rancher 部署应用 记录
<p>1.rancher中快速部署应用
<a href="http://blog.zhenglin.work/k8s/deploy_app_in_rancher.html">http://blog.zhenglin.work/k8s/deploy_app_in_rancher.html</a>
<a href="https://www.cnblogs.com/ExMan/p/13719195.html">https://www.cnblogs.com/ExMan/p/13719195.html</a></p>
<p>kubectl 常用的命令总结
<a href="http://blog.zhenglin.work/k8s/kubectl-user-instruction.html">http://blog.zhenglin.work/k8s/kubectl-user-instruction.html</a></p>
<p>2.修改pods数量
Global/user-cluster, Edit进入集群编辑页面，
点 ‘Edit as YAML’ 按钮，在YAML文件中找到111-112附近，内容如下:</p>
<pre><code class="language-bash">111 # # 修改节点最大 Pod 数量
112 # max-pods: '250' </code></pre>
<p>把112这一行的注释去掉即可。</p>
<p>3.Kubernetes 外部 IP Service 类型(浮动IP：Floating IP + Anycast IP)
<a href="https://segmentfault.com/a/1190000023567011">https://segmentfault.com/a/1190000023567011</a>
Rancher容器网络－Floating IP解决方案
<a href="http://www.2cto.com/net/201704/619681_2.html">http://www.2cto.com/net/201704/619681_2.html</a>
浮动IP，负载平衡器或Rancher的最佳做法是什么？
<a href="https://forums.rancher.com/t/floating-ips-load-balancers-or-what-would-be-the-best-practice-in-rancher/8819">https://forums.rancher.com/t/floating-ips-load-balancers-or-what-would-be-the-best-practice-in-rancher/8819</a>
<a href="https://docs.projectcalico.org/networking/add-floating-ip">https://docs.projectcalico.org/networking/add-floating-ip</a></p>
<p>单节点只能起109个容器，这个怎么改呀
<a href="https://github.com/k3s-io/k3s/issues/697">https://github.com/k3s-io/k3s/issues/697</a>
Ksd:
如果你是 curl安装的，你就添加一个 INSTALL_K3S_EXEC="--kubelet-arg=max-pods=1022"
Ksd:
<a href="https://docs.rancher.cn/docs/k3s/installation/install-options/how-to-flags/_index#%E7%A4%BA%E4%BE%8B-b%EF%BC%9Ainstall_k3s_exec">https://docs.rancher.cn/docs/k3s/installation/install-options/how-to-flags/_index#%E7%A4%BA%E4%BE%8B-b%EF%BC%9Ainstall_k3s_exec</a>
你在 K3S systemd 的配置文件中添加 --kubelet-arg=max-pods=1022
k3s.service的这个文件</p>
<pre><code class="language-bash">[Service]
Type=notify
EnvironmentFile=/etc/systemd/system/k3s.service.env
KillMode=process
Delegate=yes
# Having non-zero Limit*s causes performance problems due to accounting overhead
# in the kernel. We recommend using cgroups to do container-local accounting.
LimitNOFILE=1048576
LimitNPROC=infinity
LimitCORE=infinity
TasksMax=infinity
TimeoutStartSec=0
Restart=always
RestartSec=5s
ExecStartPre=-/sbin/modprobe br_netfilter
ExecStartPre=-/sbin/modprobe overlay
ExecStart=/usr/local/bin/k3s \
server \
'--kubelet-arg=max-pods=1022' \</code></pre>
<p>master 和agent节点都需要改是吧？
对
rke续签CA和全部服务证书报的错
答：rke cert rotate --rotate-ca</p>