Smartphy

從 qingwei personal wiki
於 2018年5月22日 (二) 07:27 由 Qingwei (對話 | 貢獻) 所做的修訂 (→‎download)
跳到： 導覽， 搜尋

Android

發佈release版本的apk

a.  Get the latest code from the git repository (android-app.git)
b.  Install Android Studio (If you have not already done so)
c.  Open the project in Android Studio.
d.  Try a debug build before the official release to make sure code compiles & run it either on emulator/actual phone.
e.  Go to build.grade file to edit version/build number (/Users/vidhsubr/IdeaProjects/android_app/android-app/app/build.gradle)
f.  Now to make an official release, please click on build drop down menu, choose option "Generate Signed APK"
g.  A pop up window appears; you'd need to fill in the below details
h.  choose app module in the pop up and hit "next"
i.  Keystore path will be "<project-path>/android-app/android.jks"; For ex : /Users/vidhsubr/IdeaProjects/android-app/android.jks
j.  Key store password: cisco123
k.  Key alias: myandroidkey
l.  Key password: cisco123; hit next
m.  choose apk destination folder; For ex /Users/vidhsubr/IdeaProjects/android_app/android-app/app
n.  Build type is "release"
o.  signature versions are v1 & v2
p.  hit finish
q.  If everything compiles well & there are no errors then "app-release.apk" file will be found in the previously specified location.

CBR 配置

真RPD ipv6 online

CRDC-NG-E09#sh run int tenGigabitEthernet 4/1/7
Load for five secs: 4%/1%; one minute: 6%; five minutes: 6%
Time source is NTP, 20:59:25.271 PST Mon Mar 5 2018

Building configuration...

Current configuration : 170 bytes
!
interface TenGigabitEthernet4/1/7
 ip address 60.10.10.5 255.255.255.0
 ipv6 address 2001:60:10:10::5/64
 ipv6 enable
 ipv6 dhcp relay destination 2001:20:1:1::33  ### 删掉这一行 不能IPV6上线，切换到IPV4上线
end

# 前提是RPD的DS要选DS controller，这个DS配置了IPV6地址， 如下7/1/0
CRDC-NG-E09#sh run int tenGigabitEthernet 7/1/0 
Load for five secs: 4%/1%; one minute: 8%; five minutes: 7%
Time source is NTP, 20:58:47.282 PST Mon Mar 5 2018

Building configuration...

Current configuration : 223 bytes
!
interface TenGigabitEthernet7/1/0
 vrf forwarding lcha-710
 ip address 60.10.10.1 255.255.255.0
 ip helper-address global 20.1.0.33
 cdp enable
 ipv6 address 2001:60:10:10::1/64
 ipv6 enable
 ipv6 nd ra suppress all
end

video

profile

CRDC-NG-E09#sh run | s profile 67
cable downstream controller-profile 67
 rf-chan 32
  type VIDEO ASYNC
  qam-profile 1
  rf-output NORMAL

support virtual splitting

cable downstream controller-profile 68
 multicast-pool 62   # !!!
 rf-chan 64
  type VIDEO ASYNC
  qam-profile 1
  rf-output NORMAL
cable virtual-service-group sg-6-0-3 downstream-cable 6/0/2 profile 68
cable virtual-service-group video-service-group downstream-cable 7/0/1 profile 68

# pool 配置
CRDC-NG-E09#sh run | s multi
cable depi multicast pool 62
 ip address 239.1.1.0 255.255.255.0
cable depi multicast pool 66
 ipv6 address FF3A::8000:0/120
cable depi multicast pool 127
 ip address 227.226.225.0 255.255.255.0

LCHA switch over

CRDC-NG-E09#redundancy linecard-group switchover from slot 6

看到online信息

CRDC-NG-E09#sh run | s logging
logging host 10.74.54.188 transport tcp port 8514

UI

啟動

BASE_URL=10.74.54.185 BASE_PORT=9000 HOST=10.74.54.185 npm start

# 如何kill
ps -ef | grep webpack
qingwyan  9392  9376 15 22:40 pts/21   00:00:48 node /home/qingwyan/workspace/robot-ui/node_modules/.bin/webpack-dev-server --inline --progress

build UI container

# 先到robot-ui目录下
git submodule init
git submodule update --remote

/bin/npm install
/bin/npm run build

MJ=1
D=`date +%y%m%d`
B=0
N=`git rev-parse --short HEAD`
TAG=$MJ.$B.$N.$D

sudo docker build -t dockerhub.cisco.com/robot-dockerprod/robot-ui:developv$TAG .

替換robot里的UI

# build UI container 后，build好的Image可以通过docker images看到
[qingwyan@crdc-sdn-ucs1 robot-ui]$ sudo docker images 
REPOSITORY                                                      TAG                          IMAGE ID            CREATED             SIZE
dockerhub.cisco.com/robot-dockerprod/robot-ui                   developv1.0.46a5914.180117   563a2bb56717        54 minutes ago      869.2 MB

# 保存tar到本地
docker save -o ui.tar 563a2bb56717
 
# 将tar cp到robot 里
# 之后导入本地镜像到robot里
sudo docker load --input ui.tar

# 命名刚刚导入的image
docker tag 563a2bb56717 localbuild/ui

# 之后的操作跟rpd 设置Debug 模式一样

iptable

container里ping不通外面

sudo iptables -t nat -A POSTROUTING -s  192.168.125.0/24 -o ens192 -j MASQUERADE

常用命令

# 删一条iptable
sudo iptables -t nat -D PREROUTING 1
# 删所有iptable
iptables -F -t nat
# 查看iptable
iptables -t nat -L --line-number | more

GCP container

快速進入docker的腳本

KEY=k8s_gcp  # gcp 换成 rpd

container_id=$(docker ps | grep $KEY | awk {'print $1'})
docker exec -it ${container_id} bash

# 一句话
docker exec -it $(docker ps | grep k8s_gcp | awk {'print $1'}) bash

debug

docker pull "dockerhub.cisco.com/cabu-cmts-orch-docker/gcp-service:debug"
# 查看log
tail -f /var/log/robot/gcp-service_stdout.log

設置debug模式

kubectl get pods --all-namespaces
kubectl exec --namespace=kube-system -it <robot-orch-container> bash
# 进入orch container 后， 修改orchmanifest.json

robotctl spec   - generates the new yaml file for the containers
robotctl update – my guess updates pod.conf
robotctl stop – stop existing default namespace containers, sometimes I needed to wait for few seconds till they stopped
kubectl get pods --all-namespaces         --- check that all default namespace containers stopped
robotctl start – start all containers
kubectl get pods --all-namespaces         --- verify all containers RUNNING
# copy image
scp qingwyan@10.75.12.21:/home/qingwyan/workspace3/gcp-service/spring/target/gcp-service-0.1-SNAPSHOT.jar .
# 启动
java -jar -Xdebug -Xrunjdwp:transport=dt_socket,server=y,suspend=n,address=5006 gcp-service-0.1-SNAPSHOT.jar

一條命令kill java進程

ps -ef | grep java | grep -v grep | awk {'print $2'} | xargs kill -9
# grep -v grep 过滤掉auto出现的进程号

swagger ui

  • 1.6.6 version(robot)

https://10.74.54.185:30604/rpd-service-manager/swagger-ui.html#

rpd container

debug

# 查看日志
tail -f /var/log/robot/rpd-service-manager_stdout.log

# 拉debug image
docker pull "dockerhub.cisco.com/cabu-cmts-orch-docker/rpd-service-manager:debug"

# copy debug image
scp qingwyan@10.75.12.21:/home/qingwyan/workspace3/rpd-service-manager/spring/target/rpd-service-manager-0.1-SNAPSHOT.jar .
# run debug image
java -jar -Xdebug -Xrunjdwp:transport=dt_socket,server=y,suspend=n,address=5005 rpd-service-manager-0.1-SNAPSHOT.jar

container里stop process

supervisorctl start rpd-service-manager
supervisorctl stop rpd-service-manager

nso

docker run

1 sudo docker run --publish 8083:8080 -itd dockerhub.cisco.com/robot-docker-v2/robot-cfgsvc

常用命令

1 # 进入CLI
2 bash-4.2# ncs_cli -u admin -C

traces

1 admin@ncs% set devices global-settings trace raw
2 # 保存位置
3 /var/log/robot/ncs-logs/xxx.trace

讀配置

curl 命令讀配置

# 读的是集合
curl -i -u admin:admin -H "Accept:application/vnd.yang.collection+json" -s http://127.0.0.1:8080/api/config/devices/device/f0bf1cdf-bee1-3525-817c-b593fcb8c8bb/config/ios:cable/virtual-service-group

show 命令讀配置

# 从NSO上读配置
admin@ncs# show running-config devices device f0bf1cdf-bee1-3525-817c-b593fcb8c8bb config ios:cable virtual-service-group 
devices device f0bf1cdf-bee1-3525-817c-b593fcb8c8bb
 config
  ios:cable virtual-service-group sg-6-0-3 downstream-cable 6/0/2 profile 68
  ios:cable virtual-service-group sg-7-0-2-d downstream-cable 7/0/2 profile 67
  ios:cable virtual-service-group sg-7-0-8 downstream-cable 7/0/8 profile 68
  ios:cable virtual-service-group test_sg downstream-cable 7/0/5 profile 67
  ios:cable virtual-service-group video-service-group downstream-cable 7/0/1 profile 68
 !
!

debug手段

Unknown SSH host key

 1 admin@ncs# devices device f0bf1cdf-bee1-3525-817c-b593fcb8c8bb sync-from
 2 result false
 3 info Failed to authenticate towards device f0bf1cdf-bee1-3525-817c-b593fcb8c8bb: Unknown SSH host key
 4 
 5 admin@ncs# devices device f0bf1cdf-bee1-3525-817c-b593fcb8c8bb ssh fetch-host-keys 
 6 result updated
 7 fingerprint {
 8     algorithm ssh-rsa
 9     value 9e:7d:f8:c4:14:b9:87:4c:f0:a9:bf:36:41:65:80:ab
10 }

狀態為disable

1 admin@ncs# show devices device state
2 devices device 2c678746-c18d-3f32-8bcc-74a64c76da91
3  state oper-state     disabled
4  state oper-state-error-tag noconnection
5 # 查看错误原因
6 admin@ncs# devices device 2c678746-c18d-3f32-8bcc-74a64c76da91 sync-from 
7 result false
8 info Failed to authenticate towards device 2c678746-c18d-3f32-8bcc-74a64c76da91: Bad password for local/remote user admin/admin Auth failed

no space left on device

# robot 里
cd /mnt/robot_logfs/robot-cfgsvc/1/ncs-logs

rm -rf rollback*

重啟NSO

bash-4.2# cd ncs-run/
bash-4.2# ls
README.ncs  ncs-cdb  ncs.conf  ncs.conf.bak  packages  scripts	state  target
bash-4.2# ncs --stop
bash-4.2# ncs

看log

bash-4.2# pwd
/var/log/robot/ncs-logs
vi ncs-java-vm.log

手動添加一個device

ncs_cli -u admin
configure

set devices authgroups group lab default-map
set devices authgroups group lab default-map remote-name admin
set devices authgroups group lab default-map remote-password lab

set devices device E09 address 20.6.6.2
set devices device E09 port 22
set devices device E09 authgroup lab
set devices device E09 device-type cli
set devices device E09 device-type cli ned-id cisco-ios
set devices device E09 state admin-state unlocked

commit
request devices device E09 sync-from

rollback 配置

cd ncs-run/
vi ncs.conf
<rollback>
    <enabled>true</enabled>
    <directory>/var/log/robot/ncs-logs</directory>
    <history-size>500</history-size>
  </rollback>

NED team support

add cmd 開case

https://techzone.cisco.com/t5/Network-Services-Orchestrator/How-to-Open-a-TAC-NSO-NED-Case/ta-p/955288

download

  • ncs

https://earth.tail-f.com:8443/ncs-pkgs/cisco-ios/4.5.5/

  • nso

https://earth.tail-f.com:8443/ncs/

build NSO docker

# upload to engci
curl --noproxy cisco.com -i -X PUT -u robot-dev-deployer:qqyiwhc7ts6um8sw -v --retry 999 --retry-max-time 0 -T ncs-4.5.5-cisco-ios-5.8.2.tar.gz "http:/engci-maven-master.cisco.com/artifactory/robot-dev-thirdparty/com.cisco/cfgsvc/nso-4.5.5/ncs-4.5.5-cisco-ios-5.8.2.tar.gz"

# :downloadNso
Download http://engci-maven-master.cisco.com/artifactory/robot-dev-thirdparty/com.cisco/cfgsvc/nso-4.4.2/nso-4.4.2.linux.x86_64.installer.bin

cassandra

在liunx上run

docker run --publish 9042:9042 --publish 9160:9160 -d --name cassandra --net robot_network -e CASSANDRA_START_RPC=true -it cassandra:3.10

dhcp

ipv6

啟動ipv6 server

sudo /usr/local/sbin/dhcpd -6 -d -cf /etc/dhcp/dhcpd6.conf sdn-br

k8s

常用命令

# 删除一个pod
kubectl delete deployment rpd-service-manager

# apply 一个yaml
kubectl apply -f rpd-service-manager.yaml

# 查看所有运行的pods
kubectl get pods --all-namespaces

robot下單獨換cobtainer

kubectl delete service robot-ui
kubectl delete deployment robot-ui

kubectl apply -f robot-ui_service.yaml
kubectl apply -f robot-ui.yaml

從k8s_robot-orch下cp yaml文件

KEY=k8s_robot-orch

container_id=$(docker ps | grep $KEY | awk {'print $1'})
docker cp ${container_id}:/root/rpd-service-manager.yaml .

docker

doker ps 不需要用sudo

# Cannot connect to the Docker daemon. Is the docker daemon running on this host?
sudo groupadd docker
sudo usermod -aG docker $(whoami)
# restart VM才生效？

設置docker代理

cat /etc/systemd/system/docker.service.d/http-proxy.conf
[Service]
Environment="HTTP_PROXY=http://64.103.36.133:8080" "HTTPS_PROXY=http://64.103.36.133:8080" "NO_PROXY=localhost,.cisco.com, sdntools.cisco.com,10.124.8.66"
[jifu@crdc-sdn-ucs1 robot-ui]$ 

then

systemctl daemon-reload
service docker restart

test

pyats運行環境

source /auto/sjc-ccebuild-ar5/pyats-gate/env.sh 
source /auto/pysw/ats64/env.sh
source ~/proxy.sh

cnr8

設置CCAP core

Design > DHCP Settings > Policies
# 搜索 2001:60
# 进入 Edit DHCP Policy 2001:60:10:10-E09
# 修改 Configured Options, 替换 ccap-cores 后的 2001:10:74:54::186
(enterprise-id 4491((tftp-servers 32 2001:20:1:1::33)(ccap-cores 61 2001:20:6:1::2)(rfc868-servers 37 2001:20:1:1::33)(syslog-servers 34 2001:20:1:1::33)(time-offset 38 8h)(cablelabs-client-configuration 2170 )))
#!!! 修改好后save, return回上级页面也要save!!!

restart dhcp

Deploy > DHCP > DHCP Server

kvm

restart guest server

virsh shutdown sp1210

virsh start sp1210

# 强制shutdown
virsh destroy sp1210
取自 "http://106.14.10.253:8190/index.php?title=Smartphy&oldid=325"