.415] [10]   -logexporter-gcs-path string
I1108 22:50:51.415] [10]     	Path to the GCS artifacts directory to dump logs from nodes. Logexporter gets enabled if this is non-empty.
I1108 22:50:51.416] [10]   -logtostderr
I1108 22:50:51.416] [10]     	log to standard error instead of files (default true)
I1108 22:50:51.416] [10]   -machine_id_file string
I1108 22:50:51.416] [10]     	Comma-separated list of files to check for machine-id. Use the first one that exists. (default "/etc/machine-id,/var/lib/dbus/machine-id")
I1108 22:50:51.416] [10]   -master-os-distro string
I1108 22:50:51.416] [10]     	The OS distribution of cluster master (debian, ubuntu, gci, coreos, or custom). (default "debian")
I1108 22:50:51.416] [10]   -master-tag string
I1108 22:50:51.417] [10]     	Network tags used on master instances. Valid only for gce, gke
... skipping 93 lines ...
I1108 22:50:51.433] [11]     	If true, adds the file directory to the header
I1108 22:50:51.433] [11]   -allow-gathering-profiles
I1108 22:50:51.434] [11]     	If set to true framework will allow to gather CPU/memory allocation pprof profiles from the master. (default true)
I1108 22:50:51.434] [11]   -allowed-not-ready-nodes int
I1108 22:50:51.434] [11]     	If non-zero, framework will allow for that many non-ready nodes when checking for all ready nodes.
I1108 22:50:51.434] [11]   -alsologtostderr
I1108 22:50:51.434] [11]     	log to standard error as well as files
I1108 22:50:51.434] [11]   -application_metrics_count_limit int
I1108 22:50:51.434] [11]     	Max number of application metrics to store (per container) (default 100)
I1108 22:50:51.435] [11]   -boot_id_file string
I1108 22:50:51.435] [11]     	Comma-separated list of files to check for boot-id. Use the first one that exists. (default "/proc/sys/kernel/random/boot_id")
I1108 22:50:51.435] [11]   -cert-dir string
I1108 22:50:51.435] [11]     	Path to the directory containing the certs. Default is empty, which doesn't use certs.
I1108 22:50:51.435] [11]   -clean-start
I1108 22:50:51.435] [11]     	If true, purge all namespaces except default and system before running tests. This serves to Cleanup test namespaces from failed/interrupted e2e runs in a long-lived cluster.
I1108 22:50:51.435] [11]   -cloud-config-file string
I1108 22:50:51.435] [11]     	Cloud config file.  Only required if provider is azure.
I1108 22:50:51.436] [11]   -cloud-provider-gce-lb-src-cidrs value
I1108 22:50:51.436] [11]     	CIDRs opened in GCE firewall for LB traffic proxy & health checks (default 130.211.0.0/22,209.85.152.0/22,209.85.204.0/22,35.191.0.0/16)
I1108 22:50:51.436] [11]   -cluster-ip-range string
I1108 22:50:51.436] [11]     	A CIDR notation IP range from which to assign IPs in the cluster. (default "10.64.0.0/14")
... skipping 91 lines ...
I1108 22:50:51.450] [11]     	If set, ginkgo will emit node output to files when running in parallel.
I1108 22:50:51.451] [11]   -ginkgo.dryRun
I1108 22:50:51.451] [11]     	If set, ginkgo will walk the test hierarchy without actually running anything.  Best paired with -v.
I1108 22:50:51.451] [11]   -ginkgo.failFast
I1108 22:50:51.451] [11]     	If set, ginkgo will stop running a test suite after a failure occurs.
I1108 22:50:51.451] [11]   -ginkgo.failOnPending
I1108 22:50:51.451] [11]     	If set, ginkgo will mark the test suite as failed if any specs are pending.
I1108 22:50:51.451] [11]   -ginkgo.flakeAttempts int
I1108 22:50:51.451] [11]     	Make up to this many attempts to run each spec. Please note that if any of the attempts succeed, the suite will not be failed. But any failures will still be recorded. (default 1)
I1108 22:50:51.452] [11]   -ginkgo.focus string
I1108 22:50:51.452] [11]     	If set, ginkgo will only run specs that match this regular expression.
I1108 22:50:51.452] [11]   -ginkgo.noColor
I1108 22:50:51.452] [11]     	If set, suppress color output in default reporter.
I1108 22:50:51.452] [11]   -ginkgo.noisyPendings
I1108 22:50:51.452] [11]     	If set, default reporter will shout about pending tests. (default true)
... skipping 71 lines ...
I1108 22:50:51.463] [11]     	If non-empty, use this log file
I1108 22:50:51.463] [11]   -log_file_max_size uint
I1108 22:50:51.463] [11]     	Defines the maximum size a log file can grow to. Unit is megabytes. If the value is 0, the maximum file size is unlimited. (default 1800)
I1108 22:50:51.464] [11]   -logexporter-gcs-path string
I1108 22:50:51.464] [11]     	Path to the GCS artifacts directory to dump logs from nodes. Logexporter gets enabled if this is non-empty.
I1108 22:50:51.464] [11]   -logtostderr
I1108 22:50:51.464] [11]     	log to standard error instead of files (default true)
I1108 22:50:51.464] [11]   -machine_id_file string
I1108 22:50:51.464] [11]     	Comma-separated list of files to check for machine-id. Use the first one that exists. (default "/etc/machine-id,/var/lib/dbus/machine-id")
I1108 22:50:51.465] [11]   -master-os-distro string
I1108 22:50:51.465] [11]     	The OS distribution of cluster master (debian, ubuntu, gci, coreos, or custom). (default "debian")
I1108 22:50:51.465] [11]   -master-tag string
I1108 22:50:51.465] [11]     	Network tags used on master instances. Valid only for gce, gke
... skipping 93 lines ...
I1108 22:50:51.480] [12]     	If true, adds the file directory to the header
I1108 22:50:51.480] [12]   -allow-gathering-profiles
I1108 22:50:51.480] [12]     	If set to true framework will allow to gather CPU/memory allocation pprof profiles from the master. (default true)
I1108 22:50:51.480] [12]   -allowed-not-ready-nodes int
I1108 22:50:51.480] [12]     	If non-zero, framework will allow for that many non-ready nodes when checking for all ready nodes.
I1108 22:50:51.480] [12]   -alsologtostderr
I1108 22:50:51.481] [12]     	log to standard error as well as files
I1108 22:50:51.481] [12]   -application_metrics_count_limit int
I1108 22:50:51.481] [12]     	Max number of application metrics to store (per container) (default 100)
I1108 22:50:51.481] [12]   -boot_id_file string
I1108 22:50:51.481] [12]     	Comma-separated list of files to check for boot-id. Use the first one that exists. (default "/proc/sys/kernel/random/boot_id")
I1108 22:50:51.481] [12]   -cert-dir string
I1108 22:50:51.481] [12]     	Path to the directory containing the certs. Default is empty, which doesn't use certs.
I1108 22:50:51.482] [12]   -clean-start
I1108 22:50:51.482] [12]     	If true, purge all namespaces except default and system before running tests. This serves to Cleanup test namespaces from failed/interrupted e2e runs in a long-lived cluster.
I1108 22:50:51.482] [12]   -cloud-config-file string
I1108 22:50:51.482] [12]     	Cloud config file.  Only required if provider is azure.
I1108 22:50:51.482] [12]   -cloud-provider-gce-lb-src-cidrs value
I1108 22:50:51.482] [12]     	CIDRs opened in GCE firewall for LB traffic proxy & health checks (default 130.211.0.0/22,209.85.152.0/22,209.85.204.0/22,35.191.0.0/16)
I1108 22:50:51.483] [12]   -cluster-ip-range string
I1108 22:50:51.483] [12]     	A CIDR notation IP range from which to assign IPs in the cluster. (default "10.64.0.0/14")
... skipping 91 lines ...
I1108 22:50:51.496] [12]     	If set, ginkgo will emit node output to files when running in parallel.
I1108 22:50:51.496] [12]   -ginkgo.dryRun
I1108 22:50:51.496] [12]     	If set, ginkgo will walk the test hierarchy without actually running anything.  Best paired with -v.
I1108 22:50:51.496] [12]   -ginkgo.failFast
I1108 22:50:51.496] [12]     	If set, ginkgo will stop running a test suite after a failure occurs.
I1108 22:50:51.497] [12]   -ginkgo.failOnPending
I1108 22:50:51.497] [12]     	If set, ginkgo will mark the test suite as failed if any specs are pending.
I1108 22:50:51.497] [12]   -ginkgo.flakeAttempts int
I1108 22:50:51.497] [12]     	Make up to this many attempts to run each spec. Please note that if any of the attempts succeed, the suite will not be failed. But any failures will still be recorded. (default 1)
I1108 22:50:51.497] [12]   -ginkgo.focus string
I1108 22:50:51.497] [12]     	If set, ginkgo will only run specs that match this regular expression.
I1108 22:50:51.497] [12]   -ginkgo.noColor
I1108 22:50:51.498] [12]     	If set, suppress color output in default reporter.
I1108 22:50:51.498] [12]   -ginkgo.noisyPendings
I1108 22:50:51.498] [12]     	If set, default reporter will shout about pending tests. (default true)
... skipping 71 lines ...
I1108 22:50:51.508] [12]     	If non-empty, use this log file
I1108 22:50:51.509] [12]   -log_file_max_size uint
I1108 22:50:51.509] [12]     	Defines the maximum size a log file can grow to. Unit is megabytes. If the value is 0, the maximum file size is unlimited. (default 1800)
I1108 22:50:51.509] [12]   -logexporter-gcs-path string
I1108 22:50:51.509] [12]     	Path to the GCS artifacts directory to dump logs from nodes. Logexporter gets enabled if this is non-empty.
I1108 22:50:51.509] [12]   -logtostderr
I1108 22:50:51.509] [12]     	log to standard error instead of files (default true)
I1108 22:50:51.509] [12]   -machine_id_file string
I1108 22:50:51.510] [12]     	Comma-separated list of files to check for machine-id. Use the first one that exists. (default "/etc/machine-id,/var/lib/dbus/machine-id")
I1108 22:50:51.510] [12]   -master-os-distro string
I1108 22:50:51.510] [12]     	The OS distribution of cluster master (debian, ubuntu, gci, coreos, or custom). (default "debian")
I1108 22:50:51.510] [12]   -master-tag string
I1108 22:50:51.510] [12]     	Network tags used on master instances. Valid only for gce, gke
... skipping 93 lines ...
I1108 22:50:51.524] [13]     	If true, adds the file directory to the header
I1108 22:50:51.524] [13]   -allow-gathering-profiles
I1108 22:50:51.524] [13]     	If set to true framework will allow to gather CPU/memory allocation pprof profiles from the master. (default true)
I1108 22:50:51.524] [13]   -allowed-not-ready-nodes int
I1108 22:50:51.525] [13]     	If non-zero, framework will allow for that many non-ready nodes when checking for all ready nodes.
I1108 22:50:51.525] [13]   -alsologtostderr
I1108 22:50:51.525] [13]     	log to standard error as well as files
I1108 22:50:51.525] [13]   -application_metrics_count_limit int
I1108 22:50:51.525] [13]     	Max number of application metrics to store (per container) (default 100)
I1108 22:50:51.525] [13]   -boot_id_file string
I1108 22:50:51.526] [13]     	Comma-separated list of files to check for boot-id. Use the first one that exists. (default "/proc/sys/kernel/random/boot_id")
I1108 22:50:51.526] [13]   -cert-dir string
I1108 22:50:51.526] [13]     	Path to the directory containing the certs. Default is empty, which doesn't use certs.
I1108 22:50:51.526] [13]   -clean-start
I1108 22:50:51.526] [13]     	If true, purge all namespaces except default and system before running tests. This serves to Cleanup test namespaces from failed/interrupted e2e runs in a long-lived cluster.
I1108 22:50:51.526] [13]   -cloud-config-file string
I1108 22:50:51.526] [13]     	Cloud config file.  Only required if provider is azure.
I1108 22:50:51.527] [13]   -cloud-provider-gce-lb-src-cidrs value
I1108 22:50:51.527] [13]     	CIDRs opened in GCE firewall for LB traffic proxy & health checks (default 130.211.0.0/22,209.85.152.0/22,209.85.204.0/22,35.191.0.0/16)
I1108 22:50:51.527] [13]   -cluster-ip-range string
I1108 22:50:51.527] [13]     	A CIDR notation IP range from which to assign IPs in the cluster. (default "10.64.0.0/14")
... skipping 91 lines ...
I1108 22:50:51.540] [13]     	If set, ginkgo will emit node output to files when running in parallel.
I1108 22:50:51.540] [13]   -ginkgo.dryRun
I1108 22:50:51.540] [13]     	If set, ginkgo will walk the test hierarchy without actually running anything.  Best paired with -v.
I1108 22:50:51.540] [13]   -ginkgo.failFast
I1108 22:50:51.540] [13]     	If set, ginkgo will stop running a test suite after a failure occurs.
I1108 22:50:51.540] [13]   -ginkgo.failOnPending
I1108 22:50:51.540] [13]     	If set, ginkgo will mark the test suite as failed if any specs are pending.
I1108 22:50:51.540] [13]   -ginkgo.flakeAttempts int
I1108 22:50:51.541] [13]     	Make up to this many attempts to run each spec. Please note that if any of the attempts succeed, the suite will not be failed. But any failures will still be recorded. (default 1)
I1108 22:50:51.541] [13]   -ginkgo.focus string
I1108 22:50:51.541] [13]     	If set, ginkgo will only run specs that match this regular expression.
I1108 22:50:51.541] [13]   -ginkgo.noColor
I1108 22:50:51.541] [13]     	If set, suppress color output in default reporter.
I1108 22:50:51.541] [13]   -ginkgo.noisyPendings
I1108 22:50:51.541] [13]     	If set, default reporter will shout about pending tests. (default true)
... skipping 71 lines ...
I1108 22:50:51.548] [13]     	If non-empty, use this log file
I1108 22:50:51.548] [13]   -log_file_max_size uint
I1108 22:50:51.548] [13]     	Defines the maximum size a log file can grow to. Unit is megabytes. If the value is 0, the maximum file size is unlimited. (default 1800)
I1108 22:50:51.549] [13]   -logexporter-gcs-path string
I1108 22:50:51.549] [13]     	Path to the GCS artifacts directory to dump logs from nodes. Logexporter gets enabled if this is non-empty.
I1108 22:50:51.549] [13]   -logtostderr
I1108 22:50:51.549] [13]     	log to standard error instead of files (default true)
I1108 22:50:51.549] [13]   -machine_id_file string
I1108 22:50:51.549] [13]     	Comma-separated list of files to check for machine-id. Use the first one that exists. (default "/etc/machine-id,/var/lib/dbus/machine-id")
I1108 22:50:51.549] [13]   -master-os-distro string
I1108 22:50:51.549] [13]     	The OS distribution of cluster master (debian, ubuntu, gci, coreos, or custom). (default "debian")
I1108 22:50:51.549] [13]   -master-tag string
I1108 22:50:51.550] [13]     	Network tags used on master instances. Valid only for gce, gke
... skipping 93 lines ...
I1108 22:50:51.559] [14]     	If true, adds the file directory to the header
I1108 22:50:51.560] [14]   -allow-gathering-profiles
I1108 22:50:51.560] [14]     	If set to true framework will allow to gather CPU/memory allocation pprof profiles from the master. (default true)
I1108 22:50:51.560] [14]   -allowed-not-ready-nodes int
I1108 22:50:51.560] [14]     	If non-zero, framework will allow for that many non-ready nodes when checking for all ready nodes.
I1108 22:50:51.560] [14]   -alsologtostderr
I1108 22:50:51.560] [14]     	log to standard error as well as files
I1108 22:50:51.560] [14]   -application_metrics_count_limit int
I1108 22:50:51.560] [14]     	Max number of application metrics to store (per container) (default 100)
I1108 22:50:51.560] [14]   -boot_id_file string
I1108 22:50:51.561] [14]     	Comma-separated list of files to check for boot-id. Use the first one that exists. (default "/proc/sys/kernel/random/boot_id")
I1108 22:50:51.561] [14]   -cert-dir string
I1108 22:50:51.561] [14]     	Path to the directory containing the certs. Default is empty, which doesn't use certs.
I1108 22:50:51.561] [14]   -clean-start
I1108 22:50:51.561] [14]     	If true, purge all namespaces except default and system before running tests. This serves to Cleanup test namespaces from failed/interrupted e2e runs in a long-lived cluster.
I1108 22:50:51.561] [14]   -cloud-config-file string
I1108 22:50:51.561] [14]     	Cloud config file.  Only required if provider is azure.
I1108 22:50:51.561] [14]   -cloud-provider-gce-lb-src-cidrs value
I1108 22:50:51.561] [14]     	CIDRs opened in GCE firewall for LB traffic proxy & health checks (default 130.211.0.0/22,209.85.152.0/22,209.85.204.0/22,35.191.0.0/16)
I1108 22:50:51.562] [14]   -cluster-ip-range string
I1108 22:50:51.562] [14]     	A CIDR notation IP range from which to assign IPs in the cluster. (default "10.64.0.0/14")
... skipping 91 lines ...
I1108 22:50:51.571] [14]     	If set, ginkgo will emit node output to files when running in parallel.
I1108 22:50:51.571] [14]   -ginkgo.dryRun
I1108 22:50:51.571] [14]     	If set, ginkgo will walk the test hierarchy without actually running anything.  Best paired with -v.
I1108 22:50:51.572] [14]   -ginkgo.failFast
I1108 22:50:51.572] [14]     	If set, ginkgo will stop running a test suite after a failure occurs.
I1108 22:50:51.572] [14]   -ginkgo.failOnPending
I1108 22:50:51.572] [14]     	If set, ginkgo will mark the test suite as failed if any specs are pending.
I1108 22:50:51.572] [14]   -ginkgo.flakeAttempts int
I1108 22:50:51.572] [14]     	Make up to this many attempts to run each spec. Please note that if any of the attempts succeed, the suite will not be failed. But any failures will still be recorded. (default 1)
I1108 22:50:51.572] [14]   -ginkgo.focus string
I1108 22:50:51.572] [14]     	If set, ginkgo will only run specs that match this regular expression.
I1108 22:50:51.572] [14]   -ginkgo.noColor
I1108 22:50:51.572] [14]     	If set, suppress color output in default reporter.
I1108 22:50:51.573] [14]   -ginkgo.noisyPendings
I1108 22:50:51.573] [14]     	If set, default reporter will shout about pending tests. (default true)
... skipping 71 lines ...
I1108 22:50:51.581] [14]     	If non-empty, use this log file
I1108 22:50:51.581] [14]   -log_file_max_size uint
I1108 22:50:51.582] [14]     	Defines the maximum size a log file can grow to. Unit is megabytes. If the value is 0, the maximum file size is unlimited. (default 1800)
I1108 22:50:51.582] [14]   -logexporter-gcs-path string
I1108 22:50:51.582] [14]     	Path to the GCS artifacts directory to dump logs from nodes. Logexporter gets enabled if this is non-empty.
I1108 22:50:51.582] [14]   -logtostderr
I1108 22:50:51.582] [14]     	log to standard error instead of files (default true)
I1108 22:50:51.582] [14]   -machine_id_file string
I1108 22:50:51.582] [14]     	Comma-separated list of files to check for machine-id. Use the first one that exists. (default "/etc/machine-id,/var/lib/dbus/machine-id")
I1108 22:50:51.583] [14]   -master-os-distro string
I1108 22:50:51.583] [14]     	The OS distribution of cluster master (debian, ubuntu, gci, coreos, or custom). (default "debian")
I1108 22:50:51.583] [14]   -master-tag string
I1108 22:50:51.583] [14]     	Network tags used on master instances. Valid only for gce, gke
... skipping 93 lines ...
I1108 22:50:51.595] [15]     	If true, adds the file directory to the header
I1108 22:50:51.595] [15]   -allow-gathering-profiles
I1108 22:50:51.596] [15]     	If set to true framework will allow to gather CPU/memory allocation pprof profiles from the master. (default true)
I1108 22:50:51.596] [15]   -allowed-not-ready-nodes int
I1108 22:50:51.596] [15]     	If non-zero, framework will allow for that many non-ready nodes when checking for all ready nodes.
I1108 22:50:51.596] [15]   -alsologtostderr
I1108 22:50:51.596] [15]     	log to standard error as well as files
I1108 22:50:51.596] [15]   -application_metrics_count_limit int
I1108 22:50:51.596] [15]     	Max number of application metrics to store (per container) (default 100)
I1108 22:50:51.596] [15]   -boot_id_file string
I1108 22:50:51.596] [15]     	Comma-separated list of files to check for boot-id. Use the first one that exists. (default "/proc/sys/kernel/random/boot_id")
I1108 22:50:51.596] [15]   -cert-dir string
I1108 22:50:51.597] [15]     	Path to the directory containing the certs. Default is empty, which doesn't use certs.
I1108 22:50:51.597] [15]   -clean-start
I1108 22:50:51.597] [15]     	If true, purge all namespaces except default and system before running tests. This serves to Cleanup test namespaces from failed/interrupted e2e runs in a long-lived cluster.
I1108 22:50:51.597] [15]   -cloud-config-file string
I1108 22:50:51.597] [15]     	Cloud config file.  Only required if provider is azure.
I1108 22:50:51.597] [15]   -cloud-provider-gce-lb-src-cidrs value
I1108 22:50:51.597] [15]     	CIDRs opened in GCE firewall for LB traffic proxy & health checks (default 130.211.0.0/22,209.85.152.0/22,209.85.204.0/22,35.191.0.0/16)
I1108 22:50:51.597] [15]   -cluster-ip-range string
I1108 22:50:51.597] [15]     	A CIDR notation IP range from which to assign IPs in the cluster. (default "10.64.0.0/14")
... skipping 91 lines ...
I1108 22:50:51.607] [15]     	If set, ginkgo will emit node output to files when running in parallel.
I1108 22:50:51.607] [15]   -ginkgo.dryRun
I1108 22:50:51.607] [15]     	If set, ginkgo will walk the test hierarchy without actually running anything.  Best paired with -v.
I1108 22:50:51.607] [15]   -ginkgo.failFast
I1108 22:50:51.607] [15]     	If set, ginkgo will stop running a test suite after a failure occurs.
I1108 22:50:51.608] [15]   -ginkgo.failOnPending
I1108 22:50:51.608] [15]     	If set, ginkgo will mark the test suite as failed if any specs are pending.
I1108 22:50:51.608] [15]   -ginkgo.flakeAttempts int
I1108 22:50:51.608] [15]     	Make up to this many attempts to run each spec. Please note that if any of the attempts succeed, the suite will not be failed. But any failures will still be recorded. (default 1)
I1108 22:50:51.608] [15]   -ginkgo.focus string
I1108 22:50:51.608] [15]     	If set, ginkgo will only run specs that match this regular expression.
I1108 22:50:51.608] [15]   -ginkgo.noColor
I1108 22:50:51.608] [15]     	If set, suppress color output in default reporter.
I1108 22:50:51.608] [15]   -ginkgo.noisyPendings
I1108 22:50:51.609] [15]     	If set, default reporter will shout about pending tests. (default true)
... skipping 71 lines ...
I1108 22:50:51.616] [15]     	If non-empty, use this log file
I1108 22:50:51.616] [15]   -log_file_max_size uint
I1108 22:50:51.616] [15]     	Defines the maximum size a log file can grow to. Unit is megabytes. If the value is 0, the maximum file size is unlimited. (default 1800)
I1108 22:50:51.616] [15]   -logexporter-gcs-path string
I1108 22:50:51.616] [15]     	Path to the GCS artifacts directory to dump logs from nodes. Logexporter gets enabled if this is non-empty.
I1108 22:50:51.616] [15]   -logtostderr
I1108 22:50:51.616] [15]     	log to standard error instead of files (default true)
I1108 22:50:51.616] [15]   -machine_id_file string
I1108 22:50:51.616] [15]     	Comma-separated list of files to check for machine-id. Use the first one that exists. (default "/etc/machine-id,/var/lib/dbus/machine-id")
I1108 22:50:51.617] [15]   -master-os-distro string
I1108 22:50:51.617] [15]     	The OS distribution of cluster master (debian, ubuntu, gci, coreos, or custom). (default "debian")
I1108 22:50:51.617] [15]   -master-tag string
I1108 22:50:51.617] [15]     	Network tags used on master instances. Valid only for gce, gke
... skipping 93 lines ...
I1108 22:50:51.627] [16]     	If true, adds the file directory to the header
I1108 22:50:51.627] [16]   -allow-gathering-profiles
I1108 22:50:51.627] [16]     	If set to true framework will allow to gather CPU/memory allocation pprof profiles from the master. (default true)
I1108 22:50:51.627] [16]   -allowed-not-ready-nodes int
I1108 22:50:51.627] [16]     	If non-zero, framework will allow for that many non-ready nodes when checking for all ready nodes.
I1108 22:50:51.627] [16]   -alsologtostderr
I1108 22:50:51.627] [16]     	log to standard error as well as files
I1108 22:50:51.627] [16]   -application_metrics_count_limit int
I1108 22:50:51.627] [16]     	Max number of application metrics to store (per container) (default 100)
I1108 22:50:51.627] [16]   -boot_id_file string
I1108 22:50:51.628] [16]     	Comma-separated list of files to check for boot-id. Use the first one that exists. (default "/proc/sys/kernel/random/boot_id")
I1108 22:50:51.628] [16]   -cert-dir string
I1108 22:50:51.628] [16]     	Path to the directory containing the certs. Default is empty, which doesn't use certs.
I1108 22:50:51.628] [16]   -clean-start
I1108 22:50:51.628] [16]     	If true, purge all namespaces except default and system before running tests. This serves to Cleanup test namespaces from failed/interrupted e2e runs in a long-lived cluster.
I1108 22:50:51.628] [16]   -cloud-config-file string
I1108 22:50:51.628] [16]     	Cloud config file.  Only required if provider is azure.
I1108 22:50:51.628] [16]   -cloud-provider-gce-lb-src-cidrs value
I1108 22:50:51.629] [16]     	CIDRs opened in GCE firewall for LB traffic proxy & health checks (default 130.211.0.0/22,209.85.152.0/22,209.85.204.0/22,35.191.0.0/16)
I1108 22:50:51.629] [16]   -cluster-ip-range string
I1108 22:50:51.629] [16]     	A CIDR notation IP range from which to assign IPs in the cluster. (default "10.64.0.0/14")
... skipping 91 lines ...
I1108 22:50:51.638] [16]     	If set, ginkgo will emit node output to files when running in parallel.
I1108 22:50:51.638] [16]   -ginkgo.dryRun
I1108 22:50:51.639] [16]     	If set, ginkgo will walk the test hierarchy without actually running anything.  Best paired with -v.
I1108 22:50:51.639] [16]   -ginkgo.failFast
I1108 22:50:51.639] [16]     	If set, ginkgo will stop running a test suite after a failure occurs.
I1108 22:50:51.639] [16]   -ginkgo.failOnPending
I1108 22:50:51.639] [16]     	If set, ginkgo will mark the test suite as failed if any specs are pending.
I1108 22:50:51.639] [16]   -ginkgo.flakeAttempts int
I1108 22:50:51.639] [16]     	Make up to this many attempts to run each spec. Please note that if any of the attempts succeed, the suite will not be failed. But any failures will still be recorded. (default 1)
I1108 22:50:51.639] [16]   -ginkgo.focus string
I1108 22:50:51.639] [16]     	If set, ginkgo will only run specs that match this regular expression.
I1108 22:50:51.640] [16]   -ginkgo.noColor
I1108 22:50:51.640] [16]     	If set, suppress color output in default reporter.
I1108 22:50:51.640] [16]   -ginkgo.noisyPendings
I1108 22:50:51.640] [16]     	If set, default reporter will shout about pending tests. (default true)
... skipping 71 lines ...
I1108 22:50:51.647] [16]     	If non-empty, use this log file
I1108 22:50:51.647] [16]   -log_file_max_size uint
I1108 22:50:51.647] [16]     	Defines the maximum size a log file can grow to. Unit is megabytes. If the value is 0, the maximum file size is unlimited. (default 1800)
I1108 22:50:51.647] [16]   -logexporter-gcs-path string
I1108 22:50:51.647] [16]     	Path to the GCS artifacts directory to dump logs from nodes. Logexporter gets enabled if this is non-empty.
I1108 22:50:51.647] [16]   -logtostderr
I1108 22:50:51.648] [16]     	log to standard error instead of files (default true)
I1108 22:50:51.648] [16]   -machine_id_file string
I1108 22:50:51.648] [16]     	Comma-separated list of files to check for machine-id. Use the first one that exists. (default "/etc/machine-id,/var/lib/dbus/machine-id")
I1108 22:50:51.648] [16]   -master-os-distro string
I1108 22:50:51.648] [16]     	The OS distribution of cluster master (debian, ubuntu, gci, coreos, or custom). (default "debian")
I1108 22:50:51.648] [16]   -master-tag string
I1108 22:50:51.648] [16]     	Network tags used on master instances. Valid only for gce, gke
... skipping 93 lines ...
I1108 22:50:51.658] [17]     	If true, adds the file directory to the header
I1108 22:50:51.658] [17]   -allow-gathering-profiles
I1108 22:50:51.658] [17]     	If set to true framework will allow to gather CPU/memory allocation pprof profiles from the master. (default true)
I1108 22:50:51.658] [17]   -allowed-not-ready-nodes int
I1108 22:50:51.658] [17]     	If non-zero, framework will allow for that many non-ready nodes when checking for all ready nodes.
I1108 22:50:51.658] [17]   -alsologtostderr
I1108 22:50:51.659] [17]     	log to standard error as well as files
I1108 22:50:51.659] [17]   -application_metrics_count_limit int
I1108 22:50:51.659] [17]     	Max number of application metrics to store (per container) (default 100)
I1108 22:50:51.659] [17]   -boot_id_file string
I1108 22:50:51.659] [17]     	Comma-separated list of files to check for boot-id. Use the first one that exists. (default "/proc/sys/kernel/random/boot_id")
I1108 22:50:51.659] [17]   -cert-dir string
I1108 22:50:51.659] [17]     	Path to the directory containing the certs. Default is empty, which doesn't use certs.
I1108 22:50:51.659] [17]   -clean-start
I1108 22:50:51.659] [17]     	If true, purge all namespaces except default and system before running tests. This serves to Cleanup test namespaces from failed/interrupted e2e runs in a long-lived cluster.
I1108 22:50:51.660] [17]   -cloud-config-file string
I1108 22:50:51.660] [17]     	Cloud config file.  Only required if provider is azure.
I1108 22:50:51.660] [17]   -cloud-provider-gce-lb-src-cidrs value
I1108 22:50:51.660] [17]     	CIDRs opened in GCE firewall for LB traffic proxy & health checks (default 130.211.0.0/22,209.85.152.0/22,209.85.204.0/22,35.191.0.0/16)
I1108 22:50:51.660] [17]   -cluster-ip-range string
I1108 22:50:51.660] [17]     	A CIDR notation IP range from which to assign IPs in the cluster. (default "10.64.0.0/14")
... skipping 91 lines ...
I1108 22:50:51.670] [17]     	If set, ginkgo will emit node output to files when running in parallel.
I1108 22:50:51.670] [17]   -ginkgo.dryRun
I1108 22:50:51.670] [17]     	If set, ginkgo will walk the test hierarchy without actually running anything.  Best paired with -v.
I1108 22:50:51.670] [17]   -ginkgo.failFast
I1108 22:50:51.670] [17]     	If set, ginkgo will stop running a test suite after a failure occurs.
I1108 22:50:51.670] [17]   -ginkgo.failOnPending
I1108 22:50:51.670] [17]     	If set, ginkgo will mark the test suite as failed if any specs are pending.
I1108 22:50:51.670] [17]   -ginkgo.flakeAttempts int
I1108 22:50:51.671] [17]     	Make up to this many attempts to run each spec. Please note that if any of the attempts succeed, the suite will not be failed. But any failures will still be recorded. (default 1)
I1108 22:50:51.671] [17]   -ginkgo.focus string
I1108 22:50:51.671] [17]     	If set, ginkgo will only run specs that match this regular expression.
I1108 22:50:51.671] [17]   -ginkgo.noColor
I1108 22:50:51.671] [17]     	If set, suppress color output in default reporter.
I1108 22:50:51.671] [17]   -ginkgo.noisyPendings
I1108 22:50:51.671] [17]     	If set, default reporter will shout about pending tests. (default true)
... skipping 71 lines ...
I1108 22:50:51.678] [17]     	If non-empty, use this log file
I1108 22:50:51.678] [17]   -log_file_max_size uint
I1108 22:50:51.679] [17]     	Defines the maximum size a log file can grow to. Unit is megabytes. If the value is 0, the maximum file size is unlimited. (default 1800)
I1108 22:50:51.679] [17]   -logexporter-gcs-path string
I1108 22:50:51.679] [17]     	Path to the GCS artifacts directory to dump logs from nodes. Logexporter gets enabled if this is non-empty.
I1108 22:50:51.679] [17]   -logtostderr
I1108 22:50:51.679] [17]     	log to standard error instead of files (default true)
I1108 22:50:51.679] [17]   -machine_id_file string
I1108 22:50:51.679] [17]     	Comma-separated list of files to check for machine-id. Use the first one that exists. (default "/etc/machine-id,/var/lib/dbus/machine-id")
I1108 22:50:51.679] [17]   -master-os-distro string
I1108 22:50:51.679] [17]     	The OS distribution of cluster master (debian, ubuntu, gci, coreos, or custom). (default "debian")
I1108 22:50:51.680] [17]   -master-tag string
I1108 22:50:51.680] [17]     	Network tags used on master instances. Valid only for gce, gke
... skipping 93 lines ...
I1108 22:50:51.689] [18]     	If true, adds the file directory to the header
I1108 22:50:51.689] [18]   -allow-gathering-profiles
I1108 22:50:51.690] [18]     	If set to true framework will allow to gather CPU/memory allocation pprof profiles from the master. (default true)
I1108 22:50:51.690] [18]   -allowed-not-ready-nodes int
I1108 22:50:51.690] [18]     	If non-zero, framework will allow for that many non-ready nodes when checking for all ready nodes.
I1108 22:50:51.690] [18]   -alsologtostderr
I1108 22:50:51.690] [18]     	log to standard error as well as files
I1108 22:50:51.690] [18]   -application_metrics_count_limit int
I1108 22:50:51.690] [18]     	Max number of application metrics to store (per container) (default 100)
I1108 22:50:51.690] [18]   -boot_id_file string
I1108 22:50:51.691] [18]     	Comma-separated list of files to check for boot-id. Use the first one that exists. (default "/proc/sys/kernel/random/boot_id")
I1108 22:50:51.691] [18]   -cert-dir string
I1108 22:50:51.691] [18]     	Path to the directory containing the certs. Default is empty, which doesn't use certs.
I1108 22:50:51.691] [18]   -clean-start
I1108 22:50:51.691] [18]     	If true, purge all namespaces except default and system before running tests. This serves to Cleanup test namespaces from failed/interrupted e2e runs in a long-lived cluster.
I1108 22:50:51.691] [18]   -cloud-config-file string
I1108 22:50:51.691] [18]     	Cloud config file.  Only required if provider is azure.
I1108 22:50:51.691] [18]   -cloud-provider-gce-lb-src-cidrs value
I1108 22:50:51.691] [18]     	CIDRs opened in GCE firewall for LB traffic proxy & health checks (default 130.211.0.0/22,209.85.152.0/22,209.85.204.0/22,35.191.0.0/16)
I1108 22:50:51.692] [18]   -cluster-ip-range string
I1108 22:50:51.692] [18]     	A CIDR notation IP range from which to assign IPs in the cluster. (default "10.64.0.0/14")
... skipping 91 lines ...
I1108 22:50:51.702] [18]     	If set, ginkgo will emit node output to files when running in parallel.
I1108 22:50:51.702] [18]   -ginkgo.dryRun
I1108 22:50:51.702] [18]     	If set, ginkgo will walk the test hierarchy without actually running anything.  Best paired with -v.
I1108 22:50:51.702] [18]   -ginkgo.failFast
I1108 22:50:51.702] [18]     	If set, ginkgo will stop running a test suite after a failure occurs.
I1108 22:50:51.702] [18]   -ginkgo.failOnPending
I1108 22:50:51.702] [18]     	If set, ginkgo will mark the test suite as failed if any specs are pending.
I1108 22:50:51.702] [18]   -ginkgo.flakeAttempts int
I1108 22:50:51.703] [18]     	Make up to this many attempts to run each spec. Please note that if any of the attempts succeed, the suite will not be failed. But any failures will still be recorded. (default 1)
I1108 22:50:51.703] [18]   -ginkgo.focus string
I1108 22:50:51.703] [18]     	If set, ginkgo will only run specs that match this regular expression.
I1108 22:50:51.703] [18]   -ginkgo.noColor
I1108 22:50:51.703] [18]     	If set, suppress color output in default reporter.
I1108 22:50:51.703] [18]   -ginkgo.noisyPendings
I1108 22:50:51.703] [18]     	If set, default reporter will shout about pending tests. (default true)
... skipping 71 lines ...
I1108 22:50:51.710] [18]     	If non-empty, use this log file
I1108 22:50:51.710] [18]   -log_file_max_size uint
I1108 22:50:51.710] [18]     	Defines the maximum size a log file can grow to. Unit is megabytes. If the value is 0, the maximum file size is unlimited. (default 1800)
I1108 22:50:51.711] [18]   -logexporter-gcs-path string
I1108 22:50:51.711] [18]     	Path to the GCS artifacts directory to dump logs from nodes. Logexporter gets enabled if this is non-empty.
I1108 22:50:51.711] [18]   -logtostderr
I1108 22:50:51.711] [18]     	log to standard error instead of files (default true)
I1108 22:50:51.711] [18]   -machine_id_file string
I1108 22:50:51.711] [18]     	Comma-separated list of files to check for machine-id. Use the first one that exists. (default "/etc/machine-id,/var/lib/dbus/machine-id")
I1108 22:50:51.711] [18]   -master-os-distro string
I1108 22:50:51.711] [18]     	The OS distribution of cluster master (debian, ubuntu, gci, coreos, or custom). (default "debian")
I1108 22:50:51.711] [18]   -master-tag string
I1108 22:50:51.712] [18]     	Network tags used on master instances. Valid only for gce, gke
... skipping 93 lines ...
I1108 22:50:51.721] [19]     	If true, adds the file directory to the header
I1108 22:50:51.722] [19]   -allow-gathering-profiles
I1108 22:50:51.722] [19]     	If set to true framework will allow to gather CPU/memory allocation pprof profiles from the master. (default true)
I1108 22:50:51.722] [19]   -allowed-not-ready-nodes int
I1108 22:50:51.722] [19]     	If non-zero, framework will allow for that many non-ready nodes when checking for all ready nodes.
I1108 22:50:51.722] [19]   -alsologtostderr
I1108 22:50:51.722] [19]     	log to standard error as well as files
I1108 22:50:51.722] [19]   -application_metrics_count_limit int
I1108 22:50:51.722] [19]     	Max number of application metrics to store (per container) (default 100)
I1108 22:50:51.722] [19]   -boot_id_file string
I1108 22:50:51.723] [19]     	Comma-separated list of files to check for boot-id. Use the first one that exists. (default "/proc/sys/kernel/random/boot_id")
I1108 22:50:51.723] [19]   -cert-dir string
I1108 22:50:51.723] [19]     	Path to the directory containing the certs. Default is empty, which doesn't use certs.
I1108 22:50:51.723] [19]   -clean-start
I1108 22:50:51.723] [19]     	If true, purge all namespaces except default and system before running tests. This serves to Cleanup test namespaces from failed/interrupted e2e runs in a long-lived cluster.
I1108 22:50:51.723] [19]   -cloud-config-file string
I1108 22:50:51.723] [19]     	Cloud config file.  Only required if provider is azure.
I1108 22:50:51.723] [19]   -cloud-provider-gce-lb-src-cidrs value
I1108 22:50:51.724] [19]     	CIDRs opened in GCE firewall for LB traffic proxy & health checks (default 130.211.0.0/22,209.85.152.0/22,209.85.204.0/22,35.191.0.0/16)
I1108 22:50:51.724] [19]   -cluster-ip-range string
I1108 22:50:51.724] [19]     	A CIDR notation IP range from which to assign IPs in the cluster. (default "10.64.0.0/14")
... skipping 91 lines ...
I1108 22:50:51.734] [19]     	If set, ginkgo will emit node output to files when running in parallel.
I1108 22:50:51.734] [19]   -ginkgo.dryRun
I1108 22:50:51.734] [19]     	If set, ginkgo will walk the test hierarchy without actually running anything.  Best paired with -v.
I1108 22:50:51.734] [19]   -ginkgo.failFast
I1108 22:50:51.734] [19]     	If set, ginkgo will stop running a test suite after a failure occurs.
I1108 22:50:51.734] [19]   -ginkgo.failOnPending
I1108 22:50:51.734] [19]     	If set, ginkgo will mark the test suite as failed if any specs are pending.
I1108 22:50:51.735] [19]   -ginkgo.flakeAttempts int
I1108 22:50:51.735] [19]     	Make up to this many attempts to run each spec. Please note that if any of the attempts succeed, the suite will not be failed. But any failures will still be recorded. (default 1)
I1108 22:50:51.735] [19]   -ginkgo.focus string
I1108 22:50:51.735] [19]     	If set, ginkgo will only run specs that match this regular expression.
I1108 22:50:51.735] [19]   -ginkgo.noColor
I1108 22:50:51.735] [19]     	If set, suppress color output in default reporter.
I1108 22:50:51.735] [19]   -ginkgo.noisyPendings
I1108 22:50:51.735] [19]     	If set, default reporter will shout about pending tests. (default true)
... skipping 71 lines ...
I1108 22:50:51.746] [19]     	If non-empty, use this log file
I1108 22:50:51.746] [19]   -log_file_max_size uint
I1108 22:50:51.746] [19]     	Defines the maximum size a log file can grow to. Unit is megabytes. If the value is 0, the maximum file size is unlimited. (default 1800)
I1108 22:50:51.747] [19]   -logexporter-gcs-path string
I1108 22:50:51.747] [19]     	Path to the GCS artifacts directory to dump logs from nodes. Logexporter gets enabled if this is non-empty.
I1108 22:50:51.747] [19]   -logtostderr
I1108 22:50:51.747] [19]     	log to standard error instead of files (default true)
I1108 22:50:51.747] [19]   -machine_id_file string
I1108 22:50:51.747] [19]     	Comma-separated list of files to check for machine-id. Use the first one that exists. (default "/etc/machine-id,/var/lib/dbus/machine-id")
I1108 22:50:51.747] [19]   -master-os-distro string
I1108 22:50:51.748] [19]     	The OS distribution of cluster master (debian, ubuntu, gci, coreos, or custom). (default "debian")
I1108 22:50:51.748] [19]   -master-tag string
I1108 22:50:51.748] [19]     	Network tags used on master instances. Valid only for gce, gke
... skipping 93 lines ...
I1108 22:50:51.762] [20]     	If true, adds the file directory to the header
I1108 22:50:51.762] [20]   -allow-gathering-profiles
I1108 22:50:51.762] [20]     	If set to true framework will allow to gather CPU/memory allocation pprof profiles from the master. (default true)
I1108 22:50:51.762] [20]   -allowed-not-ready-nodes int
I1108 22:50:51.763] [20]     	If non-zero, framework will allow for that many non-ready nodes when checking for all ready nodes.
I1108 22:50:51.763] [20]   -alsologtostderr
I1108 22:50:51.763] [20]     	log to standard error as well as files
I1108 22:50:51.763] [20]   -application_metrics_count_limit int
I1108 22:50:51.763] [20]     	Max number of application metrics to store (per container) (default 100)
I1108 22:50:51.763] [20]   -boot_id_file string
I1108 22:50:51.763] [20]     	Comma-separated list of files to check for boot-id. Use the first one that exists. (default "/proc/sys/kernel/random/boot_id")
I1108 22:50:51.764] [20]   -cert-dir string
I1108 22:50:51.764] [20]     	Path to the directory containing the certs. Default is empty, which doesn't use certs.
I1108 22:50:51.764] [20]   -clean-start
I1108 22:50:51.764] [20]     	If true, purge all namespaces except default and system before running tests. This serves to Cleanup test namespaces from failed/interrupted e2e runs in a long-lived cluster.
I1108 22:50:51.764] [20]   -cloud-config-file string
I1108 22:50:51.764] [20]     	Cloud config file.  Only required if provider is azure.
I1108 22:50:51.764] [20]   -cloud-provider-gce-lb-src-cidrs value
I1108 22:50:51.765] [20]     	CIDRs opened in GCE firewall for LB traffic proxy & health checks (default 130.211.0.0/22,209.85.152.0/22,209.85.204.0/22,35.191.0.0/16)
I1108 22:50:51.765] [20]   -cluster-ip-range string
I1108 22:50:51.765] [20]     	A CIDR notation IP range from which to assign IPs in the cluster. (default "10.64.0.0/14")
... skipping 91 lines ...
I1108 22:50:51.778] [20]     	If set, ginkgo will emit node output to files when running in parallel.
I1108 22:50:51.778] [20]   -ginkgo.dryRun
I1108 22:50:51.778] [20]     	If set, ginkgo will walk the test hierarchy without actually running anything.  Best paired with -v.
I1108 22:50:51.779] [20]   -ginkgo.failFast
I1108 22:50:51.779] [20]     	If set, ginkgo will stop running a test suite after a failure occurs.
I1108 22:50:51.779] [20]   -ginkgo.failOnPending
I1108 22:50:51.779] [20]     	If set, ginkgo will mark the test suite as failed if any specs are pending.
I1108 22:50:51.779] [20]   -ginkgo.flakeAttempts int
I1108 22:50:51.779] [20]     	Make up to this many attempts to run each spec. Please note that if any of the attempts succeed, the suite will not be failed. But any failures will still be recorded. (default 1)
I1108 22:50:51.779] [20]   -ginkgo.focus string
I1108 22:50:51.780] [20]     	If set, ginkgo will only run specs that match this regular expression.
I1108 22:50:51.780] [20]   -ginkgo.noColor
I1108 22:50:51.780] [20]     	If set, suppress color output in default reporter.
I1108 22:50:51.780] [20]   -ginkgo.noisyPendings
I1108 22:50:51.780] [20]     	If set, default reporter will shout about pending tests. (default true)
... skipping 71 lines ...
I1108 22:50:51.793] [20]     	If non-empty, use this log file
I1108 22:50:51.793] [20]   -log_file_max_size uint
I1108 22:50:51.793] [20]     	Defines the maximum size a log file can grow to. Unit is megabytes. If the value is 0, the maximum file size is unlimited. (default 1800)
I1108 22:50:51.794] [20]   -logexporter-gcs-path string
I1108 22:50:51.794] [20]     	Path to the GCS artifacts directory to dump logs from nodes. Logexporter gets enabled if this is non-empty.
I1108 22:50:51.794] [20]   -logtostderr
I1108 22:50:51.794] [20]     	log to standard error instead of files (default true)
I1108 22:50:51.794] [20]   -machine_id_file string
I1108 22:50:51.795] [20]     	Comma-separated list of files to check for machine-id. Use the first one that exists. (default "/etc/machine-id,/var/lib/dbus/machine-id")
I1108 22:50:51.795] [20]   -master-os-distro string
I1108 22:50:51.795] [20]     	The OS distribution of cluster master (debian, ubuntu, gci, coreos, or custom). (default "debian")
I1108 22:50:51.795] [20]   -master-tag string
I1108 22:50:51.795] [20]     	Network tags used on master instances. Valid only for gce, gke
... skipping 93 lines ...
I1108 22:50:51.810] [21]     	If true, adds the file directory to the header
I1108 22:50:51.810] [21]   -allow-gathering-profiles
I1108 22:50:51.810] [21]     	If set to true framework will allow to gather CPU/memory allocation pprof profiles from the master. (default true)
I1108 22:50:51.811] [21]   -allowed-not-ready-nodes int
I1108 22:50:51.811] [21]     	If non-zero, framework will allow for that many non-ready nodes when checking for all ready nodes.
I1108 22:50:51.811] [21]   -alsologtostderr
I1108 22:50:51.811] [21]     	log to standard error as well as files
I1108 22:50:51.811] [21]   -application_metrics_count_limit int
I1108 22:50:51.811] [21]     	Max number of application metrics to store (per container) (default 100)
I1108 22:50:51.811] [21]   -boot_id_file string
I1108 22:50:51.812] [21]     	Comma-separated list of files to check for boot-id. Use the first one that exists. (default "/proc/sys/kernel/random/boot_id")
I1108 22:50:51.812] [21]   -cert-dir string
I1108 22:50:51.812] [21]     	Path to the directory containing the certs. Default is empty, which doesn't use certs.
I1108 22:50:51.812] [21]   -clean-start
I1108 22:50:51.812] [21]     	If true, purge all namespaces except default and system before running tests. This serves to Cleanup test namespaces from failed/interrupted e2e runs in a long-lived cluster.
I1108 22:50:51.812] [21]   -cloud-config-file string
I1108 22:50:51.813] [21]     	Cloud config file.  Only required if provider is azure.
I1108 22:50:51.813] [21]   -cloud-provider-gce-lb-src-cidrs value
I1108 22:50:51.813] [21]     	CIDRs opened in GCE firewall for LB traffic proxy & health checks (default 130.211.0.0/22,209.85.152.0/22,209.85.204.0/22,35.191.0.0/16)
I1108 22:50:51.813] [21]   -cluster-ip-range string
I1108 22:50:51.813] [21]     	A CIDR notation IP range from which to assign IPs in the cluster. (default "10.64.0.0/14")
... skipping 91 lines ...
I1108 22:50:51.827] [21]     	If set, ginkgo will emit node output to files when running in parallel.
I1108 22:50:51.827] [21]   -ginkgo.dryRun
I1108 22:50:51.827] [21]     	If set, ginkgo will walk the test hierarchy without actually running anything.  Best paired with -v.
I1108 22:50:51.827] [21]   -ginkgo.failFast
I1108 22:50:51.827] [21]     	If set, ginkgo will stop running a test suite after a failure occurs.
I1108 22:50:51.827] [21]   -ginkgo.failOnPending
I1108 22:50:51.827] [21]     	If set, ginkgo will mark the test suite as failed if any specs are pending.
I1108 22:50:51.827] [21]   -ginkgo.flakeAttempts int
I1108 22:50:51.828] [21]     	Make up to this many attempts to run each spec. Please note that if any of the attempts succeed, the suite will not be failed. But any failures will still be recorded. (default 1)
I1108 22:50:51.828] [21]   -ginkgo.focus string
I1108 22:50:51.828] [21]     	If set, ginkgo will only run specs that match this regular expression.
I1108 22:50:51.828] [21]   -ginkgo.noColor
I1108 22:50:51.828] [21]     	If set, suppress color output in default reporter.
I1108 22:50:51.828] [21]   -ginkgo.noisyPendings
I1108 22:50:51.829] [21]     	If set, default reporter will shout about pending tests. (default true)
... skipping 71 lines ...
I1108 22:50:51.839] [21]     	If non-empty, use this log file
I1108 22:50:51.839] [21]   -log_file_max_size uint
I1108 22:50:51.839] [21]     	Defines the maximum size a log file can grow to. Unit is megabytes. If the value is 0, the maximum file size is unlimited. (default 1800)
I1108 22:50:51.839] [21]   -logexporter-gcs-path string
I1108 22:50:51.839] [21]     	Path to the GCS artifacts directory to dump logs from nodes. Logexporter gets enabled if this is non-empty.
I1108 22:50:51.839] [21]   -logtostderr
I1108 22:50:51.840] [21]     	log to standard error instead of files (default true)
I1108 22:50:51.840] [21]   -machine_id_file string
I1108 22:50:51.840] [21]     	Comma-separated list of files to check for machine-id. Use the first one that exists. (default "/etc/machine-id,/var/lib/dbus/machine-id")
I1108 22:50:51.840] [21]   -master-os-distro string
I1108 22:50:51.840] [21]     	The OS distribution of cluster master (debian, ubuntu, gci, coreos, or custom). (default "debian")
I1108 22:50:51.840] [21]   -master-tag string
I1108 22:50:51.841] [21]     	Network tags used on master instances. Valid only for gce, gke
... skipping 93 lines ...
I1108 22:50:51.855] [22]     	If true, adds the file directory to the header
I1108 22:50:51.855] [22]   -allow-gathering-profiles
I1108 22:50:51.856] [22]     	If set to true framework will allow to gather CPU/memory allocation pprof profiles from the master. (default true)
I1108 22:50:51.856] [22]   -allowed-not-ready-nodes int
I1108 22:50:51.856] [22]     	If non-zero, framework will allow for that many non-ready nodes when checking for all ready nodes.
I1108 22:50:51.856] [22]   -alsologtostderr
I1108 22:50:51.856] [22]     	log to standard error as well as files
I1108 22:50:51.857] [22]   -application_metrics_count_limit int
I1108 22:50:51.857] [22]     	Max number of application metrics to store (per container) (default 100)
I1108 22:50:51.857] [22]   -boot_id_file string
I1108 22:50:51.857] [22]     	Comma-separated list of files to check for boot-id. Use the first one that exists. (default "/proc/sys/kernel/random/boot_id")
I1108 22:50:51.857] [22]   -cert-dir string
I1108 22:50:51.857] [22]     	Path to the directory containing the certs. Default is empty, which doesn't use certs.
I1108 22:50:51.858] [22]   -clean-start
I1108 22:50:51.858] [22]     	If true, purge all namespaces except default and system before running tests. This serves to Cleanup test namespaces from failed/interrupted e2e runs in a long-lived cluster.
I1108 22:50:51.858] [22]   -cloud-config-file string
I1108 22:50:51.858] [22]     	Cloud config file.  Only required if provider is azure.
I1108 22:50:51.858] [22]   -cloud-provider-gce-lb-src-cidrs value
I1108 22:50:51.859] [22]     	CIDRs opened in GCE firewall for LB traffic proxy & health checks (default 130.211.0.0/22,209.85.152.0/22,209.85.204.0/22,35.191.0.0/16)
I1108 22:50:51.859] [22]   -cluster-ip-range string
I1108 22:50:51.859] [22]     	A CIDR notation IP range from which to assign IPs in the cluster. (default "10.64.0.0/14")
... skipping 91 lines ...
I1108 22:50:51.875] [22]     	If set, ginkgo will emit node output to files when running in parallel.
I1108 22:50:51.875] [22]   -ginkgo.dryRun
I1108 22:50:51.876] [22]     	If set, ginkgo will walk the test hierarchy without actually running anything.  Best paired with -v.
I1108 22:50:51.876] [22]   -ginkgo.failFast
I1108 22:50:51.876] [22]     	If set, ginkgo will stop running a test suite after a failure occurs.
I1108 22:50:51.876] [22]   -ginkgo.failOnPending
I1108 22:50:51.876] [22]     	If set, ginkgo will mark the test suite as failed if any specs are pending.
I1108 22:50:51.876] [22]   -ginkgo.flakeAttempts int
I1108 22:50:51.877] [22]     	Make up to this many attempts to run each spec. Please note that if any of the attempts succeed, the suite will not be failed. But any failures will still be recorded. (default 1)
I1108 22:50:51.877] [22]   -ginkgo.focus string
I1108 22:50:51.877] [22]     	If set, ginkgo will only run specs that match this regular expression.
I1108 22:50:51.877] [22]   -ginkgo.noColor
I1108 22:50:51.877] [22]     	If set, suppress color output in default reporter.
I1108 22:50:51.877] [22]   -ginkgo.noisyPendings
I1108 22:50:51.878] [22]     	If set, default reporter will shout about pending tests. (default true)
... skipping 71 lines ...
I1108 22:50:51.890] [22]     	If non-empty, use this log file
I1108 22:50:51.890] [22]   -log_file_max_size uint
I1108 22:50:51.890] [22]     	Defines the maximum size a log file can grow to. Unit is megabytes. If the value is 0, the maximum file size is unlimited. (default 1800)
I1108 22:50:51.890] [22]   -logexporter-gcs-path string
I1108 22:50:51.890] [22]     	Path to the GCS artifacts directory to dump logs from nodes. Logexporter gets enabled if this is non-empty.
I1108 22:50:51.890] [22]   -logtostderr
I1108 22:50:51.891] [22]     	log to standard error instead of files (default true)
I1108 22:50:51.891] [22]   -machine_id_file string
I1108 22:50:51.891] [22]     	Comma-separated list of files to check for machine-id. Use the first one that exists. (default "/etc/machine-id,/var/lib/dbus/machine-id")
I1108 22:50:51.891] [22]   -master-os-distro string
I1108 22:50:51.891] [22]     	The OS distribution of cluster master (debian, ubuntu, gci, coreos, or custom). (default "debian")
I1108 22:50:51.891] [22]   -master-tag string
I1108 22:50:51.892] [22]     	Network tags used on master instances. Valid only for gce, gke
... skipping 93 lines ...
I1108 22:50:51.907] [23]     	If true, adds the file directory to the header
I1108 22:50:51.907] [23]   -allow-gathering-profiles
I1108 22:50:51.908] [23]     	If set to true framework will allow to gather CPU/memory allocation pprof profiles from the master. (default true)
I1108 22:50:51.908] [23]   -allowed-not-ready-nodes int
I1108 22:50:51.908] [23]     	If non-zero, framework will allow for that many non-ready nodes when checking for all ready nodes.
I1108 22:50:51.908] [23]   -alsologtostderr
I1108 22:50:51.908] [23]     	log to standard error as well as files
I1108 22:50:51.908] [23]   -application_metrics_count_limit int
I1108 22:50:51.909] [23]     	Max number of application metrics to store (per container) (default 100)
I1108 22:50:51.909] [23]   -boot_id_file string
I1108 22:50:51.909] [23]     	Comma-separated list of files to check for boot-id. Use the first one that exists. (default "/proc/sys/kernel/random/boot_id")
I1108 22:50:51.909] [23]   -cert-dir string
I1108 22:50:51.909] [23]     	Path to the directory containing the certs. Default is empty, which doesn't use certs.
I1108 22:50:51.909] [23]   -clean-start
I1108 22:50:51.910] [23]     	If true, purge all namespaces except default and system before running tests. This serves to Cleanup test namespaces from failed/interrupted e2e runs in a long-lived cluster.
I1108 22:50:51.910] [23]   -cloud-config-file string
I1108 22:50:51.910] [23]     	Cloud config file.  Only required if provider is azure.
I1108 22:50:51.910] [23]   -cloud-provider-gce-lb-src-cidrs value
I1108 22:50:51.910] [23]     	CIDRs opened in GCE firewall for LB traffic proxy & health checks (default 130.211.0.0/22,209.85.152.0/22,209.85.204.0/22,35.191.0.0/16)
I1108 22:50:51.911] [23]   -cluster-ip-range string
I1108 22:50:51.911] [23]     	A CIDR notation IP range from which to assign IPs in the cluster. (default "10.64.0.0/14")
... skipping 91 lines ...
I1108 22:50:51.926] [23]     	If set, ginkgo will emit node output to files when running in parallel.
I1108 22:50:51.926] [23]   -ginkgo.dryRun
I1108 22:50:51.926] [23]     	If set, ginkgo will walk the test hierarchy without actually running anything.  Best paired with -v.
I1108 22:50:51.926] [23]   -ginkgo.failFast
I1108 22:50:51.927] [23]     	If set, ginkgo will stop running a test suite after a failure occurs.
I1108 22:50:51.927] [23]   -ginkgo.failOnPending
I1108 22:50:51.927] [23]     	If set, ginkgo will mark the test suite as failed if any specs are pending.
I1108 22:50:51.927] [23]   -ginkgo.flakeAttempts int
I1108 22:50:51.927] [23]     	Make up to this many attempts to run each spec. Please note that if any of the attempts succeed, the suite will not be failed. But any failures will still be recorded. (default 1)
I1108 22:50:51.927] [23]   -ginkgo.focus string
I1108 22:50:51.928] [23]     	If set, ginkgo will only run specs that match this regular expression.
I1108 22:50:51.928] [23]   -ginkgo.noColor
I1108 22:50:51.928] [23]     	If set, suppress color output in default reporter.
I1108 22:50:51.928] [23]   -ginkgo.noisyPendings
I1108 22:50:51.928] [23]     	If set, default reporter will shout about pending tests. (default true)
... skipping 71 lines ...
I1108 22:50:51.940] [23]     	If non-empty, use this log file
I1108 22:50:51.940] [23]   -log_file_max_size uint
I1108 22:50:51.940] [23]     	Defines the maximum size a log file can grow to. Unit is megabytes. If the value is 0, the maximum file size is unlimited. (default 1800)
I1108 22:50:51.940] [23]   -logexporter-gcs-path string
I1108 22:50:51.941] [23]     	Path to the GCS artifacts directory to dump logs from nodes. Logexporter gets enabled if this is non-empty.
I1108 22:50:51.941] [23]   -logtostderr
I1108 22:50:51.941] [23]     	log to standard error instead of files (default true)
I1108 22:50:51.941] [23]   -machine_id_file string
I1108 22:50:51.941] [23]     	Comma-separated list of files to check for machine-id. Use the first one that exists. (default "/etc/machine-id,/var/lib/dbus/machine-id")
I1108 22:50:51.941] [23]   -master-os-distro string
I1108 22:50:51.942] [23]     	The OS distribution of cluster master (debian, ubuntu, gci, coreos, or custom). (default "debian")
I1108 22:50:51.942] [23]   -master-tag string
I1108 22:50:51.942] [23]     	Network tags used on master instances. Valid only for gce, gke
... skipping 93 lines ...
I1108 22:50:51.956] [24]     	If true, adds the file directory to the header
I1108 22:50:51.956] [24]   -allow-gathering-profiles
I1108 22:50:51.956] [24]     	If set to true framework will allow to gather CPU/memory allocation pprof profiles from the master. (default true)
I1108 22:50:51.956] [24]   -allowed-not-ready-nodes int
I1108 22:50:51.957] [24]     	If non-zero, framework will allow for that many non-ready nodes when checking for all ready nodes.
I1108 22:50:51.957] [24]   -alsologtostderr
I1108 22:50:51.957] [24]     	log to standard error as well as files
I1108 22:50:51.957] [24]   -application_metrics_count_limit int
I1108 22:50:51.957] [24]     	Max number of application metrics to store (per container) (default 100)
I1108 22:50:51.957] [24]   -boot_id_file string
I1108 22:50:51.958] [24]     	Comma-separated list of files to check for boot-id. Use the first one that exists. (default "/proc/sys/kernel/random/boot_id")
I1108 22:50:51.958] [24]   -cert-dir string
I1108 22:50:51.958] [24]     	Path to the directory containing the certs. Default is empty, which doesn't use certs.
I1108 22:50:51.958] [24]   -clean-start
I1108 22:50:51.958] [24]     	If true, purge all namespaces except default and system before running tests. This serves to Cleanup test namespaces from failed/interrupted e2e runs in a long-lived cluster.
I1108 22:50:51.958] [24]   -cloud-config-file string
I1108 22:50:51.959] [24]     	Cloud config file.  Only required if provider is azure.
I1108 22:50:51.959] [24]   -cloud-provider-gce-lb-src-cidrs value
I1108 22:50:51.959] [24]     	CIDRs opened in GCE firewall for LB traffic proxy & health checks (default 130.211.0.0/22,209.85.152.0/22,209.85.204.0/22,35.191.0.0/16)
I1108 22:50:51.959] [24]   -cluster-ip-range string
I1108 22:50:51.959] [24]     	A CIDR notation IP range from which to assign IPs in the cluster. (default "10.64.0.0/14")
... skipping 91 lines ...
I1108 22:50:51.973] [24]     	If set, ginkgo will emit node output to files when running in parallel.
I1108 22:50:51.973] [24]   -ginkgo.dryRun
I1108 22:50:51.973] [24]     	If set, ginkgo will walk the test hierarchy without actually running anything.  Best paired with -v.
I1108 22:50:51.974] [24]   -ginkgo.failFast
I1108 22:50:51.974] [24]     	If set, ginkgo will stop running a test suite after a failure occurs.
I1108 22:50:51.974] [24]   -ginkgo.failOnPending
I1108 22:50:51.974] [24]     	If set, ginkgo will mark the test suite as failed if any specs are pending.
I1108 22:50:51.974] [24]   -ginkgo.flakeAttempts int
I1108 22:50:51.974] [24]     	Make up to this many attempts to run each spec. Please note that if any of the attempts succeed, the suite will not be failed. But any failures will still be recorded. (default 1)
I1108 22:50:51.974] [24]   -ginkgo.focus string
I1108 22:50:51.975] [24]     	If set, ginkgo will only run specs that match this regular expression.
I1108 22:50:51.975] [24]   -ginkgo.noColor
I1108 22:50:51.975] [24]     	If set, suppress color output in default reporter.
I1108 22:50:51.975] [24]   -ginkgo.noisyPendings
I1108 22:50:51.975] [24]     	If set, default reporter will shout about pending tests. (default true)
... skipping 71 lines ...
I1108 22:50:51.988] [24]     	If non-empty, use this log file
I1108 22:50:51.988] [24]   -log_file_max_size uint
I1108 22:50:51.988] [24]     	Defines the maximum size a log file can grow to. Unit is megabytes. If the value is 0, the maximum file size is unlimited. (default 1800)
I1108 22:50:51.988] [24]   -logexporter-gcs-path string
I1108 22:50:51.989] [24]     	Path to the GCS artifacts directory to dump logs from nodes. Logexporter gets enabled if this is non-empty.
I1108 22:50:51.989] [24]   -logtostderr
I1108 22:50:51.989] [24]     	log to standard error instead of files (default true)
I1108 22:50:51.989] [24]   -machine_id_file string
I1108 22:50:51.989] [24]     	Comma-separated list of files to check for machine-id. Use the first one that exists. (default "/etc/machine-id,/var/lib/dbus/machine-id")
I1108 22:50:51.990] [24]   -master-os-distro string
I1108 22:50:51.990] [24]     	The OS distribution of cluster master (debian, ubuntu, gci, coreos, or custom). (default "debian")
I1108 22:50:51.990] [24]   -master-tag string
I1108 22:50:51.990] [24]     	Network tags used on master instances. Valid only for gce, gke
... skipping 93 lines ...
I1108 22:50:52.009] [25]     	If true, adds the file directory to the header
I1108 22:50:52.009] [25]   -allow-gathering-profiles
I1108 22:50:52.010] [25]     	If set to true framework will allow to gather CPU/memory allocation pprof profiles from the master. (default true)
I1108 22:50:52.010] [25]   -allowed-not-ready-nodes int
I1108 22:50:52.010] [25]     	If non-zero, framework will allow for that many non-ready nodes when checking for all ready nodes.
I1108 22:50:52.010] [25]   -alsologtostderr
I1108 22:50:52.010] [25]     	log to standard error as well as files
I1108 22:50:52.011] [25]   -application_metrics_count_limit int
I1108 22:50:52.011] [25]     	Max number of application metrics to store (per container) (default 100)
I1108 22:50:52.011] [25]   -boot_id_file string
I1108 22:50:52.011] [25]     	Comma-separated list of files to check for boot-id. Use the first one that exists. (default "/proc/sys/kernel/random/boot_id")
I1108 22:50:52.011] [25]   -cert-dir string
I1108 22:50:52.012] [25]     	Path to the directory containing the certs. Default is empty, which doesn't use certs.
I1108 22:50:52.012] [25]   -clean-start
I1108 22:50:52.012] [25]     	If true, purge all namespaces except default and system before running tests. This serves to Cleanup test namespaces from failed/interrupted e2e runs in a long-lived cluster.
I1108 22:50:52.012] [25]   -cloud-config-file string
I1108 22:50:52.012] [25]     	Cloud config file.  Only required if provider is azure.
I1108 22:50:52.013] [25]   -cloud-provider-gce-lb-src-cidrs value
I1108 22:50:52.013] [25]     	CIDRs opened in GCE firewall for LB traffic proxy & health checks (default 130.211.0.0/22,209.85.152.0/22,209.85.204.0/22,35.191.0.0/16)
I1108 22:50:52.013] [25]   -cluster-ip-range string
I1108 22:50:52.013] [25]     	A CIDR notation IP range from which to assign IPs in the cluster. (default "10.64.0.0/14")
... skipping 91 lines ...
I1108 22:50:52.032] [25]     	If set, ginkgo will emit node output to files when running in parallel.
I1108 22:50:52.032] [25]   -ginkgo.dryRun
I1108 22:50:52.032] [25]     	If set, ginkgo will walk the test hierarchy without actually running anything.  Best paired with -v.
I1108 22:50:52.032] [25]   -ginkgo.failFast
I1108 22:50:52.032] [25]     	If set, ginkgo will stop running a test suite after a failure occurs.
I1108 22:50:52.033] [25]   -ginkgo.failOnPending
I1108 22:50:52.033] [25]     	If set, ginkgo will mark the test suite as failed if any specs are pending.
I1108 22:50:52.033] [25]   -ginkgo.flakeAttempts int
I1108 22:50:52.033] [25]     	Make up to this many attempts to run each spec. Please note that if any of the attempts succeed, the suite will not be failed. But any failures will still be recorded. (default 1)
I1108 22:50:52.034] [25]   -ginkgo.focus string
I1108 22:50:52.034] [25]     	If set, ginkgo will only run specs that match this regular expression.
I1108 22:50:52.034] [25]   -ginkgo.noColor
I1108 22:50:52.034] [25]     	If set, suppress color output in default reporter.
I1108 22:50:52.034] [25]   -ginkgo.noisyPendings
I1108 22:50:52.034] [25]     	If set, default reporter will shout about pending tests. (default true)
... skipping 71 lines ...
I1108 22:50:52.049] [25]     	If non-empty, use this log file
I1108 22:50:52.049] [25]   -log_file_max_size uint
I1108 22:50:52.049] [25]     	Defines the maximum size a log file can grow to. Unit is megabytes. If the value is 0, the maximum file size is unlimited. (default 1800)
I1108 22:50:52.049] [25]   -logexporter-gcs-path string
I1108 22:50:52.050] [25]     	Path to the GCS artifacts directory to dump logs from nodes. Logexporter gets enabled if this is non-empty.
I1108 22:50:52.050] [25]   -logtostderr
I1108 22:50:52.050] [25]     	log to standard error instead of files (default true)
I1108 22:50:52.050] [25]   -machine_id_file string
I1108 22:50:52.050] [25]     	Comma-separated list of files to check for machine-id. Use the first one that exists. (default "/etc/machine-id,/var/lib/dbus/machine-id")
I1108 22:50:52.051] [25]   -master-os-distro string
I1108 22:50:52.051] [25]     	The OS distribution of cluster master (debian, ubuntu, gci, coreos, or custom). (default "debian")
I1108 22:50:52.051] [25]   -master-tag string
I1108 22:50:52.051] [25]     	Network tags used on master instances. Valid only for gce, gke
... skipping 93 lines ...
I1108 22:50:52.070] [26]     	If true, adds the file directory to the header
I1108 22:50:52.070] [26]   -allow-gathering-profiles
I1108 22:50:52.070] [26]     	If set to true framework will allow to gather CPU/memory allocation pprof profiles from the master. (default true)
I1108 22:50:52.070] [26]   -allowed-not-ready-nodes int
I1108 22:50:52.071] [26]     	If non-zero, framework will allow for that many non-ready nodes when checking for all ready nodes.
I1108 22:50:52.071] [26]   -alsologtostderr
I1108 22:50:52.071] [26]     	log to standard error as well as files
I1108 22:50:52.071] [26]   -application_metrics_count_limit int
I1108 22:50:52.071] [26]     	Max number of application metrics to store (per container) (default 100)
I1108 22:50:52.072] [26]   -boot_id_file string
I1108 22:50:52.072] [26]     	Comma-separated list of files to check for boot-id. Use the first one that exists. (default "/proc/sys/kernel/random/boot_id")
I1108 22:50:52.072] [26]   -cert-dir string
I1108 22:50:52.072] [26]     	Path to the directory containing the certs. Default is empty, which doesn't use certs.
I1108 22:50:52.072] [26]   -clean-start
I1108 22:50:52.073] [26]     	If true, purge all namespaces except default and system before running tests. This serves to Cleanup test namespaces from failed/interrupted e2e runs in a long-lived cluster.
I1108 22:50:52.073] [26]   -cloud-config-file string
I1108 22:50:52.073] [26]     	Cloud config file.  Only required if provider is azure.
I1108 22:50:52.073] [26]   -cloud-provider-gce-lb-src-cidrs value
I1108 22:50:52.074] [26]     	CIDRs opened in GCE firewall for LB traffic proxy & health checks (default 130.211.0.0/22,209.85.152.0/22,209.85.204.0/22,35.191.0.0/16)
I1108 22:50:52.074] [26]   -cluster-ip-range string
I1108 22:50:52.074] [26]     	A CIDR notation IP range from which to assign IPs in the cluster. (default "10.64.0.0/14")
... skipping 91 lines ...
I1108 22:50:52.092] [26]     	If set, ginkgo will emit node output to files when running in parallel.
I1108 22:50:52.093] [26]   -ginkgo.dryRun
I1108 22:50:52.093] [26]     	If set, ginkgo will walk the test hierarchy without actually running anything.  Best paired with -v.
I1108 22:50:52.093] [26]   -ginkgo.failFast
I1108 22:50:52.093] [26]     	If set, ginkgo will stop running a test suite after a failure occurs.
I1108 22:50:52.093] [26]   -ginkgo.failOnPending
I1108 22:50:52.094] [26]     	If set, ginkgo will mark the test suite as failed if any specs are pending.
I1108 22:50:52.094] [26]   -ginkgo.flakeAttempts int
I1108 22:50:52.094] [26]     	Make up to this many attempts to run each spec. Please note that if any of the attempts succeed, the suite will not be failed. But any failures will still be recorded. (default 1)
I1108 22:50:52.094] [26]   -ginkgo.focus string
I1108 22:50:52.094] [26]     	If set, ginkgo will only run specs that match this regular expression.
I1108 22:50:52.095] [26]   -ginkgo.noColor
I1108 22:50:52.095] [26]     	If set, suppress color output in default reporter.
I1108 22:50:52.095] [26]   -ginkgo.noisyPendings
I1108 22:50:52.095] [26]     	If set, default reporter will shout about pending tests. (default true)
... skipping 71 lines ...
I1108 22:50:52.109] [26]     	If non-empty, use this log file
I1108 22:50:52.109] [26]   -log_file_max_size uint
I1108 22:50:52.109] [26]     	Defines the maximum size a log file can grow to. Unit is megabytes. If the value is 0, the maximum file size is unlimited. (default 1800)
I1108 22:50:52.109] [26]   -logexporter-gcs-path string
I1108 22:50:52.110] [26]     	Path to the GCS artifacts directory to dump logs from nodes. Logexporter gets enabled if this is non-empty.
I1108 22:50:52.110] [26]   -logtostderr
I1108 22:50:52.110] [26]     	log to standard error instead of files (default true)
I1108 22:50:52.110] [26]   -machine_id_file string
I1108 22:50:52.110] [26]     	Comma-separated list of files to check for machine-id. Use the first one that exists. (default "/etc/machine-id,/var/lib/dbus/machine-id")
I1108 22:50:52.111] [26]   -master-os-distro string
I1108 22:50:52.111] [26]     	The OS distribution of cluster master (debian, ubuntu, gci, coreos, or custom). (default "debian")
I1108 22:50:52.111] [26]   -master-tag string
I1108 22:50:52.111] [26]     	Network tags used on master instances. Valid only for gce, gke
... skipping 93 lines ...
I1108 22:50:52.129] [27]     	If true, adds the file directory to the header
I1108 22:50:52.129] [27]   -allow-gathering-profiles
I1108 22:50:52.129] [27]     	If set to true framework will allow to gather CPU/memory allocation pprof profiles from the master. (default true)
I1108 22:50:52.129] [27]   -allowed-not-ready-nodes int
I1108 22:50:52.130] [27]     	If non-zero, framework will allow for that many non-ready nodes when checking for all ready nodes.
I1108 22:50:52.130] [27]   -alsologtostderr
I1108 22:50:52.130] [27]     	log to standard error as well as files
I1108 22:50:52.130] [27]   -application_metrics_count_limit int
I1108 22:50:52.130] [27]     	Max number of application metrics to store (per container) (default 100)
I1108 22:50:52.131] [27]   -boot_id_file string
I1108 22:50:52.131] [27]     	Comma-separated list of files to check for boot-id. Use the first one that exists. (default "/proc/sys/kernel/random/boot_id")
I1108 22:50:52.131] [27]   -cert-dir string
I1108 22:50:52.131] [27]     	Path to the directory containing the certs. Default is empty, which doesn't use certs.
I1108 22:50:52.131] [27]   -clean-start
I1108 22:50:52.132] [27]     	If true, purge all namespaces except default and system before running tests. This serves to Cleanup test namespaces from failed/interrupted e2e runs in a long-lived cluster.
I1108 22:50:52.132] [27]   -cloud-config-file string
I1108 22:50:52.132] [27]     	Cloud config file.  Only required if provider is azure.
I1108 22:50:52.132] [27]   -cloud-provider-gce-lb-src-cidrs value
I1108 22:50:52.132] [27]     	CIDRs opened in GCE firewall for LB traffic proxy & health checks (default 130.211.0.0/22,209.85.152.0/22,209.85.204.0/22,35.191.0.0/16)
I1108 22:50:52.133] [27]   -cluster-ip-range string
I1108 22:50:52.133] [27]     	A CIDR notation IP range from which to assign IPs in the cluster. (default "10.64.0.0/14")
... skipping 91 lines ...
I1108 22:50:52.152] [27]     	If set, ginkgo will emit node output to files when running in parallel.
I1108 22:50:52.152] [27]   -ginkgo.dryRun
I1108 22:50:52.152] [27]     	If set, ginkgo will walk the test hierarchy without actually running anything.  Best paired with -v.
I1108 22:50:52.152] [27]   -ginkgo.failFast
I1108 22:50:52.152] [27]     	If set, ginkgo will stop running a test suite after a failure occurs.
I1108 22:50:52.152] [27]   -ginkgo.failOnPending
I1108 22:50:52.153] [27]     	If set, ginkgo will mark the test suite as failed if any specs are pending.
I1108 22:50:52.153] [27]   -ginkgo.flakeAttempts int
I1108 22:50:52.153] [27]     	Make up to this many attempts to run each spec. Please note that if any of the attempts succeed, the suite will not be failed. But any failures will still be recorded. (default 1)
I1108 22:50:52.153] [27]   -ginkgo.focus string
I1108 22:50:52.154] [27]     	If set, ginkgo will only run specs that match this regular expression.
I1108 22:50:52.154] [27]   -ginkgo.noColor
I1108 22:50:52.154] [27]     	If set, suppress color output in default reporter.
I1108 22:50:52.154] [27]   -ginkgo.noisyPendings
I1108 22:50:52.154] [27]     	If set, default reporter will shout about pending tests. (default true)
... skipping 71 lines ...
I1108 22:50:52.168] [27]     	If non-empty, use this log file
I1108 22:50:52.168] [27]   -log_file_max_size uint
I1108 22:50:52.169] [27]     	Defines the maximum size a log file can grow to. Unit is megabytes. If the value is 0, the maximum file size is unlimited. (default 1800)
I1108 22:50:52.169] [27]   -logexporter-gcs-path string
I1108 22:50:52.169] [27]     	Path to the GCS artifacts directory to dump logs from nodes. Logexporter gets enabled if this is non-empty.
I1108 22:50:52.169] [27]   -logtostderr
I1108 22:50:52.169] [27]     	log to standard error instead of files (default true)
I1108 22:50:52.170] [27]   -machine_id_file string
I1108 22:50:52.170] [27]     	Comma-separated list of files to check for machine-id. Use the first one that exists. (default "/etc/machine-id,/var/lib/dbus/machine-id")
I1108 22:50:52.170] [27]   -master-os-distro string
I1108 22:50:52.170] [27]     	The OS distribution of cluster master (debian, ubuntu, gci, coreos, or custom). (default "debian")
I1108 22:50:52.170] [27]   -master-tag string
I1108 22:50:52.171] [27]     	Network tags used on master instances. Valid only for gce, gke
... skipping 93 lines ...
I1108 22:50:52.190] [28]     	If true, adds the file directory to the header
I1108 22:50:52.190] [28]   -allow-gathering-profiles
I1108 22:50:52.191] [28]     	If set to true framework will allow to gather CPU/memory allocation pprof profiles from the master. (default true)
I1108 22:50:52.191] [28]   -allowed-not-ready-nodes int
I1108 22:50:52.191] [28]     	If non-zero, framework will allow for that many non-ready nodes when checking for all ready nodes.
I1108 22:50:52.191] [28]   -alsologtostderr
I1108 22:50:52.191] [28]     	log to standard error as well as files
I1108 22:50:52.192] [28]   -application_metrics_count_limit int
I1108 22:50:52.192] [28]     	Max number of application metrics to store (per container) (default 100)
I1108 22:50:52.192] [28]   -boot_id_file string
I1108 22:50:52.192] [28]     	Comma-separated list of files to check for boot-id. Use the first one that exists. (default "/proc/sys/kernel/random/boot_id")
I1108 22:50:52.192] [28]   -cert-dir string
I1108 22:50:52.193] [28]     	Path to the directory containing the certs. Default is empty, which doesn't use certs.
I1108 22:50:52.194] [28]   -clean-start
I1108 22:50:52.194] [28]     	If true, purge all namespaces except default and system before running tests. This serves to Cleanup test namespaces from failed/interrupted e2e runs in a long-lived cluster.
I1108 22:50:52.194] [28]   -cloud-config-file string
I1108 22:50:52.194] [28]     	Cloud config file.  Only required if provider is azure.
I1108 22:50:52.194] [28]   -cloud-provider-gce-lb-src-cidrs value
I1108 22:50:52.195] [28]     	CIDRs opened in GCE firewall for LB traffic proxy & health checks (default 130.211.0.0/22,209.85.152.0/22,209.85.204.0/22,35.191.0.0/16)
I1108 22:50:52.195] [28]   -cluster-ip-range string
I1108 22:50:52.195] [28]     	A CIDR notation IP range from which to assign IPs in the cluster. (default "10.64.0.0/14")
... skipping 91 lines ...
I1108 22:50:52.216] [28]     	If set, ginkgo will emit node output to files when running in parallel.
I1108 22:50:52.216] [28]   -ginkgo.dryRun
I1108 22:50:52.216] [28]     	If set, ginkgo will walk the test hierarchy without actually running anything.  Best paired with -v.
I1108 22:50:52.216] [28]   -ginkgo.failFast
I1108 22:50:52.217] [28]     	If set, ginkgo will stop running a test suite after a failure occurs.
I1108 22:50:52.217] [28]   -ginkgo.failOnPending
I1108 22:50:52.217] [28]     	If set, ginkgo will mark the test suite as failed if any specs are pending.
I1108 22:50:52.217] [28]   -ginkgo.flakeAttempts int
I1108 22:50:52.218] [28]     	Make up to this many attempts to run each spec. Please note that if any of the attempts succeed, the suite will not be failed. But any failures will still be recorded. (default 1)
I1108 22:50:52.218] [28]   -ginkgo.focus string
I1108 22:50:52.218] [28]     	If set, ginkgo will only run specs that match this regular expression.
I1108 22:50:52.218] [28]   -ginkgo.noColor
I1108 22:50:52.218] [28]     	If set, suppress color output in default reporter.
I1108 22:50:52.219] [28]   -ginkgo.noisyPendings
I1108 22:50:52.219] [28]     	If set, default reporter will shout about pending tests. (default true)
... skipping 71 lines ...
I1108 22:50:52.237] [28]     	If non-empty, use this log file
I1108 22:50:52.238] [28]   -log_file_max_size uint
I1108 22:50:52.238] [28]     	Defines the maximum size a log file can grow to. Unit is megabytes. If the value is 0, the maximum file size is unlimited. (default 1800)
I1108 22:50:52.238] [28]   -logexporter-gcs-path string
I1108 22:50:52.239] [28]     	Path to the GCS artifacts directory to dump logs from nodes. Logexporter gets enabled if this is non-empty.
I1108 22:50:52.239] [28]   -logtostderr
I1108 22:50:52.239] [28]     	log to standard error instead of files (default true)
I1108 22:50:52.239] [28]   -machine_id_file string
I1108 22:50:52.240] [28]     	Comma-separated list of files to check for machine-id. Use the first one that exists. (default "/etc/machine-id,/var/lib/dbus/machine-id")
I1108 22:50:52.240] [28]   -master-os-distro string
I1108 22:50:52.240] [28]     	The OS distribution of cluster master (debian, ubuntu, gci, coreos, or custom). (default "debian")
I1108 22:50:52.240] [28]   -master-tag string
I1108 22:50:52.240] [28]     	Network tags used on master instances. Valid only for gce, gke
... skipping 93 lines ...
I1108 22:50:52.260] [29]     	If true, adds the file directory to the header
I1108 22:50:52.260] [29]   -allow-gathering-profiles
I1108 22:50:52.260] [29]     	If set to true framework will allow to gather CPU/memory allocation pprof profiles from the master. (default true)
I1108 22:50:52.261] [29]   -allowed-not-ready-nodes int
I1108 22:50:52.261] [29]     	If non-zero, framework will allow for that many non-ready nodes when checking for all ready nodes.
I1108 22:50:52.261] [29]   -alsologtostderr
I1108 22:50:52.261] [29]     	log to standard error as well as files
I1108 22:50:52.261] [29]   -application_metrics_count_limit int
I1108 22:50:52.262] [29]     	Max number of application metrics to store (per container) (default 100)
I1108 22:50:52.262] [29]   -boot_id_file string
I1108 22:50:52.262] [29]     	Comma-separated list of files to check for boot-id. Use the first one that exists. (default "/proc/sys/kernel/random/boot_id")
I1108 22:50:52.262] [29]   -cert-dir string
I1108 22:50:52.262] [29]     	Path to the directory containing the certs. Default is empty, which doesn't use certs.
I1108 22:50:52.262] [29]   -clean-start
I1108 22:50:52.263] [29]     	If true, purge all namespaces except default and system before running tests. This serves to Cleanup test namespaces from failed/interrupted e2e runs in a long-lived cluster.
I1108 22:50:52.263] [29]   -cloud-config-file string
I1108 22:50:52.263] [29]     	Cloud config file.  Only required if provider is azure.
I1108 22:50:52.263] [29]   -cloud-provider-gce-lb-src-cidrs value
I1108 22:50:52.264] [29]     	CIDRs opened in GCE firewall for LB traffic proxy & health checks (default 130.211.0.0/22,209.85.152.0/22,209.85.204.0/22,35.191.0.0/16)
I1108 22:50:52.264] [29]   -cluster-ip-range string
I1108 22:50:52.264] [29]     	A CIDR notation IP range from which to assign IPs in the cluster. (default "10.64.0.0/14")
... skipping 91 lines ...
I1108 22:50:52.281] [29]     	If set, ginkgo will emit node output to files when running in parallel.
I1108 22:50:52.282] [29]   -ginkgo.dryRun
I1108 22:50:52.282] [29]     	If set, ginkgo will walk the test hierarchy without actually running anything.  Best paired with -v.
I1108 22:50:52.282] [29]   -ginkgo.failFast
I1108 22:50:52.282] [29]     	If set, ginkgo will stop running a test suite after a failure occurs.
I1108 22:50:52.282] [29]   -ginkgo.failOnPending
I1108 22:50:52.283] [29]     	If set, ginkgo will mark the test suite as failed if any specs are pending.
I1108 22:50:52.283] [29]   -ginkgo.flakeAttempts int
I1108 22:50:52.283] [29]     	Make up to this many attempts to run each spec. Please note that if any of the attempts succeed, the suite will not be failed. But any failures will still be recorded. (default 1)
I1108 22:50:52.283] [29]   -ginkgo.focus string
I1108 22:50:52.283] [29]     	If set, ginkgo will only run specs that match this regular expression.
I1108 22:50:52.283] [29]   -ginkgo.noColor
I1108 22:50:52.284] [29]     	If set, suppress color output in default reporter.
I1108 22:50:52.284] [29]   -ginkgo.noisyPendings
I1108 22:50:52.284] [29]     	If set, default reporter will shout about pending tests. (default true)
... skipping 71 lines ...
I1108 22:50:52.301] [29]     	If non-empty, use this log file
I1108 22:50:52.301] [29]   -log_file_max_size uint
I1108 22:50:52.301] [29]     	Defines the maximum size a log file can grow to. Unit is megabytes. If the value is 0, the maximum file size is unlimited. (default 1800)
I1108 22:50:52.302] [29]   -logexporter-gcs-path string
I1108 22:50:52.302] [29]     	Path to the GCS artifacts directory to dump logs from nodes. Logexporter gets enabled if this is non-empty.
I1108 22:50:52.302] [29]   -logtostderr
I1108 22:50:52.302] [29]     	log to standard error instead of files (default true)
I1108 22:50:52.302] [29]   -machine_id_file string
I1108 22:50:52.303] [29]     	Comma-separated list of files to check for machine-id. Use the first one that exists. (default "/etc/machine-id,/var/lib/dbus/machine-id")
I1108 22:50:52.303] [29]   -master-os-distro string
I1108 22:50:52.303] [29]     	The OS distribution of cluster master (debian, ubuntu, gci, coreos, or custom). (default "debian")
I1108 22:50:52.303] [29]   -master-tag string
I1108 22:50:52.304] [29]     	Network tags used on master instances. Valid only for gce, gke
... skipping 93 lines ...
I1108 22:50:52.326] [30]     	If true, adds the file directory to the header
I1108 22:50:52.326] [30]   -allow-gathering-profiles
I1108 22:50:52.326] [30]     	If set to true framework will allow to gather CPU/memory allocation pprof profiles from the master. (default true)
I1108 22:50:52.326] [30]   -allowed-not-ready-nodes int
I1108 22:50:52.327] [30]     	If non-zero, framework will allow for that many non-ready nodes when checking for all ready nodes.
I1108 22:50:52.327] [30]   -alsologtostderr
I1108 22:50:52.327] [30]     	log to standard error as well as files
I1108 22:50:52.327] [30]   -application_metrics_count_limit int
I1108 22:50:52.327] [30]     	Max number of application metrics to store (per container) (default 100)
I1108 22:50:52.328] [30]   -boot_id_file string
I1108 22:50:52.328] [30]     	Comma-separated list of files to check for boot-id. Use the first one that exists. (default "/proc/sys/kernel/random/boot_id")
I1108 22:50:52.328] [30]   -cert-dir string
I1108 22:50:52.328] [30]     	Path to the directory containing the certs. Default is empty, which doesn't use certs.
I1108 22:50:52.329] [30]   -clean-start
I1108 22:50:52.329] [30]     	If true, purge all namespaces except default and system before running tests. This serves to Cleanup test namespaces from failed/interrupted e2e runs in a long-lived cluster.
I1108 22:50:52.329] [30]   -cloud-config-file string
I1108 22:50:52.330] [30]     	Cloud config file.  Only required if provider is azure.
I1108 22:50:52.330] [30]   -cloud-provider-gce-lb-src-cidrs value
I1108 22:50:52.330] [30]     	CIDRs opened in GCE firewall for LB traffic proxy & health checks (default 130.211.0.0/22,209.85.152.0/22,209.85.204.0/22,35.191.0.0/16)
I1108 22:50:52.330] [30]   -cluster-ip-range string
I1108 22:50:52.331] [30]     	A CIDR notation IP range from which to assign IPs in the cluster. (default "10.64.0.0/14")
... skipping 91 lines ...
I1108 22:50:52.349] [30]     	If set, ginkgo will emit node output to files when running in parallel.
I1108 22:50:52.349] [30]   -ginkgo.dryRun
I1108 22:50:52.350] [30]     	If set, ginkgo will walk the test hierarchy without actually running anything.  Best paired with -v.
I1108 22:50:52.350] [30]   -ginkgo.failFast
I1108 22:50:52.350] [30]     	If set, ginkgo will stop running a test suite after a failure occurs.
I1108 22:50:52.350] [30]   -ginkgo.failOnPending
I1108 22:50:52.350] [30]     	If set, ginkgo will mark the test suite as failed if any specs are pending.
I1108 22:50:52.351] [30]   -ginkgo.flakeAttempts int
I1108 22:50:52.351] [30]     	Make up to this many attempts to run each spec. Please note that if any of the attempts succeed, the suite will not be failed. But any failures will still be recorded. (default 1)
I1108 22:50:52.351] [30]   -ginkgo.focus string
I1108 22:50:52.351] [30]     	If set, ginkgo will only run specs that match this regular expression.
I1108 22:50:52.351] [30]   -ginkgo.noColor
I1108 22:50:52.352] [30]     	If set, suppress color output in default reporter.
I1108 22:50:52.352] [30]   -ginkgo.noisyPendings
I1108 22:50:52.352] [30]     	If set, default reporter will shout about pending tests. (default true)
... skipping 71 lines ...
I1108 22:50:52.366] [30]     	If non-empty, use this log file
I1108 22:50:52.366] [30]   -log_file_max_size uint
I1108 22:50:52.367] [30]     	Defines the maximum size a log file can grow to. Unit is megabytes. If the value is 0, the maximum file size is unlimited. (default 1800)
I1108 22:50:52.367] [30]   -logexporter-gcs-path string
I1108 22:50:52.367] [30]     	Path to the GCS artifacts directory to dump logs from nodes. Logexporter gets enabled if this is non-empty.
I1108 22:50:52.367] [30]   -logtostderr
I1108 22:50:52.367] [30]     	log to standard error instead of files (default true)
I1108 22:50:52.368] [30]   -machine_id_file string
I1108 22:50:52.368] [30]     	Comma-separated list of files to check for machine-id. Use the first one that exists. (default "/etc/machine-id,/var/lib/dbus/machine-id")
I1108 22:50:52.368] [30]   -master-os-distro string
I1108 22:50:52.368] [30]     	The OS distribution of cluster master (debian, ubuntu, gci, coreos, or custom). (default "debian")
I1108 22:50:52.368] [30]   -master-tag string
I1108 22:50:52.369] [30]     	Network tags used on master instances. Valid only for gce, gke
... skipping 86 lines ...
I1108 22:50:52.386] [30]   -vmodule value
I1108 22:50:52.387] [30]     	comma-separated list of pattern=N settings for file-filtered logging
I1108 22:50:52.387] [30]   -volume-dir string
I1108 22:50:52.387] [30]     	Path to the directory containing the kubelet volumes. (default "/var/lib/kubelet")
I1108 22:50:52.387] 
I1108 22:50:52.387] Ginkgo ran 1 suite in 12.841764921s
I1108 22:50:52.388] Test Suite Failed
I1108 22:50:52.388] Checking for custom logdump instances, if any
I1108 22:50:52.388] Sourcing kube-util.sh
I1108 22:50:52.388] Detecting project
I1108 22:50:52.388] Project: k8s-jkns-pr-gce-gpus
I1108 22:50:52.388] Network Project: k8s-jkns-pr-gce-gpus
I1108 22:50:52.389] Zone: us-west1-b
I1108 22:50:52.389] Dumping logs from master locally to '/workspace/_artifacts'
W1108 22:50:52.489] !!! Error in ./hack/ginkgo-e2e.sh:146
W1108 22:50:52.491]   Error in ./hack/ginkgo-e2e.sh:146. '"${ginkgo}" "${ginkgo_args[@]:+${ginkgo_args[@]}}" "${e2e_test}" -- "${auth_config[@]:+${auth_config[@]}}" --ginkgo.flakeAttempts="${FLAKE_ATTEMPTS}" --host="${KUBE_MASTER_URL}" --provider="${KUBERNETES_PROVIDER}" --gce-project="${PROJECT:-}" --gce-zone="${ZONE:-}" --gce-region="${REGION:-}" --gce-multizone="${MULTIZONE:-false}" --gke-cluster="${CLUSTER_NAME:-}" --kube-master="${KUBE_MASTER:-}" --cluster-tag="${CLUSTER_ID:-}" --cloud-config-file="${CLOUD_CONFIG:-}" --repo-root="${KUBE_ROOT}" --node-instance-group="${NODE_INSTANCE_GROUP:-}" --prefix="${KUBE_GCE_INSTANCE_PREFIX:-e2e}" --network="${KUBE_GCE_NETWORK:-${KUBE_GKE_NETWORK:-e2e}}" --node-tag="${NODE_TAG:-}" --master-tag="${MASTER_TAG:-}" --cluster-monitoring-mode="${KUBE_ENABLE_CLUSTER_MONITORING:-standalone}" --prometheus-monitoring="${KUBE_ENABLE_PROMETHEUS_MONITORING:-false}" --dns-domain="${KUBE_DNS_DOMAIN:-cluster.local}" --ginkgo.slowSpecThreshold="${GINKGO_SLOW_SPEC_THRESHOLD:-300}" ${KUBE_CONTAINER_RUNTIME:+"--container-runtime=${KUBE_CONTAINER_RUNTIME}"} ${MASTER_OS_DISTRIBUTION:+"--master-os-distro=${MASTER_OS_DISTRIBUTION}"} ${NODE_OS_DISTRIBUTION:+"--node-os-distro=${NODE_OS_DISTRIBUTION}"} ${NUM_NODES:+"--num-nodes=${NUM_NODES}"} ${E2E_REPORT_DIR:+"--report-dir=${E2E_REPORT_DIR}"} ${E2E_REPORT_PREFIX:+"--report-prefix=${E2E_REPORT_PREFIX}"} "${@:-}"' exited with status 1
W1108 22:50:52.491] Call stack:
W1108 22:50:52.491]   1: ./hack/ginkgo-e2e.sh:146 main(...)
W1108 22:50:52.491] Exiting with status 1
W1108 22:50:52.492] 2019/11/08 22:50:52 process.go:155: Step './hack/ginkgo-e2e.sh --ginkgo.focus=\[Feature:GPUDevicePlugin\] --minStartupPods=8 --report-dir=/workspace/_artifacts --disable-log-dump=true --cluster-ip-range=10.64.0.0/14' finished in 13.860345404s
W1108 22:50:52.492] 2019/11/08 22:50:52 e2e.go:534: Dumping logs locally to: /workspace/_artifacts
W1108 22:50:52.492] 2019/11/08 22:50:52 process.go:153: Running: ./cluster/log-dump/log-dump.sh /workspace/_artifacts
... skipping 5 lines ...
W1108 22:51:27.578] 
W1108 22:51:27.579] Specify --start=41229 in the next get-serial-port-output invocation to get only the new output starting from here.
W1108 22:51:30.180] scp: /var/log/cluster-autoscaler.log*: No such file or directory
W1108 22:51:30.258] scp: /var/log/fluentd.log*: No such file or directory
W1108 22:51:30.259] scp: /var/log/kubelet.cov*: No such file or directory
W1108 22:51:30.259] scp: /var/log/startupscript.log*: No such file or directory
W1108 22:51:30.263] ERROR: (gcloud.compute.scp) [/usr/bin/scp] exited with return code [1].
I1108 22:51:30.364] Dumping logs from nodes locally to '/workspace/_artifacts'
I1108 22:51:30.364] Detecting nodes in the cluster
I1108 22:52:08.617] Changing logfiles to be world-readable for download
I1108 22:52:09.003] Changing logfiles to be world-readable for download
I1108 22:52:09.322] Changing logfiles to be world-readable for download
I1108 22:52:09.387] Changing logfiles to be world-readable for download
... skipping 10 lines ...
W1108 22:52:13.818] 
W1108 22:52:13.819] Specify --start=44717 in the next get-serial-port-output invocation to get only the new output starting from here.
W1108 22:52:14.743] scp: /var/log/fluentd.log*: No such file or directory
W1108 22:52:14.743] scp: /var/log/node-problem-detector.log*: No such file or directory
W1108 22:52:14.743] scp: /var/log/kubelet.cov*: No such file or directory
W1108 22:52:14.743] scp: /var/log/startupscript.log*: No such file or directory
W1108 22:52:14.753] ERROR: (gcloud.compute.scp) [/usr/bin/scp] exited with return code [1].
W1108 22:52:15.252] scp: /var/log/fluentd.log*: No such file or directory
W1108 22:52:15.252] scp: /var/log/node-problem-detector.log*: No such file or directory
W1108 22:52:15.252] scp: /var/log/kubelet.cov*: No such file or directory
W1108 22:52:15.252] scp: /var/log/startupscript.log*: No such file or directory
W1108 22:52:15.255] ERROR: (gcloud.compute.scp) [/usr/bin/scp] exited with return code [1].
W1108 22:52:15.511] scp: /var/log/fluentd.log*: No such file or directory
W1108 22:52:15.511] scp: /var/log/node-problem-detector.log*: No such file or directory
W1108 22:52:15.512] scp: /var/log/kubelet.cov*: No such file or directory
W1108 22:52:15.512] scp: /var/log/startupscript.log*: No such file or directory
W1108 22:52:15.515] ERROR: (gcloud.compute.scp) [/usr/bin/scp] exited with return code [1].
W1108 22:52:16.022] scp: /var/log/fluentd.log*: No such file or directory
W1108 22:52:16.023] scp: /var/log/node-problem-detector.log*: No such file or directory
W1108 22:52:16.023] scp: /var/log/kubelet.cov*: No such file or directory
W1108 22:52:16.023] scp: /var/log/startupscript.log*: No such file or directory
W1108 22:52:16.026] ERROR: (gcloud.compute.scp) [/usr/bin/scp] exited with return code [1].
W1108 22:52:19.632] INSTANCE_GROUPS=e2e-e04178beac-8baae-minion-group
W1108 22:52:19.632] NODE_NAMES=e2e-e04178beac-8baae-minion-group-62l5 e2e-e04178beac-8baae-minion-group-df8l e2e-e04178beac-8baae-minion-group-jbvx e2e-e04178beac-8baae-minion-group-tz9p
I1108 22:52:20.596] Failures for e2e-e04178beac-8baae-minion-group
W1108 22:52:21.648] 2019/11/08 22:52:21 process.go:155: Step './cluster/log-dump/log-dump.sh /workspace/_artifacts' finished in 1m29.458499519s
W1108 22:52:21.648] 2019/11/08 22:52:21 process.go:153: Running: ./hack/e2e-internal/e2e-down.sh
W1108 22:52:21.696] Project: k8s-jkns-pr-gce-gpus
... skipping 15 lines ...
W1108 22:52:37.991] Deleting Managed Instance Group...
W1108 22:55:10.518] .................................Deleted [https://www.googleapis.com/compute/v1/projects/k8s-jkns-pr-gce-gpus/zones/us-west1-b/instanceGroupManagers/e2e-e04178beac-8baae-minion-group].
W1108 22:55:10.519] done.
W1108 22:55:16.734] Deleted [https://www.googleapis.com/compute/v1/projects/k8s-jkns-pr-gce-gpus/global/instanceTemplates/e2e-e04178beac-8baae-minion-template].
W1108 22:55:23.606] Deleted [https://www.googleapis.com/compute/v1/projects/k8s-jkns-pr-gce-gpus/global/instanceTemplates/e2e-e04178beac-8baae-windows-node-template].
I1108 22:55:28.478] Removing etcd replica, name: e2e-e04178beac-8baae-master, port: 2379, result: 52
I1108 22:55:30.149] {"message":"Internal Server Error"}Removing etcd replica, name: e2e-e04178beac-8baae-master, port: 4002, result: 0
W1108 22:55:34.716] Updated [https://www.googleapis.com/compute/v1/projects/k8s-jkns-pr-gce-gpus/zones/us-west1-b/instances/e2e-e04178beac-8baae-master].
W1108 22:57:57.001] Deleted [https://www.googleapis.com/compute/v1/projects/k8s-jkns-pr-gce-gpus/zones/us-west1-b/instances/e2e-e04178beac-8baae-master].
W1108 22:58:13.048] Deleted [https://www.googleapis.com/compute/v1/projects/k8s-jkns-pr-gce-gpus/global/firewalls/e2e-e04178beac-8baae-master-https].
W1108 22:58:13.383] Deleted [https://www.googleapis.com/compute/v1/projects/k8s-jkns-pr-gce-gpus/global/firewalls/e2e-e04178beac-8baae-master-etcd].
W1108 22:58:14.199] Deleted [https://www.googleapis.com/compute/v1/projects/k8s-jkns-pr-gce-gpus/global/firewalls/e2e-e04178beac-8baae-minion-all].
W1108 22:58:19.958] Deleted [https://www.googleapis.com/compute/v1/projects/k8s-jkns-pr-gce-gpus/regions/us-west1/addresses/e2e-e04178beac-8baae-master-ip].
... skipping 12 lines ...
I1108 22:59:26.004] Property "users.k8s-jkns-pr-gce-gpus_e2e-e04178beac-8baae-basic-auth" unset.
I1108 22:59:26.144] Property "contexts.k8s-jkns-pr-gce-gpus_e2e-e04178beac-8baae" unset.
I1108 22:59:26.148] Cleared config for k8s-jkns-pr-gce-gpus_e2e-e04178beac-8baae from /workspace/.kube/config
I1108 22:59:26.148] Done
W1108 22:59:26.161] 2019/11/08 22:59:26 process.go:155: Step './hack/e2e-internal/e2e-down.sh' finished in 7m4.501149087s
W1108 22:59:26.162] 2019/11/08 22:59:26 process.go:96: Saved XML output to /workspace/_artifacts/junit_runner.xml.
W1108 22:59:26.162] 2019/11/08 22:59:26 main.go:319: Something went wrong: encountered 1 errors: [error during ./hack/ginkgo-e2e.sh --ginkgo.focus=\[Feature:GPUDevicePlugin\] --minStartupPods=8 --report-dir=/workspace/_artifacts --disable-log-dump=true --cluster-ip-range=10.64.0.0/14: exit status 1]
W1108 22:59:26.163] Traceback (most recent call last):
W1108 22:59:26.163]   File "/workspace/./test-infra/jenkins/../scenarios/kubernetes_e2e.py", line 778, in <module>
W1108 22:59:26.163]     main(parse_args())
W1108 22:59:26.163]   File "/workspace/./test-infra/jenkins/../scenarios/kubernetes_e2e.py", line 626, in main
W1108 22:59:26.163]     mode.start(runner_args)
W1108 22:59:26.163]   File "/workspace/./test-infra/jenkins/../scenarios/kubernetes_e2e.py", line 262, in start
W1108 22:59:26.164]     check_env(env, self.command, *args)
W1108 22:59:26.164]   File "/workspace/./test-infra/jenkins/../scenarios/kubernetes_e2e.py", line 111, in check_env
W1108 22:59:26.164]     subprocess.check_call(cmd, env=env)
W1108 22:59:26.164]   File "/usr/lib/python2.7/subprocess.py", line 190, in check_call
W1108 22:59:26.164]     raise CalledProcessError(retcode, cmd)
W1108 22:59:26.164] subprocess.CalledProcessError: Command '('kubetest', '--dump=/workspace/_artifacts', '--gcp-service-account=/etc/service-account/service-account.json', '--build=bazel', '--stage=gs://kubernetes-release-pull/ci/pull-kubernetes-e2e-gce-device-plugin-gpu', '--up', '--down', '--test', '--provider=gce', '--cluster=e2e-e04178beac-8baae', '--gcp-network=e2e-e04178beac-8baae', '--extract=local', '--gcp-node-image=gci', '--gcp-nodes=4', '--gcp-project=k8s-jkns-pr-gce-gpus', '--gcp-zone=us-west1-b', '--ginkgo-parallel=30', '--test_args=--ginkgo.focus=\\[Feature:GPUDevicePlugin\\] --minStartupPods=8', '--timeout=60m')' returned non-zero exit status 1
E1108 22:59:26.165] Command failed
I1108 22:59:26.165] process 577 exited with code 1 after 22.3m
E1108 22:59:26.165] FAIL: pull-kubernetes-e2e-gce-device-plugin-gpu
I1108 22:59:26.165] Call:  gcloud auth activate-service-account --key-file=/etc/service-account/service-account.json
W1108 22:59:26.622] Activated service account credentials for: [pr-kubekins@kubernetes-jenkins-pull.iam.gserviceaccount.com]
I1108 22:59:26.670] process 43246 exited with code 0 after 0.0m
I1108 22:59:26.671] Call:  gcloud config get-value account
I1108 22:59:26.945] process 43258 exited with code 0 after 0.0m
I1108 22:59:26.946] Will upload results to gs://kubernetes-jenkins/pr-logs using pr-kubekins@kubernetes-jenkins-pull.iam.gserviceaccount.com
I1108 22:59:26.946] Upload result and artifacts...
I1108 22:59:26.946] Gubernator results at https://gubernator.k8s.io/build/kubernetes-jenkins/pr-logs/pull/85019/pull-kubernetes-e2e-gce-device-plugin-gpu/1192933722979045376
I1108 22:59:26.947] Call:  gsutil ls gs://kubernetes-jenkins/pr-logs/pull/85019/pull-kubernetes-e2e-gce-device-plugin-gpu/1192933722979045376/artifacts
W1108 22:59:27.799] CommandException: One or more URLs matched no objects.
E1108 22:59:27.902] Command failed
I1108 22:59:27.902] process 43270 exited with code 1 after 0.0m
W1108 22:59:27.902] Remote dir gs://kubernetes-jenkins/pr-logs/pull/85019/pull-kubernetes-e2e-gce-device-plugin-gpu/1192933722979045376/artifacts not exist yet
I1108 22:59:27.903] Call:  gsutil -m -q -o GSUtil:use_magicfile=True cp -r -c -z log,txt,xml /workspace/_artifacts gs://kubernetes-jenkins/pr-logs/pull/85019/pull-kubernetes-e2e-gce-device-plugin-gpu/1192933722979045376/artifacts
I1108 22:59:30.172] process 43414 exited with code 0 after 0.0m
I1108 22:59:30.173] Call:  git rev-parse HEAD
I1108 22:59:30.178] process 44073 exited with code 0 after 0.0m
... skipping 20 lines ...