开发用的devstack (by quqi99)

已于 2024-05-08 15:43:58 修改
阅读量1.9k 收藏 2
点赞数
分类专栏： OpenStack Non-Networking 文章标签： devstack
于 2019-07-29 06:50:20 首次发布
版权声明：本文为博主原创文章，遵循 CC 4.0 BY-SA 版权协议，转载请附上原文出处链接和本声明。
本文链接：https://blog.csdn.net/quqi99/article/details/97622336
版权

作者：张华 发表于：2014-05-15
版权声明：可以任意转载，转载时请务必以超链接形式标明文章原始出处和作者信息及本版权声明 (http://blog.csdn.net/quqi99 )

导入镜像

镜像是在kvm上做的，做完转成vmware镜像的方法如下：

qemu-img convert -f qcow2 -O vmdk o7k.qcow2 o7k.vmdk

可在win10上下载vmware player (player是免费版), 如 VMware-player-16.1.2-17966106.exe - https://www.vmware.com/cn/products/workstation-player.html
vmware player免费版没有直接导入镜像的功能，但可以这样导入：

  • 创建一个新虚机，"Create a New Virtual Machine -> I will install the operating system later"进去之后随便创建一块硬盘（注意此处选择Store virtual dsik as a single file）
  • 虚机创建好之后，再在虚机的设置中将上面创建的硬盘删除，然后再新建一块硬盘(Hard Disk)，此时可以选择导入已有的镜像(Use an existing virtual disk)
  • 虚机设置使用8G内存及2个vCPU, 其他用默认即可。
  • 另外非常重要的一点是，虚机应该使用UEFI模式而不是BIOS模式。但免费版的vmware player也没有提供此项配置。但我们可以在它的 .vmx 文件 （.vmx是虚机文件，.vmdk是虚机的硬盘)的最后一行添加下行即可(可参见：https://www.serverwatch.com/guides/enabling-uefi-on-virtual-machines/）:
firmware="efi"
  • 然后启动虚机后可能没有IP, 那是因为网卡名字可能改化了。需修改netplan的配置文件/etc/netplan/00-installer-config.yaml改成正确的网卡名称(网卡名称可通过ip addr show获取), 然后运行sudo netplan apply即可获得IP

快速运行

sudo mkdir -p /bak/openstack && sudo chown -R $(whoami) /bak/openstack && cd /bak/openstack
sudo chown -R $(whoami) ~/.config
git clone https://github.com/openstack/devstack.git
cd devstack && wget https://raw.githubusercontent.com/zhhuabj/exercise/master/o7k/local.conf
#sudo systemctl stop "devstack@*"
#./clean.sh
./stack.sh
source accrc/admin/admin

[可选] apt加速

如果网速很慢，可由一人搭建apt-cacher-ng server, 其他人用它来避免使用apt-get时的网络拥挤。

sudo apt install apt-cacher-ng -y
echo 'PassThroughPattern: .*' |sudo tee -a /etc/apt-cacher-ng/acng.conf
sudo systemctl restart apt-cacher-ng.service && sudo systemctl enable apt-cacher-ng.service
du -sh /var/cache/apt-cacher-ng/
#vim /var/lib/dpkg/info/apt-cacher-ng.postinst
#dpkg --configure apt-cacher-ng

#Use apt cache proxy
echo 'Acquire::http::Proxy "http://proxy:3142";' | sudo tee /etc/apt/apt.conf.d/01acng

如果想要换目录的话才需要

#change the dir from /var/cache/apt-cacher-ng/ to /mnt/udisk/apt-cacher-ng
cat << EOF |sudo tee -a /etc/fstab
#use blkid to see uuid
UUID="d63d7251-ec3d-4ef5-aa92-f3d4c480f20c" /mnt/udisk   ext4    defaults    0  2
EOF
mkfs.ext4 -F -L udisk /dev/sdb1
mkdir /mnt/udisk/apt-cacher-ng
chown -R apt-cacher-ng:apt-cacher-ng /mnt/udisk/apt-cacher-ng
sudo sed -i 's/CacheDir: \/var\/cache\/apt-cacher-ng/CacheDir: \/mnt\/udisk\/apt-cacher-ng/g' /etc/apt-cacher-ng/acng.conf
du -sh /mnt/udisk/apt-cacher-ng

[可选]pypi加速

对于加速pypi, 可使用devpi缓存

#https://opensource.com/article/18/7/setting-devpi
sudo apt install python3-pip -y
sudo pip install devpi-server devpi-web
devpi-init --serverdir /bak/work/devpi
devpi-server --serverdir /bak/work/devpi --host 0.0.0.0

pip install --trusted-host 192.168.2.6 -i http://192.168.2.6:3141/root/pypi/+simple/ httpie

export PIP_INDEX_URL=http://192.168.2.6:3141/root/pypi/+simple/
pip install httpie

unset PIP_INDEX_URL
mkdir -p ~/.pip && cat > ~/.pip/pip.conf << EOF
[global]
index-url = http://192.168.2.6:3141/root/pypi/+simple/
[search]
index = http://192.168.2.6:3141/root/pypi/
[install]
trusted-host=192.168.2.6
disable-pip-version-check = true
timeout = 120
EOF

也可以使用阿里等第三方的:

mkdir -p ~/.pip
cat << EOF |tee ~/.pip/pip.conf
[global]
trusted-host=mirrors.aliyun.com
index-url = http://mirrors.aliyun.com/pypi/simple
disable-pip-version-check = true
timeout = 120
EOF

或者在devstack中添加： PYPI_ALTERNATIVE_URL=http://mirrors.aliyun.com/pypi/simple

[可选] 创建虚机

例如，在win10上使用hypver-v创建一个7G(初始2G并动态分配即可)，40G硬件的ubuntu 20.04的名为o7k的虚机, 禁用check point减小硬盘使用：

  • Win10中搜索’hyper-v’打开’Turn Windows features on or off’将’Hyper-v’,　‘Virtual Machine Platform’, 'Windows Hypervior Platform’都打开
  • 下载ubuntu 20.04 -　http://mirrors.aliyun.com/ubuntu-releases/20.04/ubuntu-20.04.2-live-server-amd64.iso
  • 在’hyper-v manager’中点击’Create Virtual Machine’, 然后从’local　installtion source’中选择iso,　Disable ‘Secure Boot’, 内存7000M(初始2G并动态分配即可), 禁用检查点，其它设置默认
  • 打开虚机里面的kvm嵌套虚拟化(参考：https://superuser.com/questions/1330271/configuring-kvm-inside-ubuntu-hyper-v-client), 先下载：https://raw.githubusercontent.com/MicrosoftDocs/Virtualization-Documentation/master/hyperv-tools/Nested/Enable-NestedVm.ps1　然后以管理员打开powershell, 先运行＂Set-ExecutionPolicy Bypass" 以允许在powershell中运行脚本，然后运行：.\Enable-NestedVm.ps1 o7k
  • [可选]安装docker用docker - https://hub.docker.com/editions/community/docker-ce-desktop-windows/，将devstack运行在docker容器中可参考：　https://github.com/slimjim2234/docker-devstack
  • [可选]安装完了，可以压缩:
# stop vm first
diskpart
select vdisk file="D:\images\o7k.vhdx"
compact vdisk
#compress it by winrar, then move it to D:\www\sftp, then we can download it by: sftp ftpuser@192.168.2.9X
  • 将vhdx格式转化为其他格式
unrar x o7k.rar
qemu-img convert o7k.vhdx -O qcow2 o7k.qcow2 #kvm
qemu-img convert o7k.vhdx -O raw o7k.raw
qemu-img convert o7k.vhdx -O vmdk o7k.vmdk   #vmware
qemu-img convert o7k.vhdx -O vdi o7k.vdi     #virtualbox

例如，使用kvm来使用o7k.qcow2时，会遇到＂booting from hard disk ＂错误，那是因为hyper-v是默认使用UEFI来创建虚机的，那应该让KVM也支持UEFI:

sudo apt install ovmf && sudo systemctl restart libvirtd
ls /usr/share/OVMF/OVMF_CODE.fd
然后在virt-manager安前装在＂概况＂中将Firmware选择"UEFI x86_64: /usr/shar/OVMF/OVMF_CODE.fd"即可．

[可选]创建stack用户

ssh登录进虚机创建stack用户，此步可选, 注意，若添加了stack用户, 记得运行’sudo su - stack’之后做pip相关的设置:

echo 'ubuntu     ALL=(ALL) NOPASSWD:ALL' |sudo tee -a /etc/sudoers
echo 'precedence ::ffff:0:0/96 100' |sudo tee -a /etc/gai.conf
# don't add # when coping via vim
echo 'autocmd FileType * setlocal formatoptions-=c formatoptions-=r formatoptions-=o' |tee ~/.vimrc

sudo useradd -s /bin/bash -d /opt/stack -m stack
echo "stack ALL=(ALL) NOPASSWD: ALL" | sudo tee /etc/sudoers.d/stack
sudo su - stack && cd /opt/stack

[可选] 设置镜像

NOTE: 设置镜像太容易引发各种奇奇怪怪的包依赖问题，我本次未使用各种镜像．

  • 使用deb镜像．NOTE：使用镜像，如果镜像不同步，可能会引入更多的包依赖问题．
    我此次实际使用的是:cn.archive.ubuntu.com，你也可以使用：mirrors.cloud.tencent.com
sudo cp /etc/apt/sources.list /etc/apt/sources.list_bak
sudo sed -i s/archive.ubuntu.com/cn.archive.ubuntu.com/g /etc/apt/sources.list
sudo apt clean all
sudo apt update
  • 使用pip镜像．NOTE: 使用镜像，如果镜像质量不好不同步，可能会引入更多的包依赖问题．
    例如，我这次使用douban的pypi镜像时就遇到了下列包依赖的问题：
ERROR: Could not find a version that satisfies the requirement cliff===3.8.0 (from -c /bak/openstack/requirements/upper-constraints.txt (line 249)) (from versions: none)
ERROR: No matching distribution found for cliff===3.8.0 (from -c /bak/openstack/requirements/upper-constraints.txt (line 249))

所以我此次实际未使用镜像．附录１有本次使用douban镜像的步骤，仅供参考：

  • github（如git@github.com:openstack/nova.git）有可能被间歇性封锁, 建议尽量使用https协议避免使用git协议．也可考虑trystack的openstack代码镜像(http://git.trystack.cn ), 当然它的代码及时性就不好说了．真要开发肯定不能用这种，搭建测试环境暂时可以用．

运行devstack创建OpenStack环境

#NOTE: this can cause bastion unable to be SSHed
#juju deploy ubuntu openstack --series=focal --config hostname=openstack --constraints "mem=8G cores=2 root-disk=60G"
sudo mkdir -p /bak/openstack && sudo chown -R $(whoami) /bak/openstack && cd /bak/openstack
git clone https://github.com/openstack/devstack.git && cd devstack
#wget https://raw.githubusercontent.com/zhhuabj/exercise/master/o7k/local.conf
sudo bash -c 'cat > local.conf' <<EOF
[[local|localrc]]

#OFFLINE=True
DEST=/bak/openstack
LIBVIRT_TYPE=qemu
mkdir -p $DEST
unset http_proxy
unset https_proxy
#sudo route del -net 10.0.1.0/24 gw 192.168.101.3 > /dev/null 2>&1
sudo apt -y install openvswitch-switch qemu-kvm libvirt-clients libvirt-daemon-system bridge-utils
#sudo virsh net-destroy default
sudo ovs-vsctl del-br br-data
sudo ovs-vsctl -- --may-exist add-br br-data 
sudo ip addr add 172.16.1.1/24 dev br-data >/dev/null 2>&1 &
sudo chown -R $(whoami) /opt/stack/
#sudo ovs-vsctl -- --may-exist add-port br-data eth0 -- set interface eth0 type=internal

# ENABLED_SERVICES=n-cpu,rabbit,neutron,q-agt
GIT_BASE="https://git.openstack.org"
#enable_plugin neutron-lbaas https://git.openstack.org/openstack/neutron-lbaas
ENABLED_SERVICES=rabbit,mysql,key,g-api,g-reg
ENABLED_SERVICES+=,n-api,n-crt,n-obj,n-cpu,n-cond,n-sch,n-cauth,n-novnc,placement-api,n-super-cond,n-cond-cell1
ENABLED_SERVICES+=,q-svc,q-agt,q-dhcp,q-l3,q-meta,neutron
disable_service q-metering
Q_USE_ROOTWRAP=True
Q_USE_ROOTWRAP_DAEMON=False
ENABLE_IDENTITY_V2=False
enable_service placement-api
heartbeat_timeout_threshold=7200
Q_ML2_TENANT_NETWORK_TYPE=flat,vlan,gre,vxlan
Q_ML2_PLUGIN_TYPE_DRIVERS=flat,vlan,gre,vxlan
# must use public
OVS_BRIDGE_MAPPINGS=public:br-data
PUBLIC_NETWORK_NAME=public

# Neutron Agent configuration
Q_USE_NAMESPACE=True
Q_ALLOW_OVERLAPPING_IP=True
# Neutron ML2 configuration
Q_PLUGIN=ml2
#Q_DVR_MODE=dvr_snat
Q_DVR_MODE=legacy
# VLAN
ENABLE_TENANT_VLANS=True
ML2_VLAN_RANGES=public:1000:2999
# GRE
ENABLE_TENANT_TUNNELS=True
TENANT_TUNNEL_RANGE=1000:2000
Q_ML2_PLUGIN_VXLAN_TYPE_OPTIONS=(vni_ranges=1000:2000)

# Avanced ML2 configuration
Q_AGENT=openvswitch
Q_ML2_PLUGIN_MECHANISM_DRIVERS=openvswitch,linuxbridge,l2population

# OVN
#enable_plugin networking-ovn http://git.openstack.org/openstack/networking-ovn
#enable_service ovn-northd
#enable_service ovn-controller
# We have to disable the neutron L2 agent. OVN does not use the L2 agent.
#disable_service q-agt

# Swift
#ENABLED_SERVICES+=,s-proxy,s-object,s-container,s-account
#sudo apt -y install libapache2-mod-wsgi
VOLUME_BACKING_FILE_SIZE=500M
SWIFT_HASH=66a3d6b56c1f479c8b4e70ab5c2000f5
SWIFT_REPLICAS=1
SWIFT_DATA_DIR=$DEST/data/swift
NOVA_VNC_ENABLED=True

# Ceph
sudo rm -rf /var/lib/ceph/mon/ceph-`hostname`
#ENABLED_SERVICES+=,ceph
CEPH_LOOPBACK_DISK_SIZE=3G
CEPH_CONF=/etc/ceph/ceph.conf
CEPH_REPLICAS=1

# Cinder - Block Device Service
#ENABLED_SERVICES+=,cinder,c-api,c-vol,c-sch,c-bak
CINDER_DRIVER=ceph
CINDER_CEPH_USER=cinder
CINDER_CEPH_POOL=volumes
CINDER_CEPH_UUID=6d52eb95-12f3-47e3-9eb9-0c1fe4142426
CINDER_BAK_CEPH_POOL=backeups
CINDER_BAK_CEPH_USER=cind-backeups
CINDER_ENABLED_BACKENDS=ceph,lvm
GLANCE_CEPH_USER=glance
GLANCE_CEPH_POOL=images
NOVA_CEPH_POOL=vmz

HOST_IP=172.16.1.1
SERVICE_HOST=$HOST_IP
MYSQL_HOST=$SERVICE_HOST
#QPID_HOST=$SERVICE_HOST
RABBIT_HOST=$SERVICE_HOST
GLANCE_HOSTPORT=$SERVICE_HOST:9292
Q_HOST=$SERVICE_HOST

FIXED_RANGE=10.0.1.0/24
#113 -- 118, http://jodies.de/ipcalc
FLOATING_RANGE=192.168.101.0/24
Q_FLOATING_ALLOCATION_POOL=start=192.168.101.3,end=192.168.101.100
PUBLIC_NETWORK_GATEWAY=192.168.101.1
NETWORK_GATEWAY=10.0.1.1
PUBLIC_BRIDGE=br-data
# sudo ovs-vsctl add-port br-ex eth0
OVS_PHYSICAL_BRIDGE=br-data
# https://review.openstack.org/#/c/87987/
IP_VERSION=4
DATABASE_USER=root
DATABASE_PASSWORD=password
ADMIN_PASSWORD=password
SERVICE_PASSWORD=password
RABBIT_PASSWORD=password
SERVICE_TOKEN=ADMIN
LOGFILE=$DEST/logs/stack.log
ENABLE_DEBUG_LOG_LEVEL=False
SYSLOG=False
VERBOSE=True
SCREEN_LOGDIR=$DEST/logs
LOG_COLOR=False
Q_USE_DEBUG_COMMAND=False
KEYSTONE_DEPLOY=mod_wsgi
USE_SSL=False
disable_service tls-proxy
USE_PYTHON3=True
DOWNLOAD_DEFAULT_IMAGES=False
IMAGE_URLS="http://download.cirros-cloud.net/0.5.2/cirros-0.5.2-x86_64-disk.img"
#IPSEC_PACKAGE=strongswan

# don't use trystack git mirror, but just use noVNC from trystack
#GIT_BASE=http://git.trystack.cn
# git call failed: [git clone' https://github.com/novnc/noVNC.git
NOVNC_REPO=http://git.trystack.cn/kanaka/noVNC.git
#SPICE_REPO=http://git.trystack.cn/git/spice/sice-html5.gi

disable_service etcd3
disable_service memory_tracker

#devstack NET_ID 'Failure creating NET_ID for private
#NEUTRON_CREATE_INITIAL_NETWORKS=False

#HORIZON_REPO=https://github.com/openstack/horizon
#HORIZON_BRANCH=stable/grizzly
#CINDER_REPO=https://github.com/openstack/cinder.git
#CINDER_BRANCH=stable/grizzly
#CINDERCLIENT_REPO=https://github.com/openstack/python-cinderclient.git
#CINDERCLIENT_BRANCH=stable/grizzly
#NOVA_REPO=https://github.com/openstack/nova.git
#NOVA_BRANCH=stable/grizzly
#NOVACLIENT_REPO=https://github.com/openstack/python-novaclient.git
#NOVACLIENT_BRANCH=stable/grizzly
#GLANCE_REPO=https://github.com/openstack/glance.git
#GLANCE_BRANCH=stable/grizzly
#GLANCECLIENT_REPO=https://github.com/openstack/python-glanceclient.git
#KEYSTONE_REPO=https://github.com/openstack/keystone.git
#KEYSTONE_BRANCH=stable/grizzly
#KEYSTONECLIENT_REPO=https://github.com/openstack/python-keystoneclient.git
#KEYSTONECLIENT_BRANCH=stable/grizzly
#NEUTRON_REPO=https://github.com/openstack/neutron.git
#NEUTRON_BRANCH=stable/grizzly
#NEUTRONCLIENT_REPO=https://github.com/openstack/python-neutronclient.git
#NEUTRONCLIENT_BRANCH=stable/grizzly
#CEILOMETER_REPO=git://git.openstack.org/openstack/ceilometer.git
#CEILOMETERCLIENT_REPO=git://git.openstack.org/openstack/python-ceilometerclient.git
#HEAT_REPO=git://git.openstack.org/openstack/heat.git
#HEAT_BRANCH=master
#SWIFT_REPO=git://git.openstack.org/openstack/swift.git
#SWIFT_BRANCH=master
#OPENSTACKCLIENT_REPO=git://git.openstack.org/openstack/python-openstackclient.git
#OPENSTACKCLIENT_BRANCH=master
#SOLUM_REPO=git://git.openstack.org/stackforge/solum.git
#SOLUM_BRANCH=master

# Enable Ceilometer
#enable_service ceilometer-acompute
#enable_service ceilometer-acentral
#enable_service ceilometer-anotification
#enable_service ceilometer-collector
#enable_service ceilometer-api
#enable_service ceilometer-alarm-notifier
#enable_service ceilometer-alarm-evaluator
#CEILOMETER_PIPELINE_INTERVAL=60

# Enable Solum
#enable_service solum
#enable_service solum-api
#enable_service solum-build-api
#enable_service solum-deployer
#enable_service solum-conductor
#enable_service solum-worker
#SOLUM_INSTALL_CEDARISH=True

# Enable Marconi
#disable_service marconi-server
#enable_service zaqar-server

#https://wiki.openstack.org/wiki/Docker
#export INSTALLDIR=/bak/openstack && cd $INSTALLDIR/nova-docker
#git clone https://git.openstack.org/stackforge/nova-docker
#./contrib/devstack/prepare_devstack.sh
#VIRT_DRIVER=docker
#DEFAULT_IMAGE_NAME=cirros
#NON_STANDARD_REQS=1
#IMAGE_URLS=""

#Enable tacker
#TACKER_BRANCH=master
#enable_plugin heat https://git.openstack.org/openstack/heat $TACKER_BRANCH
#enable_plugin networking-sfc https://git.openstack.org/openstack/networking-sfc $TACKER_BRANCH
#enable_plugin barbican https://git.openstack.org/openstack/barbican $TACKER_BRANCH
#enable_plugin mistral https://git.openstack.org/openstack/mistral $TACKER_BRANCH
#enable_plugin tacker https://git.openstack.org/openstack/tacker $TACKER_BRANCH
EOF

然后使用＂./stack.sh＂运行它，注意非常重要的一点，如果想要重新运行避免出莫名其妙的错误的话，必须：

#re-run
sudo systemctl stop "devstack@*"
./clean.sh
./stack.sh

Use it

source accrc/admin/admin
$ env |grep OS_
OS_PROJECT_DOMAIN_ID=default
OS_CACERT=
OS_AUTH_URL=http://172.16.1.1/identity
OS_USER_DOMAIN_ID=default
OS_USERNAME=admin
OS_PROJECT_NAME=admin
OS_PASSWORD=password

source <(openstack complete)
openstack complete |sudo tee /etc/bash_completion.d/openstack.bash_completion > /dev/null

systemctl -a |grep devstack
sudo tail -f /var/log/syslog
sudo journalctl -f --unit devstack@n-cpu.service

# only need to create network/subnet staffs when NEUTRON_CREATE_INITIAL_NETWORKS=true
openstack router create provider-router
openstack network create private
openstack subnet create --subnet-range 10.0.1.0/24 --network private --allocation-pool start=10.0.1.50,end=10.0.1.100 --gateway 10.0.1.1 private-subnet
openstack network create --provider-network-type flat --external --share --provider-physical-network physnet1 ext_net
openstack subnet create --allocation-pool start=192.168.101.3,end=192.168.101.100 --subnet-range 192.168.101.0/24 --dhcp --gateway 192.168.101.1 --ip-version 4 --network ext_net ext_net_subnet
#openstack subnet set ext_net_subnet --dhcp  #just simulate ext_net so use dhcp
openstack router add subnet provider-router private-subnet
openstack router set --external-gateway ext_net provider-router

cat << EOF > user-data
#cloud-config
user: ubuntu
password: password
chpasswd: { expire: False }
EOF
# in case you don't have a key
#ssh-keygen -q -N ""
openstack keypair create --public-key ~/.ssh/id_rsa.pub mykey
openstack --debug server create --image cirros-0.5.2-x86_64-disk --flavor m1.tiny --key-name mykey --nic net-id=$(openstack network list | awk '/private/ {print $2}') --user-data ./user-data --config-drive true i1

#add FIP
public_network=$(openstack network show public -f value -c id)
fip=$(openstack floating ip create $public_network -f value -c floating_ip_address)
fix_ip=$(openstack server list |grep 'private=' |awk -F '=' '{print $2}' |awk '{print $1}')
openstack floating ip set $fip --fixed-ip-address $fix_ip --port $(openstack port list --fixed-ip ip-address=$fix_ip -c id -f value)

#add SG rules
sg_id=$(openstack security group list --project=$(openstack server show i1 -f value -c project_id) |grep default |awk '{print $2}')
openstack security group rule create --protocol tcp --dst-port 22 $sg_id
openstack security group rule create --protocol icmp $sg_id
sudo ip netns exec qrouter-6aef358e-7cbe-4d77-af70-ed13fd11f621 ping $fix_ip
openstack console log show i1
#we just simulate fip so we should ping fip from qg-xxx interface
sudo ip netns exec qrouter-6aef358e-7cbe-4d77-af70-ed13fd11f621 ping $fip

Debug it

sudo su - stack
cd /bak/openstack/devstack
sudo systemctl --failed
sudo systemctl list-units |grep devstack |grep failed |awk '{print $2}'
sudo systemctl status devstack@*
source accrc/admin/admin
openstack server list

# if above doesn't work, we may need to rerun devstack
sudo systemctl stop devstack@*
./stack.sh
openstack keypair create --public-key ~/.ssh/id_rsa.pub mykey
# 有时创建虚机失败时，换个名字如i2就好了　
openstack --debug server create --image cirros-0.5.2-x86_64-disk --flavor m1.tiny --key-name mykey --nic net-id=$(openstack network list | awk '/private/ {print $2}') i1   
# if fails, pls check
sudo systemctl --failed
sudo tail -f /var/log/syslog
sudo systemctl -a |grep devstack@
sudo systemctl restart devstack@n-cpu.service
sudo journalctl -f --unit devstack@n-cpu.service
see - https://docs.openstack.org/devstack/latest/systemd.html

附录1 - 使用pip镜像

pypi.douban.com的pip镜像质量不高，同步不及时，导致我在此次使用过程中遇到了下列包依赖问题．

ERROR: Could not find a version that satisfies the requirement cliff===3.8.0 (from -c /bak/openstack/requirements/upper-constraints.txt (line 249)) (from versions: none)
ERROR: No matching distribution found for cliff===3.8.0 (from -c /bak/openstack/requirements/upper-constraints.txt (line 249))

去掉它就好了，下面只是记录使用douban镜像的过程，仅作参考：
NOTE: 一定要使用pip镜像时也可以试试清华的（pip config set global.index-url https://pypi.tuna.tsinghua.edu.cn/simple　）

当前用户或stack用户添加：

mkdir ~/.pip/
cat << EOF > ~/.pip/pip.conf
[global]
index-url = http://pypi.douban.com/simple
[install]
trusted-host=pypi.douban.com
disable-pip-version-check = true
timeout = 120
EOF
此时，"sudo pip3 install -r requirements.txt＂还是未生效，那是因为需要继续在root用户中也添加：
sudo mkdir /root/.pip
sudo bash -c 'cat > /root/.pip/pip.conf' << EOF
[global]
index-url = http://pypi.douban.com/simple
[install]
trusted-host=pypi.douban.com
disable-pip-version-check = true
timeout = 120
EOF
或者使用：
[global]
trusted-host=mirrors.aliyun.com
index-url = http://mirrors.aliyun.com/pypi/simple
disable-pip-version-check = true
timeout = 120

若不使用pip镜像，例如，运行下列命令时，
sudo -H LC_ALL=en_US.UTF-8 http_proxy= https_proxy= no_proxy= PIP_FIND_LINKS= SETUPTOOLS_SYS_PATH_TECHNIQUE=rewrite python3.8 -m pip install -c /bak/openstack/requirements/upper-constraints.txt 'glance-store[cinder]!=0.29.0'
会报这种错误：
urllib3.exceptions.ReadTimeoutError: HTTPSConnectionPool(host='files.pythonhosted.org', port=443): Read timed out.
可以局部通过如下添加＂--index-url https://pypi.douban.com/simple＂解决：
sudo -H LC_ALL=en_US.UTF-8 http_proxy= https_proxy= no_proxy= PIP_FIND_LINKS= SETUPTOOLS_SYS_PATH_TECHNIQUE=rewrite python3.8 -m pip install -c /bak/openstack/requirements/upper-constraints.txt 'glance-store[cinder]!=0.29.0' --index-url https://pypi.douban.com/simple
对于小包可以通过添加超时时间解决（pip --default-timeout=100 install -U pip）

附录2 - 其他可能的问题

1, 当将devstack中的OFFLINE设置为True后重新运行./stack.sh报下列错：
apache2.service: Failed with result 'start-limit-hit'
Soluation:
从这个网页找到了答案：https://www.hiroom2.com/2017/02/18/linux-systemd-s-start-request-repeated-too-quickly-for-xxx-service/
修改文件（/lib/systemd/system/apache2.service）添加下设置禁用掉systemd对重启服务次数的限制．
StartLimitIntervalSec=0
StartLimitBurst=0
然后运行
sudo systemctl daemon-reload
sudo systemctl reset-failed apache2.service

2, 运行＂cd /bak/openstack/horizon && /usr/bin/python3.8 manage.py compilemessages＂报下列错：
ModuleNotFoundError: No module named 'django.utils.log.NullHandler'; 'django.utils.log' is not a package
Solution:
sed -i "s/django.utils.log.NullHandler/logging.NullHandler/g" ./openstack_dashboard/local/local_settings.py

3, 运行＂sudo ebtables -t broute -L＂报下列错：
table `broute' is incompatible, use 'nft' tool
Solution:
Try Switching to iptable-legacy
sudo apt update && sudo apt upgrade
sudo apt -y install iptables arptables ebtables
sudo update-alternatives --set iptables /usr/sbin/iptables-legacy || true
sudo update-alternatives --set ip6tables /usr/sbin/ip6tables-legacy || true
sudo update-alternatives --set arptables /usr/sbin/arptables-legacy || true
sudo update-alternatives --set ebtables /usr/sbin/ebtables-legacy || true

20210915更新　- 运行devstack镜像注意点

#注意点
1, 实验用的虚机镜像是UEFI格式的，所以导入虚机镜像时记得使用UEFI。
2, 虚机启动后没有自动获得IP的问题，那是因为：
   a, 物理机libvirt的default网络若使用了默认的192.168.122.0/24网段，虚机内部也安装了libvirt会用这个默认网段。
　    虚机内部使用virsh net-edit default修改一个不同网段并重启虚机即可。
   b, 若虚机内的网卡名不叫eth0, 如enp1s0.
      需修改/etc/netplan/00-installer-config.yaml将eth0改为enps1s0, 然后运行sudo netplan apply即可获得IP
3, 最好devstack运行成功之后禁用它的重新开机后的自动重启。因为devstack机器重启之后总是会遇到各种问题，索性每次重运行一次 
　 sudo systemctl -a |grep devstack@ |xargs sudo systemctl disable
   sudo systemctl -a

#重新运行, OFFLINE=true模式不work, 仍应确保使用unstack.sh而不是clean.sh
sudo su - stack
cd /bak/openstack/devstack
#./clean.sh
./unstack.sh
./stack.sh

#运行成功之后记得及时disable所有服务避免下次重启机器时再各种问题
sudo systemctl -a |grep devstack@ |xargs sudo systemctl disable

#验证
sudo systemctl --failed
sudo systemctl status devstack@*
source accrc/admin/admin
openstack server list

#使用, 有时创建虚机会失败，换个名字如i2就好了　
openstack keypair create --public-key ~/.ssh/id_rsa.pub mykey
openstack --debug server create --image cirros-0.5.2-x86_64-disk --flavor m1.tiny --key-name mykey --nic net-id=$(openstack network list | awk '/private/ {print $2}') i1

#访问虚机
PROJECT_ID=$(openstack project show --domain default admin -f value -c id)
SG_ID=$(openstack security group list --project ${PROJECT_ID} | awk '/default/ {print $2}')
openstack security group rule list $SG_ID
openstack security group rule create $SG_ID --ingress --ethertype IPv4 --protocol icmp --remote-ip 0.0.0.0/0
openstack security group rule create $SG_ID --ingress --ethertype IPv4 --protocol tcp --dst-port 22:22 --remote-ip 0.0.0.0/0
#you may fail to ssh cirros becuase dropbear key and openssh key are incompatible, change non-cirros will succeed
#sudo ip netns exec qrouter-b4a099e0-d0db-4ec8-8f3c-9d83e11d3488 ssh cirros@10.0.1.20 -v

#浮动IP
# eg: vm=10.0.1.20
public_network=$(openstack network show public -f value -c id)
fip=$(openstack floating ip create $public_network -f value -c floating_ip_address)
openstack floating ip set $fip --fixed-ip-address 10.0.1.20 --port $(openstack port list --fixed-ip ip-address=10.0.1.20 -c id -f value)

#调试
sudo systemctl --failed
sudo tail -f /var/log/syslog
sudo systemctl -a |grep devstack@
sudo systemctl restart devstack@n-cpu.service
sudo journalctl -f --unit devstack@n-cpu.service

20230403 - another local.conf

hua@minipc:/bak/openstack/devstack$ cat local.conf
[[local|localrc]]
#sudo apt -y install iptables arptables ebtables -y
#sudo update-alternatives --set iptables /usr/sbin/iptables-legacy || true
#sudo update-alternatives --set ip6tables /usr/sbin/ip6tables-legacy || true
#sudo update-alternatives --set arptables /usr/sbin/arptables-legacy || true
#sudo update-alternatives --set ebtables /usr/sbin/ebtables-legacy || true
#sudo ebtables -t broute -L
#curl https://bootstrap.pypa.io/get-pip.py -o get-pip.py && python3 ./get-pip.py
#sudo pip install --upgrade setuptools
#
# Network Design
# Physical Network: HOST_IP=192.168.99.179, used for openstack itself, eg: keystone, mysql, DB etc
# FIP Network: devstack will set its GW 192.168.101.1 in br-data, and create a SNAT rule for 192.168.101.0/24
# VM Network: openstack will create geneve private-subnet=10.0.0.0/26, and create virtual GW 10.0.0.1/26 
#git checkout -b zed origin/stable/zed
TARGET_BRANCH=stable/2023.1
HOST_IP=$(ip route list default |grep src |awk '{print $9}')
#If OVN_DNS_SERVERS is not set, VM will use 127.0.0.53, then systemd-resolv inside VM will fail to start, don't know yet
OVN_DNS_SERVERS=192.168.99.1
enable_service neutron-dns
LIBVIRT_CPU_MODE=host-passthrough
#for multi hosts, SERVICE_HOST specifies controller's HOST_IP
#ENABLED_SERVICES=n-cpu,c-vol,placement-client,ovn-controller,ovs-vswitchd,ovsdb-server,q-ovn-metadata-agent
SERVICE_HOST=$HOST_IP
MYSQL_HOST=$SERVICE_HOST
RABBIT_HOST=$SERVICE_HOST
GLANCE_HOSTPORT=$SERVICE_HOST:9292
NOVA_VNC_ENABLED=True
NOVNCPROXY_URL="http://$SERVICE_HOST:6080/vnc_lite.html"
VNCSERVER_LISTEN=$HOST_IP
VNCSERVER_PROXYCLIENT_ADDRESS=$VNCSERVER_LISTEN
ADMIN_PASSWORD=password
DATABASE_PASSWORD=$ADMIN_PASSWORD
RABBIT_PASSWORD=$ADMIN_PASSWORD
SERVICE_PASSWORD=$ADMIN_PASSWORD
#enable ml2-dns to parse the hostname of VMs
Q_ML2_PLUGIN_EXT_DRIVERS=port_security,qos,dns
#should enable SG for 'juju bootstrap'
Q_USE_SECGROUP=True
#need to add the following 3 commands in /etc/rc.local as well
#ip link set dev br-data up >/dev/null 2>&1
#ip addr add 192.168.101.1/24 dev br-data up >/dev/null 2>&1
#iptables -t nat -A POSTROUTING -s 192.168.101.0/24 ! -d 192.168.101.0/24 -j MASQUERADE >/dev/null 2>&1
#sudo systemctl enable $(find /etc/systemd/system -name 'devstack@*.service' -print)
FLOATING_RANGE=192.168.101.0/24
Q_FLOATING_ALLOCATION_POOL=start=192.168.101.10,end=192.168.101.99
PUBLIC_NETWORK_GATEWAY=192.168.101.1
#sudo ip l add name veth-br-eth0 type veth peer name veth-ex >/dev/null 2>&1
#sudo ip l set dev veth-br-eth0 up
#sudo ip l set dev veth-ex up
#sudo ip l set veth-br-eth0 master br-eth0
#PUBLIC_INTERFACE=veth-ex
#don't need veth-ex here because devstack will create SNAT for FIP subnet
Q_USE_PROVIDERNET_FOR_PUBLIC=True
OVS_PHYSICAL_BRIDGE=br-data
PUBLIC_BRIDGE=$OVS_PHYSICAL_BRIDGE
OVS_BRIDGE_MAPPINGS=public:br-data
IP_VERSION=4
ENABLE_DEBUG_LOG_LEVEL=False
SYSLOG=True
VERBOSE=True
LOG_COLOR=True
LOGDAYS=3
DOWNLOAD_DEFAULT_IMAGES=False
IMAGE_URLS="http://download.cirros-cloud.net/0.6.2/cirros-0.6.2-x86_64-disk.img"
#PYPI_ALTERNATIVE_URL=http://pypi.douban.com/simple
#GIT_BASE=http://git.trystack.cn
#enable_plugin designate https://opendev.org/openstack/designate
disable_service tempest
disable_service dstat
disable_service memory_tracker
#cp: cannot stat '/opt/stack/data/venv/etc/glance/rootwrap.*': No such file or directory
GLOBAL_VENV=False
sudo systemctl restart openvswitch-switch.service
#ln: failed to create symbolic link '/var/run/ovn/openvswitch': File exist
sudo rm -rf /var/run/ovn
# Unable to create the network. No tenant network is available for allocation. unstack.sh and then stack.sh

这次主要遇到了3个问题：

  • HOST_IP就用eth0或br-eth0的IP(192.168.99.179)就行，它是用于给openstack组件用的。FIP用192.168.101.0/24, 虚机网络用默认的10.0.0.0/26, neutron会创建virtual GW 10.0.0.1/26, devstack会创建ovs bridge br-data同时在br-data设置FIP GW=192.168.101.1/24，也会为192.168.101.0/24创建SNAT rule.
  • 用devstack创建分布式环境的话，在另一个计算节点上设置不同的HOST_IP,如192.168.99.180, 然后SERVICE_HOST指向控制节点的HOST_IP(192.168.99.179), 最后添加计算相关的服务：ENABLED_SERVICES=n-cpu,c-vol,placement-client,ovn-controller,ovs-vswitchd,ovsdb-server,q-ovn-metadata-agent
  • 另外，发现如果不设置dns-server, 它默认会使用127.0.0.53, 进虚机调试时发现systemd-resolved启动失败，最后添加下面两句解决
OVN_DNS_SERVERS=192.168.99.1
enable_service neutron-dns

#enable ml2-dns
sudo vi /etc/neutron/neutron.conf
dns_domain = devstack.lan.
sudo vi /etc/neutron/plugins/ml2/ml2_conf.ini
[ml2]
extension_drivers = port_security,qos,dns
sudo systemctl restart devstack@q-svc.service

neutron subnet-update private-subnet --dns_nameservers list=true 192.168.99.1
$ sudo ovn-nbctl list dhcp_options |grep options
options             : {classless_static_route="{169.254.169.254/32,10.0.1.2, 0.0.0.0/0,10.0.1.1}", dns_server="{192.168.99.1}", domain_name="\"devstack.lan.\"", lease_time="43200", mtu="1442", router="10.0.1.1", server_id="10.0.1.1", server_mac="fa:16:3e:7b:39:9f"}
ubuntu@i1:~$ sudo resolvectl status ens3 |grep -E 'Current DNS Server|DNS Domain'
Current DNS Server: 192.168.99.1
        DNS Domain: devstack.lan
  • 也最好启用ml2-dns来解析虚机的名字 - Q_ML2_PLUGIN_EXT_DRIVERS=port_security,qos,dns
  • 要想devstack在机器重启之后自动可用，除了设置 sudo systemctl enable $(find /etc/systemd/system -name ‘devstack@*.service’ -print) 之外，还要在/etc/rc.local (chmod +x /etc/rc.local)里设置下列三个命令来添加本来是devstack脚本添加的IP与SNAT （不设置它，会导致jujucontroller里的dns服务失败从而不法通过juju创建虚机)
ip link set dev br-data up >/dev/null 2>&1
ip addr add 192.168.101.1/24 dev br-data up >/dev/null 2>&1
iptables -t nat -A POSTROUTING -s 192.168.101.0/24 ! -d 192.168.101.0/24 -j MASQUERADE >/dev/null 2>&1
  • 做了上面一步的设置，机器重启之后devstack仍然不能用的话， 从日志中发现有时是devstack@etcd.serveice 没有启动，运行下列命令即可恢复devstack openstack环境
 sudo systemctl restart devstack@*
 sudo systemctl status devstack@* |grep Active
  • 若没加LIBVIRT_CPU_MODE=host-passthrough， 也可手工加：
vim /etc/nova/nova-cpu.conf
[libvirt]
live_migration_uri = qemu+ssh://hua@%s/system
#cpu_model = Nehalem
#cpu_mode = custom
cpu_mode = host-passthrough
virt_type = kvm
  • 之前遇到过的:
sudo apt -y install iptables arptables ebtables -y
sudo update-alternatives --set iptables /usr/sbin/iptables-legacy || true
sudo update-alternatives --set ip6tables /usr/sbin/ip6tables-legacy || true
sudo update-alternatives --set arptables /usr/sbin/arptables-legacy || true
sudo update-alternatives --set ebtables /usr/sbin/ebtables-legacy || true
sudo ebtables -t broute -L
  • 一定不要用USE_VENV=True, 有一个相关的lp bug说社区的CI没有测试USE_VENV=True，并且有olso和requirement这些公共库就是让所有openstack组件可以运行一套pip包的，好像记得执意用USE_VENV=True在g-api处无法正常启动。
  • 若error和tools/install_pip.sh相关，注释掉’set -o errexit’可以做为workaround. 若error和worlddump.py相关需导入setuptools
hua@minipc:/bak/openstack/devstack$ git diff
diff --git a/tools/install_pip.sh b/tools/install_pip.sh
index 7c5d4c65..7e203bcf 100755
--- a/tools/install_pip.sh
+++ b/tools/install_pip.sh
@@ -7,7 +7,7 @@
 # Assumptions:
 # - PYTHON3_VERSION refers to a version already installed
 
-set -o errexit
+#set -o errexit
 
 # Keep track of the current directory
 TOOLS_DIR=$(cd $(dirname "$0") && pwd)
diff --git a/tools/worlddump.py b/tools/worlddump.py
index e2921737..78f6ca28 100755
--- a/tools/worlddump.py
+++ b/tools/worlddump.py
@@ -19,6 +19,7 @@
 
 import argparse
 import datetime
+import setuptools
 from distutils import spawn
 import fnmatch
 import io

下列脚本可以重建混乱的pip依赖:

find . -name "*.pyc" -exec rm -rf {} \;
#echo 'deleting ~/cache/pip ...'
#rm -rf ~/.cache/pip/*
#sudo chown -R $USER  ~/.cache/pip/
echo 'deleting dist-packages ...'
sudo rm -rf /usr/local/lib/python3.10/dist-packages/oslo*
sudo rm -rf /usr/local/lib/python3.10/dist-packages/*keystone*
sudo rm -rf /usr/local/lib/python3.10/dist-packages/*glance*
sudo rm -rf /usr/local/lib/python3.10/dist-packages/*swift*
sudo rm -rf /usr/local/lib/python3.10/dist-packages/*cinder*
sudo rm -rf /usr/local/lib/python3.10/dist-packages/*nova*
sudo rm -rf /usr/local/lib/python3.10/dist-packages/*neutron*
#root_dir=/bak/openstack
root_dir=/opt/stack
for pro in requirements keystone glance cinder horizon nova neutron tempest; do
  echo $pro
  cd $pro
  git diff > diff && git checkout . && git checkout zed >/dev/null 2>&1
  git pull
  sudo pip uninstall -y -r requirements.txt
  sudo pip uninstall -y -r test-requirements.txt
  sudo pip install --upgrade -r requirements.txt
  sudo pip install --upgrade -r test-requirements.txt
  #sudo python setup.py develop
  cd $root_dir
done;

20230913 Update - Run devstack on AliCloud VM

阿里云虚机系统是22.04，不需要支持嵌套虚拟化。推荐是4 vCPUs, 8G memory, 30G dist. 但如下，安装之前实际内存用了4.8G，disk用了8.5G

stack@iZ2ze94xd36la2abgiwrftZ:~$ free -h |head -n2
               total        used        free      shared  buff/cache   available
Mem:           7.1Gi       4.8Gi       201Mi        32Mi       2.2Gi       2.0Gi
stack@iZ2ze94xd36la2abgiwrftZ:~$ df -h |grep vda |head -1
/dev/vda3        40G  8.5G   29G  23% /

下面是安装步骤（主要是github无法访问所以需禁用etcd3, 并且使用git.trystack.cn代替github, 同时使用pip mirror. 最后因为我这个阿里云打开了IPv6所以需要设置IPv4 prefer这样才能访问cirros-cloud.net顺利下载cirros2镜像)

#Create the user stack, and make sure to use this user for the test
useradd -s /bin/bash -d /opt/stack -m stack
chmod +x /opt/stack
echo "stack ALL=(ALL) NOPASSWD: ALL" | sudo tee /etc/sudoers.d/stack
#Make sure to run this command so that cirros-cloud.net can be visited normally
echo 'precedence ::ffff:0:0/96 100' |sudo tee -a /etc/gai.conf
sudo -u stack -i

#Use pip mirror to bypass the wall
mkdir -p ~/.pip
cat << EOF |tee ~/.pip/pip.conf
[global]
trusted-host=pypi.tuna.tsinghua.edu.cn
index-url = https://pypi.tuna.tsinghua.edu.cn/simple
disable-pip-version-check = true
timeout = 120
EOF
#pip3 install pbr -i https://pypi.tuna.tsinghua.edu.cn/simple --trusted-host pypi.tuna.tsinghua.edu.cn

#Run devstack
git clone http://git.trystack.cn/openstack/devstack.git -b stable/zed
cd devstack
cat << EOF |tee ./local.conf
[[local|localrc]]
# Network Design
# Physical Network: HOST_IP, used for openstack itself, eg: keystone, mysql, DB etc
# FIP Network: devstack will set its GW 192.168.101.1 in br-data, and create a SNAT rule for 192.168.101.0/24
# VM Network: openstack will create geneve private-subnet=10.0.0.0/26, and create virtual GW 10.0.0.1/26
TARGET_BRANCH=stable/zed
#aliyun doens't support nested virtualization so use qemu instead
LIBVIRT_TYPE=qemu
#LIBVIRT_CPU_MODE=host-passthrough
HOST_IP=$(ip route list default |grep src |awk '{print $9}')
#If OVN_DNS_SERVERS is not set, VM will use 127.0.0.53, then systemd-resolv inside VM will fail to start, don't know yet
OVN_DNS_SERVERS=$(resolvectl status | grep 'Current DNS Server' | awk -F':' '{print $2}' | head -n1 | sed -e 's/^[ \t]*//')
enable_service neutron-dns
SERVICE_HOST=\$HOST_IP
MYSQL_HOST=\$SERVICE_HOST
RABBIT_HOST=\$SERVICE_HOST
GLANCE_HOSTPORT=\$SERVICE_HOST:9292
NOVA_VNC_ENABLED=True
NOVNCPROXY_URL=http://\$SERVICE_HOST:6080/vnc_lite.html
VNCSERVER_LISTEN=\$HOST_IP
VNCSERVER_PROXYCLIENT_ADDRESS=\$VNCSERVER_LISTEN
ADMIN_PASSWORD=password
DATABASE_PASSWORD=\$ADMIN_PASSWORD
RABBIT_PASSWORD=\$ADMIN_PASSWORD
SERVICE_PASSWORD=\$ADMIN_PASSWORD
#enable ml2-dns to parse the domain of VMs
Q_ML2_PLUGIN_EXT_DRIVERS=port_security,qos,dns
Q_USE_SECGROUP=True
#add the following 3 commands in /etc/rc.local as well if hoping devstack to take effect after the restart
#ifconfig br-data 192.168.101.1/24 >/dev/null 2>&1 &
#iptables -t nat -D POSTROUTING -s 192.168.101.0/24 ! -d 192.168.101.0/24 -j MASQUERADE >/dev/null 2>&1 &
#iptables -t nat -A POSTROUTING -s 192.168.101.0/24 ! -d 192.168.101.0/24 -j MASQUERADE >/dev/null 2>&1 &
#sudo systemctl enable \$(find /etc/systemd/system -name 'devstack@*.service' -print)
FLOATING_RANGE=192.168.101.0/24
Q_FLOATING_ALLOCATION_POOL=start=192.168.101.10,end=192.168.101.99
PUBLIC_NETWORK_GATEWAY=192.168.101.1
Q_USE_PROVIDERNET_FOR_PUBLIC=True
OVS_PHYSICAL_BRIDGE=br-data
PUBLIC_BRIDGE=\$OVS_PHYSICAL_BRIDGE
OVS_BRIDGE_MAPPINGS=public:br-data
IP_VERSION=4
ENABLE_DEBUG_LOG_LEVEL=False
SYSLOG=True
VERBOSE=True
LOG_COLOR=True
LOGDAYS=3
DOWNLOAD_DEFAULT_IMAGES=False
#Make sure to run this command so that cirros-cloud.net can be visited normally
#echo 'precedence ::ffff:0:0/96 100' |sudo tee -a /etc/gai.conf
IMAGE_URLS="http://download.cirros-cloud.net/0.6.2/cirros-0.6.2-x86_64-disk.img"
#enable_plugin designate https://opendev.org/openstack/designate
disable_service tempest
disable_service dstat
disable_service memory_tracker
#disable etcd3 to avoid visiting github that has been blocked
disable_service etcd3
#ln: failed to create symbolic link '/var/run/ovn/openvswitch': File exist
sudo rm -rf /var/run/ovn
GIT_BASE=http://git.trystack.cn
NOVNC_REPO=http://git.trystack.cn/kanaka/noVNC.git
SPICE_REPO=http://git.trystack.cn/git/spice/spice-html5.git
EOF
#./unstack.sh && sudo systemctl restart openvswitch-switch
./stack.sh

source openrc
#when hoping to use admin user
#unset OS_SERVICE_TOKEN
#export OS_USERNAME=admin
#export OS_TENANT_NAME==admin
source <(openstack complete)
openstack complete |sudo tee /etc/bash_completion.d/openstack.bash_completion > /dev/null
openstack network create bio-net
openstack subnet create --subnet-range 10.10.122.0/24 --dhcp --dhcp --network bio-net --allocation-pool start=10.10.122.50,end=10.10.122.100 --dns-nameserver 8.8.8.8 bio-net-subnet
openstack router create bio-rtr
openstack router add subnet bio-rtr bio-net-subnet
openstack router set --external-gateway public bio-rtr

cat << EOF > user-data
#cloud-config
user: ubuntu
password: password
chpasswd: { expire: False }
EOF
# in case you don't have a key
#ssh-keygen -q -N ""
openstack keypair create --public-key ~/.ssh/id_rsa.pub mykey
openstack --debug server create --image cirros-0.6.2-x86_64-disk --flavor m1.tiny --key-name mykey --nic net-id=$(openstack network list | awk '/bio-net/ {print $2}') --user-data ./user-data --config-drive true i1

public_network=$(openstack network show public -f value -c id)
fip=$(openstack floating ip create $public_network -f value -c floating_ip_address)
fix_ip=$(openstack server list |grep 'bio-net=' |awk -F '=' '{print $2}' |awk '{print $1}')
openstack floating ip set $fip --fixed-ip-address $fix_ip --port $(openstack port list --fixed-ip ip-address=$fix_ip -c id -f value)

sg_id=$(openstack security group list --project=$(openstack server show i1 -f value -c project_id) |grep default |awk '{print $2}')
openstack security group rule create --protocol tcp --dst-port 22 $sg_id
openstack security group rule create --protocol icmp $sg_id
ping $fip
openstack console log show i1
ssh cirros@$fip

20240508更新

需要搭建一个local horizon integration ci 测试环境，所以需要用devstack搭建一个openstack环境，但今天遇到了以下几个问题：
1, cp: cannot stat ‘/opt/stack/data/venv/etc/glance/rootwrap.*’: No such file or directory
我最终使用的workaround是’GLOBAL_VENV=False’ - https://bugs.launchpad.net/devstack/+bug/2031410/comments/4

2, HttpException: 503, Unable to create the network. No tenant network is available for allocation.
我最终先运行./unstack.sh再运行./stack.sh解决了 - https://bugs.launchpad.net/devstack/+bug/1904155/comments/8

3, 这次安装的是stable/2023.1，它似乎只能使用Django3.2.12 (默认安装的是3.2.16), 先装了3.2.12后再运行stack.sh
sudo pip uninstall Django -y && pip install Django3.2.12

quqi99
关注
专栏目录
devstack 搭建OpenStack环境
x18845040365的博客
11-29 377
开发环境：Ubuntn 16.04.5 1修改阿里源 先备份 命令：sudo cp /etc/apt/sources.list /etc/apt/sources.list.copy 命令：sudo vi /etc/apt/sources.list 添加一下内容： deb http://mirrors.aliyun.com/ubuntu/ xenial main restricted unive...
study octavia workflow (by quqi99)
最新发布
技术并艺术着
08-21 881
这个bug 2043808说的是当在unordered_flow时，它下面的subtask (JobTask)出错时，它的retry (JobRetry)类中返回retry.REVERT_ALL (应revert整个flow, 即main-flow及其子任务）时被其他subtask返回的retry.REVERT（仅revert触发它的subflow, 即subflow1-jobtask或subflow2-jobtask)给覆盖了. 如上面例子的流程是：例如，
使用Devstack进行开发
weixin_34344677的博客
07-03 209
翻译自 devstack 官方文档 Developing with Devstack 检查服务 默认情况下，DevStack 的大多数服务都以 devstack@$servicename.service 的名字作为 system unit 运行。关于 system unit 可以参考Understanding system units a...
使用 pip 换源的指南
pumpkin84514的博客
06-30 3898
以上是 pip 换源的几种常用方法，涵盖了命令行临时换源、全局配置换源、通过环境变量换源以及在开发工具中配置换源的方法。根据实际需求和环境选择合适的方法来加速包的下载和安装。
Devstack搭建基础开发环境
bai0324lin的专栏
09-11 1940
经验借鉴 一、环境准备 1. 准备：       两台虚拟机，第一台（hostname:devstack）充当控制、网络、计算节点（allinone），第二台（hostname:devstack-com）充当计算节点。内存为8G、磁盘100G全部挂在/下，便于开发。安装Centos Linux release 7.2.1511，选择最简安装。OpenStack版本将对齐社区Master分支，...
开发用的devstack(by quqi99)
huof的专栏
01-27 767
作者：张华  发表于：2014-05-15 版权声明：可以任意转载，转载时请务必以超链接形式标明文章原始出处和作者信息及本版权声明 (http://blog.csdn.net/quqi99 ) 环境变量： export OS_USERNAME=admin export OS_PASSWORD=password export OS_TENANT_NAME=demo export O
play with serverless services (by quqi99)
技术并艺术着
04-17 684
云计算里的函数即服务是什么，我们学习一下如何将一个serverless services部署到云端(cloudflare workers)
integrate k8s with keystone (by quqi99)
技术并艺术着
10-29 638
作者：张华 发表于：2020-09-21 版权声明：可以任意转载，转载时请务必以超链接形式标明文章原始出处和作者信息及本版权声明 Install k8s with keystone (http) juju add-model k8s stsstack/stsstack juju deploy kubernetes-core wget https://raw.githubusercontent.com/juju-solutions/kubernetes-docs/master/assets/keystone.
my cloud test bed (by quqi99)
技术并艺术着
03-10 1561
若容器里如果上不了网，如无法访问api.snapcraft.io，是因为lxd容易默认使用了eth1上的dns=10.10.10.1，下面的配置可让eth0, eth1, eth2都默认使用dns=192.168.99.1来避免特色网络对api.snapcraft.io的污染。下列netplan配置创建了br-eth0，也让br-eth0支持wol通过魔术包唤醒，也创建了一个没有dhcp的br-maas用于maas实验。
try anbox or waydroid (by quqi99)
技术并艺术着
06-28 1283
try anbox but failed, this is the note:
如何使用Devstack进行OpenStack开发
06-08
本课程主要帮助，想成为一名OpenStack Developer却又不知道从何开始的同学们，掌握使用Devstack搭建OpenStack，以及相应的调试方法、IDE介绍等。
cirros-0.5.1-aarch64-disk.img openstack官方16m镜像最新版直接下载
03-27
不用等待官方的卡顿了，给各位学习openstack的朋友提供一个快速创建实例的机会，大家共同进步吧，无积分直接下载，搬运而已
使用devstack构建openstack的重启脚本
10-22
针对使用devstack构建的openstack，重新启动系统后，使用此脚本恢复openstack环境。 For openstack built by devstack, this shell script can be run to recover the openstack environment.
DevStack云计算平台快速搭建
mgsky1的博客
09-10 976
这学期学校开了一门课，叫云计算平台搭建，用的书是OpenStack，这个是现在比较流行的一个云计算平台，官方对这个平台的解释是OpenStack is a cloud operating system that controls large pools of compute, storage, and networking resources throughout a datacenter, al
Devstack-Queens版本日志查看
u013184378的博客
12-13 1295
日志文件查看 以nova组件为例进行说明。 在老版本的devstack下，日志记录再/opt/stack/log下，或者非devstack的openstack下使用/var/log/nova/xxx.log文件进行查看，但是新版本的devstack使用journalctl来代替日志的查看。 #journalctl -f --unit devstack@n-cpu.service //相当于之前的...
血的教训：如何使用DevStack部署OpenStack开发环境
jimoduwu的专栏
06-14 589
断断续续的搞了3个星期了，按照官方的文档各种跑；按照国内的技术博客的方法各种跑；都是各种报错，VMware虚拟机和VirtualBox虚拟机的Ubuntu系统都各安装了不下10次，真的要被DevStack折腾的怀疑人生了。搞了十多年的软件开发，没遇到过连个开发环境搭建就这么难搞的。官方发布一个可以直接Debug的VMware或者VirtualBox虚拟机的镜像不可以吗？ 说来也奇怪，写完回答后的几个小时内尝试安装竟然成功了。将VMware镜像文件共享给大家: 链接：https://pan.baidu.c
Debian Customer PPA RFC (by quqi99)
技术并艺术着
01-13 3792
作者：张华 发表于：2016-01-13 版权声明：可以任意转载，转载时请务必以超链接形式标明文章原始出处和作者信息及本版权声明 ( http://blog.csdn.net/quqi99 ) Precondition a,sudo apt install gnupg pbuilder ubuntu-dev-tools bzr-builddeb apt-file debhelper............
https://quqi.gblhgk.com/s/39330/2YCkzWmOod3LV16C
08-12
根据提供的引用信息，您分享了一些关于处理HTTPS请求的源代码和调试信息。根据引用和引用中提到的内容，您可能在调试过程中遇到了一些问题，尝试更改了一些命令和参数。引用是haproxy相关处理的源代码，该代码用于检查SSL版本并确定响应是否有效。 然而，您的问题是关于"https://quqi.gblhgk.com/s/39330/2YCkzWmOod3LV16C"这个链接的。根据提供的引用内容，我无法找到与该特定链接相关的信息。如果您有关于该链接的具体问题，请提供更多信息，以便我能够为您提供更准确的答案。<span class="em">1</span><span class="em">2</span><span class="em">3</span> #### 引用[.reference_title] - *1* *2* *3* [Use Octavia to Implement HTTPS Health Monitors (by quqi99)](https://blog.csdn.net/quqi99/article/details/85469161)[target="_blank" data-report-click={"spm":"1018.2226.3001.9630","extra":{"utm_source":"vip_chatgpt_common_search_pc_result","utm_medium":"distribute.pc_search_result.none-task-cask-2~all~insert_cask~default-1-null.142^v92^chatsearchT3_1"}}] [.reference_item style="max-width: 100%"] [ .reference_list ]
评论 1
添加红包

请填写红包祝福语或标题

个

红包个数最小为10个

元

红包金额最低5元

当前余额3.43元 前往充值 >
需支付：10.00元
成就一亿技术人!
领取后你会自动成为博主和红包主的粉丝 规则
hope_wisdom
发出的红包

打赏作者

quqi99

你的鼓励就是我创造的最大动力

¥1 ¥2 ¥4 ¥6 ¥10 ¥20
扫码支付：¥1
获取中
扫码支付

您的余额不足，请更换扫码支付或充值

打赏作者

实付元
使用余额支付
点击重新获取
扫码支付
钱包余额 0

抵扣说明：

1.余额是钱包充值的虚拟货币，按照1:1的比例进行支付金额的抵扣。
2.余额无法直接购买下载，可以购买VIP、付费专栏及课程。

余额充值