nerdalert

TSD/Grafana Dashboards

• Submariner Multi-Active Gateway Aggregate Netperf Dashboard (Also now has Prometheus reporting) http://44.203.139.9/d/r4KWO5E7r/submariner-multi-active-gateway-netperf-dashboard?orgId=1&refresh=30s&from=now-3h&to=now

• Submariner Single Gateway Aggregate Netperf Dashboard: http://44.203.139.9/d/r4KWOeR7e/submariner-single-active-gateway-netperf-dashboard?orgId=1&refresh=30s&from=now-3h&to=now

Topologies:

nerdalert

[02:03:39] (='o'=) [submariner-ec2-custom-no-active-ha]$ ansible-playbook setup-k8s.yml -vv
ansible-playbook [core 2.12.3]
  config file = /Users/brent/Downloads/code/redhat-et/apr15-performance-benchmarking/performance-benchmarking/submariner-ec2-custom-no-active-ha/ansible.cfg
  configured module search path = ['/Users/brent/.ansible/plugins/modules', '/usr/share/ansible/plugins/modules']
  ansible python module location = /Users/brent/Library/Python/3.10/lib/python/site-packages/ansible
  ansible collection location = /Users/brent/.ansible/collections:/usr/share/ansible/collections
  executable location = /Users/brent/Library/Python/3.10/bin/ansible-playbook
  python version = 3.10.3 (v3.10.3:a342a49189, Mar 16 2022, 09:34:18) [Clang 13.0.0 (clang-1300.0.29.30)]

nerdalert

[fedora@cluster2-ha1-perf-target-node ~]$ kubectl get pods --all-namespaces -o wide
NAMESPACE             NAME                                             READY   STATUS             RESTARTS         AGE   IP            NODE                            NOMINATED NODE   READINESS GATES
kube-system           local-path-provisioner-84bb864455-5zq8r          1/1     Running            0                76m   10.42.0.3     cluster2-ha1-perf-target-node   <none>           <none>
kube-system           coredns-96cc4f57d-bz2ns                          1/1     Running            0                76m   10.42.0.2     cluster2-ha1-perf-target-node   <none>           <none>
kube-system           metrics-server-ff9dbcb6c-vz9w2                   1/1     Running            0                76m   10.42.0.4     cluster2-ha1-perf-target-node   <none>           <none>
submariner-operator   submariner-operator-789c775b8-xj5tv              1/1     Running            1 (67m ago)      67m   10.42.2.2     cluster2-worker1-node     

nerdalert

[fedora@cluster2-ha1-perf-target-node ~]$ kubectl get nodes
NAME                            STATUS   ROLES                  AGE   VERSION
cluster2-worker1-node           Ready    <none>                 15m   v1.22.7+k3s1
cluster2-ha2-node               Ready    <none>                 18m   v1.22.7+k3s1
cluster2-ha1-perf-target-node   Ready    control-plane,master   21m   v1.22.7+k3s1
[fedora@cluster2-ha1-perf-target-node ~]$ kubectl get pods --all-namespaces -o wide
NAMESPACE             NAME                                             READY   STATUS             RESTARTS      AGE   IP            NODE                            NOMINATED NODE   READINESS GATES
kube-system           local-path-provisioner-84bb864455-hn6km          1/1     Running            0             21m   10.42.0.3     cluster2-ha1-perf-target-node   <none>           <none>
kube-system           coredns-96cc4f57d-j9ggn                          1/1     Running            0             21m   10.42.0.4     cluster2-ha1-perf-target-node   

nerdalert

• Node Addresses:

[brokerNode]
35.171.157.41 ansible_user=fedora ansible_connection=ssh k8s_master=10.10.0.80 hostname=cluster1-ha1-node

[secondaryGatewayNode]
18.234.103.8 ansible_user=fedora ansible_connection=ssh k8s_master=10.10.0.80 hostname=cluster1-ha2-node

nerdalert

Logs and Cluster Details from GlobalNet Crashing with active/active branch

• Makes/Images

git clone https://github.com/astoycos/submariner-operator.git
git clone https://github.com/astoycos/submariner.git
cd submariner
git checkout remotes/origin/AS-mult-active-gw-multiple-vxlan-tunnels
make images

nerdalert

fedora@cluster2-perf-target:~$ sudo tcpdump -i any host 10.42.0.18
tcpdump: data link type LINUX_SLL2
dropped privs to tcpdump
tcpdump: verbose output suppressed, use -v[v]... for full protocol decode
listening on any, link-type LINUX_SLL2 (Linux cooked v2), snapshot length 262144 bytes
08:11:59.594219 cni0  Out IP 242.0.0.7.60267 > 10.42.0.18.netperf: Flags [S], seq 1739230431, win 62377, options [mss 8911,sackOK,TS val 2341609812 ecr 0,nop,wscale 7], length 0
08:11:59.594224 veth6505e1e8 Out IP 242.0.0.7.60267 > 10.42.0.18.netperf: Flags [S], seq 1739230431, win 62377, options [mss 8911,sackOK,TS val 2341609812 ecr 0,nop,wscale 7], length 0
08:11:59.594250 veth6505e1e8 P   IP 10.42.0.18.netperf > 242.0.0.7.60267: Flags [S.], seq 2914677477, ack 1739230432, win 62293, options [mss 8911,sackOK,TS val 422794371 ecr 2341609812,nop,wscale 7], length 0
08:11:59.594255 cni0  In  IP 10.42.0.18.netperf > 242.0.0.7.60267: Flags [S.], seq 2914677477, ack 1739230432, win 62293, options [mss 8911,sackOK,TS val 422794371

nerdalert

# Iperf

bash-5.0# iperf3 -c 242.1.255.251
Connecting to host 242.1.255.251, port 5201
[  5] local 10.42.0.14 port 43472 connected to 242.1.255.251 port 5201
[ ID] Interval           Transfer     Bitrate         Retr  Cwnd
[  5]   0.00-1.00   sec   123 MBytes  1.03 Gbits/sec    9    530 KBytes
[  5]   1.00-2.00   sec   120 MBytes  1.01 Gbits/sec    7    617 KBytes

nerdalert

image.txt contains quay.io/prometheus/busybox

• Running:

sudo ./make-microshift-app-images-rpm.sh \
  --image-list ./images.txt \
  --container-dir /var/lib/containers/storage/overlay-images \
  --rpmbuild-dir /home/brent/rpmbuild

nerdalert

Install and Confiure VNC on Linux

• quick and dirty, the proper way is via a systemd unit but it can be a PITA to setup.

sudo dnf install tigervnc-server -y
touch ~/.Xauthority

• Check the output off the vncserver command for a port in the 590* range.