Dear Fabien and Fabio,
Puppet configuration is now fixed and Kubernetes 1.9.1 is running on all nodes.
However I now use internal kubernetes DNS and it seems something it killing it in my back after a few seconds on CC-IN2P3 cluster (see logs below).
FYI, I have successfully tested this setup on 10 Centos7 boxes in Openstack.
Would you please have an idea about the thing which kill kube-dns process on CC-IN2P3 cluster?
Cheers,
Fabrice
# Logs for the process inside kubedns pod:
root@ccosvms0070:~/admin# kubectl logs kube-dns-6f4fd4bdf-rb4x8 -c kubedns --namespace kube-system
I0201 13:29:23.613716 1 dns.go:48] version: 1.14.6-3-gc36cb11
I0201 13:29:23.615050 1 server.go:69] Using configuration read from directory: /kube-dns-config with period 10s
I0201 13:29:23.615128 1 server.go:112] FLAG: --alsologtostderr="false"
I0201 13:29:23.615144 1 server.go:112] FLAG: --config-dir="/kube-dns-config"
I0201 13:29:23.615153 1 server.go:112] FLAG: --config-map=""
I0201 13:29:23.615158 1 server.go:112] FLAG: --config-map-namespace="kube-system"
I0201 13:29:23.615165 1 server.go:112] FLAG: --config-period="10s"
I0201 13:29:23.615173 1 server.go:112] FLAG: --dns-bind-address="0.0.0.0"
I0201 13:29:23.615180 1 server.go:112] FLAG: --dns-port="10053"
I0201 13:29:23.615189 1 server.go:112] FLAG: --domain="cluster.local."
I0201 13:29:23.615198 1 server.go:112] FLAG: --federations=""
I0201 13:29:23.615204 1 server.go:112] FLAG: --healthz-port="8081"
I0201 13:29:23.615209 1 server.go:112] FLAG: --initial-sync-timeout="1m0s"
I0201 13:29:23.615213 1 server.go:112] FLAG: --kube-master-url=""
I0201 13:29:23.615221 1 server.go:112] FLAG: --kubecfg-file=""
I0201 13:29:23.615227 1 server.go:112] FLAG: --log-backtrace-at=":0"
I0201 13:29:23.615237 1 server.go:112] FLAG: --log-dir=""
I0201 13:29:23.615244 1 server.go:112] FLAG: --log-flush-frequency="5s"
I0201 13:29:23.615251 1 server.go:112] FLAG: --logtostderr="true"
I0201 13:29:23.615256 1 server.go:112] FLAG: --nameservers=""
I0201 13:29:23.615263 1 server.go:112] FLAG: --stderrthreshold="2"
I0201 13:29:23.615268 1 server.go:112] FLAG: --v="2"
I0201 13:29:23.615274 1 server.go:112] FLAG: --version="false"
I0201 13:29:23.615283 1 server.go:112] FLAG: --vmodule=""
I0201 13:29:23.615368 1 server.go:194] Starting SkyDNS server (0.0.0.0:10053)
I0201 13:29:23.615753 1 server.go:213] Skydns metrics enabled (/metrics:10055)
I0201 13:29:23.615781 1 dns.go:146] Starting endpointsController
I0201 13:29:23.615791 1 dns.go:149] Starting serviceController
I0201 13:29:23.615880 1 logs.go:41] skydns: ready for queries on cluster.local. for tcp://0.0.0.0:10053 [rcache 0]
I0201 13:29:23.615898 1 logs.go:41] skydns: ready for queries on cluster.local. for udp://0.0.0.0:10053 [rcache 0]
I0201 13:29:24.116020 1 dns.go:170] Initialized services and endpoints from apiserver
I0201 13:29:24.116075 1 server.go:128] Setting up Healthz Handler (/readiness)
I0201 13:29:24.116108 1 server.go:133] Setting up cache handler (/cache)
I0201 13:29:24.116121 1 server.go:119] Status HTTP port 8081
I0201 13:31:02.019718 1 server.go:153] Ignoring signal terminated (can only be terminated by SIGKILL)
# kube-dns pods has crashed 39 times in 1 hour
root@ccosvms0070:~/admin# kubectl get pods --namespace kube-system
NAME READY STATUS RESTARTS AGE
etcd-ccqservkm1 1/1 Running 0 1h
kube-apiserver-ccqservkm1 1/1 Running 0 1h
kube-controller-manager-ccqservkm1 1/1 Running 0 1h
kube-dns-6f4fd4bdf-rb4x8 0/3 CrashLoopBackOff 39 1h
kube-proxy-28ssp 1/1 Running 0 1h
kube-proxy-4kb9z 1/1 Running 0 1h
kube-proxy-67vk2 1/1 Running 0 1h
########################################################################
Use REPLY-ALL to reply to list
To unsubscribe from the QSERV-L list, click the following link:
https://listserv.slac.stanford.edu/cgi-bin/wa?SUBED1=QSERV-L&A=1
|
