未分类

解决在Mac下iTerm2终端使用sz和rz命令报错问题

我们经常使用 sz/rz 命令进行文件的上传下载，非常方便。但是在 Mac 下面就不能直接使用了需要进行配置才能使用昨天在给客户调试相关代码时，需要覆盖一些代码，使用 rz 进行上传时却报错了： rz waiting to receive.**B0100000023be50 使用 sz 下载也是报错：**B00000000000000 并且都会卡死终端一段时间解决方案解决的方案有点复杂，一点一点来看安装 lrzsz 首先需要我们安装一下 lrzsz，使用命令进行安装：

<span style="color: #000000;">brew install lrzsz</span>

1	<span style="color: #000000;">brew install lrzsz</span>

配置 iTerm2 安装完成后我们需要在 iTerm2 中使用的话，还需要一些配置进入到 /usr/local/bin 目录下，下载两个脚本文件

<span style="color: #000000;">cd /usr/local/bin 
sudo wget https://gist.githubusercontent.com/sy-records/1b3010b566af42f57fa6fa38138dd22a/raw/2bfe590665d3b0e6c8223623922474361058920c/iterm2-send-zmodem.sh 
sudo wget https://gist.githubusercontent.com/sy-records/40f4ba22e3fbdeedf58463b067798962/raw/b32d2f7ac3fa54acca81be3664797cebb724690f/iterm2-recv-zmodem.sh
sudo chmod 777 /usr/local/bin/iterm2-*</span>

1

2

3

4

<span style="color: #000000;">cd /usr/local/bin

sudo wget https://gist.githubusercontent.com/sy-records/1b3010b566af42f57fa6fa38138dd22a/raw/2bfe590665d3b0e6c8223623922474361058920c/iterm2-send-zmodem.sh

sudo wget https://gist.githubusercontent.com/sy-records/40f4ba22e3fbdeedf58463b067798962/raw/b32d2f7ac3fa54acca81be3664797cebb724690f/iterm2-recv-zmodem.sh

sudo chmod 777 /usr/local/bin/iterm2-*</span>

下载好之后我们进行 iTerm2 的配置点击 iTerm2 的设置界面 Perference -> Profiles -> Default -> Advanced -> Triggers 的 Edit Read more…

By sean, 4 years11/21/2020 ago

未分类

CentOS 安装Parallels Tools

CentOS 安装Parallels Tools 为了做到Mac和Linux之间共享文件夹，因此需要安装Parallels Tool，具体安装步骤可以参考 install parallels tool for linux guest http://download.parallels.com/desktop/v4/docs/en/Parallels_Desktop_Users_Guide/22570.htm 具体步骤如下：在Virtual Machine上点击Install Parallels Tools image 在Linux上我们会看不到/media/cdrom的内容，因此，需要手动地mkdir -p /media/cdrom，然后再mount上去，mount的命令是mount -o exec /dev/cdrom /media/cdrom。我们在/media/cdrom下看到install安装脚本了，运行su /media/cdrom/install就可以了，安装完按照提示重启centos后，就可以在/media/psf下看到对应的共享文件了，自己可以设置是可读还是可读写来着，默认是可写的。参考： http://blog.hadoopspark.com/2013/12/centos-setup-under-parallels-desktop/ http://blog.csdn.net/mmical/article/details/39972573

By sean, 4 years ago

未分类

TypeError: argument of type ‘PosixPath’ is not iterable解决办法

Django执行python manage.py migrate时报如下错误： File “/root/.python_env/.virtualenvs/venv/lib/python3.9/site-packages/django/db/backends/sqlite3/creation.py”, line 12, in is_in_memory_db return database_name == ‘:memory:’ or ‘mode=memory’ in database_name TypeError: argument of type ‘PosixPath’ is not iterable 这个问题应该是django2.0版本和python新版本不兼容导致的（我安装的是Django 2.2.13， Python 3.9.0） django2.0中，django/db/backends/sqlite3/creation.py的代码如下而最新的django3.1此处的代码为：如果不想升级django的话可以参照新版django修改即可。不过要在文件前面导入Path，即： from pathlib import Path 所以改动如下：vim /root/.python_env/.virtualenvs/venv/lib/python3.9/site-packages/django/db/backends/sqlite3/creation.py

from pathlib import Path

class DatabaseCreation(BaseDatabaseCreation):

    @staticmethod
    def is_in_memory_db(database_name):
        #return database_name == ':memory:' or 'mode=memory' in database_name
        # ref: https://blog.csdn.net/jason_cdd/article/details/108609917
        return not isinstance(database_name, Path) and (
            database_name == ':memory:' or 'mode=memory' in database_name
        )

1

2

3

4

5

6

7

8

9

10

11

from pathlib import Path

class DatabaseCreation(BaseDatabaseCreation):

@staticmethod

def is_in_memory_db(database_name):

#return database_name == ':memory:' or 'mode=memory' in database_name

# ref: https://blog.csdn.net/jason_cdd/article/details/108609917

return not isinstance(database_name, Path) and (

database_name == ':memory:' or 'mode=memory' in database_name

)

By sean, 4 years11/17/2020 ago

未分类

解决Django:SQLite 3.8.3 or later is required

在我的CentOS系统上安装了django==2.2.8并创建了一个webApps项目，使用：

$ django-admin startproject webApps

1	$ django-admin startproject webApps

但是，当我使用以下命令初始化迁移所需模型时，发生了错误：

$ python3 manage.py migrate

1	$ python3 manage.py migrate

以上命令产生了如下的错误输出： django.core.exceptions.ImproperlyConfigured: SQLite 3.8.3 or later is required (found 3.7.17). django发现Python使用的sqlite版本过低，不满足要求。因此，我尝试升级自带的sqlite。使用sqlite3 –version查看了CentOS的Sqlite为3.7，我开始用yum remove sqlite移除了当前版本，并且编译安装高版本，参考比如：

$ wget https://www.sqlite.org/2019/sqlite-autoconf-3270200.tar.gz
$ tar -zxvf sqlite-autoconf-3270200.tar.gz
$ cd sqlite-autoconf-3270200
$ ./configure --prefix=/usr/local
$ make && make install

1

2

3

4

5

$ wget https://www.sqlite.org/2019/sqlite-autoconf-3270200.tar.gz

$ tar -zxvf sqlite-autoconf-3270200.tar.gz

$ cd sqlite-autoconf-3270200

$ ./configure --prefix=/usr/local

$ make && make install

此时完成了sqlite 3.27.2的安装，但是系统默认认识3.7。因此链接新的路径：

$ mv /usr/bin/sqlite3  /usr/bin/sqlite3_7
$ ln -s /usr/local/bin/sqlite3   /usr/bin/sqlite3

1 2	$ mv /usr/bin/sqlite3 /usr/bin/sqlite3_7 $ ln -s /usr/local/bin/sqlite3 /usr/bin/sqlite3

设置共享库路径：export LD_LIBRARY_PATH=”/usr/local/lib”:LD_LIBRARY_PATH并执行生效source ~/.bashrc 这个时候，使用sqlite3 –version时便会输出版本为3.27.2。可是当我再次执行python3 manage.py migrate时仍会报错，原来python使用的sqlite还是3.7：

>>>import sqlite3                                                     
>>>sqlite3.sqlite_version
'3.7.17'

1

2

3

>>>import sqlite3

>>>sqlite3.sqlite_version

'3.7.17'

这个时候我就二丈摸不着头脑了，google了一下，解决方式包括未正确设定sqlite、尝试升级python版本、重新编译python等，重新编译太麻烦了吧，于是我发现了一个可以有效解决当前django必须使用sqlite3.8.3以上版本的问题。感谢，可以这样尝试： locate django将会输出大量包含django关键字的目录文件，可以看到django安装在哪些路径下，在我的系统上，需要找到/…/lib/python3.6/site-packages/django/db/backends/sqlite3/base.py这个脚本，django判断当前使用的sqlite版本的代码就在这里，找到以下代码块，注释掉那一行代码并更改：

def check_sqlite_version():
    # if Database.sqlite_version_info < (3, 8, 3):
    if Database.sqlite_version_info < (3, 6, 3):
        raise ImproperlyConfigured('SQLite 3.8.3 or later is required(found %s).' % Database.sqlite_version)


check_sqlite_version()

1

2

3

4

5

6

7

def check_sqlite_version():

# if Database.sqlite_version_info < (3, 8, 3):

if Database.sqlite_version_info < (3, 6, 3):

raise ImproperlyConfigured('SQLite 3.8.3 or later is required(found %s).' % Database.sqlite_version)

check_sqlite_version()

再次尝试django-admin startproject webApps便会执行成功 From: https://cloud.tencent.com/developer/article/1639633

By sean, 4 years11/10/2020 ago

未分类

build a docker-compose binary for aarch64/arm64

在uos(基于Debian)操作系统上, docker-compose也需要重新编译，过程如下：首先安装docker 获取软件包获取“Docker Compose-1.22.0”安装包。

cd /usr/local/src
git clone https://github.com/ubiquiti/docker-compose-aarch64.git

1 2	cd /usr/local/src git clone https://github.com/ubiquiti/docker-compose-aarch64.git

安装进入docker-compose源文件目录。cd /usr/local/src/docker-compose-aarch64 配置Dockerfile。vi Dockerfile更改如下：由于是直接在uos机器上编译，所以不需要进行交叉编译，注释掉RUN [ “cross-build-start” ] pip install时老是timeout，在此增加timeout时间, 配置清华源完整Dockerfile如下：

# Dockerfile to build docker-compose for aarch64
FROM arm64v8/python:3.6.5-stretch

# Add env
ENV LANG C.UTF-8

# Enable cross-build for aarch64
COPY ./vendor/qemu-bin /usr/bin/
# RUN [ "cross-build-start" ]

# Set the versions
ENV DOCKER_COMPOSE_VER 1.22.0
# docker-compose requires pyinstaller 3.3.1 (check github.com/docker/compose/requirements-build.txt)
# If this changes, you may need to modify the version of "six" below
ENV PYINSTALLER_VER 3.3.1
# "six" is needed for PyInstaller. v1.11.0 is the latest as of PyInstaller 3.3.1
ENV SIX_VER 1.11.0

# Install dependencies
# RUN apt-get update && apt-get install -y
RUN pip install six==$SIX_VER

# Compile the pyinstaller "bootloader"
# https://pyinstaller.readthedocs.io/en/stable/bootloader-building.html
WORKDIR /build/pyinstallerbootloader
RUN curl -fsSL https://github.com/pyinstaller/pyinstaller/releases/download/v$PYINSTALLER_VER/PyInstaller-$PYINSTALLER_VER.tar.gz | tar xvz >/dev/null \
    && cd PyInstaller*/bootloader \
    && python3 ./waf all

# Clone docker-compose
WORKDIR /build/dockercompose
RUN git clone https://github.com/docker/compose.git . \
    && git checkout $DOCKER_COMPOSE_VER

# Run the build steps (taken from github.com/docker/compose/script/build/linux-entrypoint)
RUN mkdir ./dist \
    && pip install --default-timeout=1000 --no-cache-dir -q -i https://pypi.tuna.tsinghua.edu.cn/simple -r requirements.txt -r requirements-build.txt \
    && ./script/build/write-git-sha \
    && pyinstaller docker-compose.spec \
    && mv dist/docker-compose ./docker-compose-$(uname -s)-$(uname -m)

# Disable cross-build for aarch64
# Note: don't disable this, since we want to run this container on x86_64, not aarch64
# RUN [ "cross-build-end" ]

# Copy out the generated binary
VOLUME /dist
CMD cp docker-compose-* /dist

1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

20

21

22

23

24

25

26

27

28

29

30

31

32

33

34

35

36

37

38

39

40

41

42

43

44

45

46

47

48

# Dockerfile to build docker-compose for aarch64

FROM arm64v8/python:3.6.5-stretch

# Add env

ENV LANG C.UTF-8

# Enable cross-build for aarch64

COPY ./vendor/qemu-bin /usr/bin/

# RUN [ "cross-build-start" ]

# Set the versions

ENV DOCKER_COMPOSE_VER 1.22.0

# docker-compose requires pyinstaller 3.3.1 (check github.com/docker/compose/requirements-build.txt)

# If this changes, you may need to modify the version of "six" below

ENV PYINSTALLER_VER 3.3.1

# "six" is needed for PyInstaller. v1.11.0 is the latest as of PyInstaller 3.3.1

ENV SIX_VER 1.11.0

# Install dependencies

# RUN apt-get update && apt-get install -y

RUN pip install six==$SIX_VER

# Compile the pyinstaller "bootloader"

# https://pyinstaller.readthedocs.io/en/stable/bootloader-building.html

WORKDIR /build/pyinstallerbootloader

RUN curl -fsSL https://github.com/pyinstaller/pyinstaller/releases/download/v$PYINSTALLER_VER/PyInstaller-$PYINSTALLER_VER.tar.gz | tar xvz >/dev/null \

&& cd PyInstaller*/bootloader \

&& python3 ./waf all

# Clone docker-compose

WORKDIR /build/dockercompose

RUN git clone https://github.com/docker/compose.git . \

&& git checkout $DOCKER_COMPOSE_VER

# Run the build steps (taken from github.com/docker/compose/script/build/linux-entrypoint)

RUN mkdir ./dist \

&& pip install --default-timeout=1000 --no-cache-dir -q -i https://pypi.tuna.tsinghua.edu.cn/simple -r requirements.txt -r requirements-build.txt \

&& ./script/build/write-git-sha \

&& pyinstaller docker-compose.spec \

&& mv dist/docker-compose ./docker-compose-$(uname -s)-$(uname -m)

# Disable cross-build for aarch64

# Note: don't disable this, since we want to run this container on x86_64, not aarch64

# RUN [ "cross-build-end" ]

# Copy out the generated binary

VOLUME /dist

CMD cp docker-compose-* /dist

安装docker-compose

cd /usr/local/src/docker-compose-aarch64
docker build . -t docker-compose-aarch64-builder

1 2	cd /usr/local/src/docker-compose-aarch64 docker build . -t docker-compose-aarch64-builder

运行和验证运行docker-compose容器

(venv) root@uos-PC:/data/nfsdata/docker-compose-aarch64# docker run --rm -it docker-compose-aarch64-builder bash
root@6b31862b8797:/build/dockercompose# find / -name "docker-compose-Linux-aarch64"
/build/dockercompose/docker-compose-Linux-aarch64
root@6b31862b8797:/build/dockercompose# cd /build/dockercompose
root@6b31862b8797:/build/dockercompose# ./docker-compose-Linux-aarch64 --version
docker-compose version 1.22.0, build e20d808e

1

2

3

4

5

6

(venv) root@uos-PC:/data/nfsdata/docker-compose-aarch64# docker run --rm -it docker-compose-aarch64-builder bash

root@6b31862b8797:/build/dockercompose# find / -name "docker-compose-Linux-aarch64"

/build/dockercompose/docker-compose-Linux-aarch64

root@6b31862b8797:/build/dockercompose# cd /build/dockercompose

root@6b31862b8797:/build/dockercompose# ./docker-compose-Linux-aarch64 --version

docker-compose version 1.22.0, build e20d808e

拷贝执行文件：

docker cp e4b58853c83d:/build/dockercompose/docker-compose-Linux-aarch64 .

1	docker cp e4b58853c83d:/build/dockercompose/docker-compose-Linux-aarch64 .

By sean, 5 years09/11/2020 ago

未分类

搭建rsync服务器同步数据

最近在弄国产化操作系统适配，由于别人公司不让用VPN, 只能通过一个公网映射端口做ssh登录，而安装代码又是放在内网的其它服务器上，代码调试很不方便，所以想到用rsync来从公网服务器上自动同步代码到内网服务器上首先搭建rsync服务：安装rsync：

apt-get install rsync

1	apt-get install rsync

在source服务器上运行命令（可以同时同步到两台内网服务器上）

nohup inotifywait -mqr -e modify --timefmt '%Y-%m-%d %H:%M:%S' \
  --format '%w %f %e %T' /data/iuap_installer | while read file; do \
    for i in 192.168.1.3 192.168.1.98; do \
      rsync -avzP -e ssh /data/iuap_installer/ root@$i:/data/iuap_installer/ \
      >/dev/null 2>&1; \
    done; \
  done &

1

2

3

4

5

6

7

nohup inotifywait -mqr -e modify --timefmt '%Y-%m-%d %H:%M:%S' \

--format '%w %f %e %T' /data/iuap_installer | while read file; do \

for i in 192.168.1.3 192.168.1.98; do \

rsync -avzP -e ssh /data/iuap_installer/ root@$i:/data/iuap_installer/ \

>/dev/null 2>&1; \

done; \

done &

如果用cron来定时同步的话，可以简单地配置ssh免密，来通过ssh同步：

for dir in /armbuild /data; do
    rsync -avzP -e "ssh -i /root/.ssh/id_rsa -p 22" $dir rsync@10.x.x.x:/ali-nas/gPaaS/48.49/
done

1

2

3

for dir in /armbuild /data; do

rsync -avzP -e "ssh -i /root/.ssh/id_rsa -p 22" $dir rsync@10.x.x.x:/ali-nas/gPaaS/48.49/

done

注：如果用rsync模块来同步的话，可以来配置rsync服务器：

cat > /etc/rsyncd.conf << EOF
# /etc/rsyncd: configuration file for rsync daemon mode
# See rsyncd.conf man page for more options.
# 传输文件使用的用户和用户组，如果是从服务器=>客户端，要保证www用户对文件有读取的权限；如果是从客户端=>服务端，要保证www对文件有写权限。
uid = root
gid = root
# 允许chroot，提升安全性，客户端连接模块，首先chroot到模块path参数指定的目录下，chroot为yes时必须使用root权限，且不能备份path路径外的链接文件
use chroot = yes
# 只读
read only = no
# 只写
write only = no
# 设定白名单，可以指定IP段（172.18.50.1/255.255.255.0）,各个Ip段用空格分开
hosts allow = 10.10.0.0/255.255.0.0,192.168.0.0/255.255.0.0
#hosts deny = *
# 允许的客户端最大连接数
max connections = 10
# 欢迎文件的路径，非必须
motd file = /etc/rsync/rsyncd.motd
# pid文件路径
pid file = /var/run/rsyncd.pid
# 记录传输文件日志
transfer logging = yes
# 日志文件格式
log format = %t %a %m %f %b
# 指定日志文件
log file = /var/log/rsync.log
# 剔除某些文件或目录，不同步
exclude = lost+found/
# 设置超时时间
timeout = 900
ignore nonreadable = yes
# 设置不需要压缩的文件
dont compress   = *.gz *.tgz *.zip *.z *.Z *.rpm *.deb *.bz2

# 模块，可以配置多个，使用如: sate@172.18.50.125::125to110
[test]
# 模块的根目录，同步目录，要注意权限
path = /data/iuap_installer
# 是否允许列出模块内容
list = no
# 忽略错误
ignore errors
# 添加注释
comment = ftp export area
# 模块验证的用户名称，可使用空格或者逗号隔开多个用户名
auth users = root
# 模块验证密码文件 可放在全局配置里
secrets file = /etc/rsyncd.secrets
# 剔除某些文件或目录，不同步
exclude = lost+found/ conf/ man/

1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

20

21

22

23

24

25

26

27

28

29

30

31

32

33

34

35

36

37

38

39

40

41

42

43

44

45

46

47

48

49

50

51

cat > /etc/rsyncd.conf << EOF

# /etc/rsyncd: configuration file for rsync daemon mode

# See rsyncd.conf man page for more options.

# 传输文件使用的用户和用户组，如果是从服务器=>客户端，要保证www用户对文件有读取的权限；如果是从客户端=>服务端，要保证www对文件有写权限。

uid = root

gid = root

# 允许chroot，提升安全性，客户端连接模块，首先chroot到模块path参数指定的目录下，chroot为yes时必须使用root权限，且不能备份path路径外的链接文件

use chroot = yes

# 只读

read only = no

# 只写

write only = no

# 设定白名单，可以指定IP段（172.18.50.1/255.255.255.0）,各个Ip段用空格分开

hosts allow = 10.10.0.0/255.255.0.0,192.168.0.0/255.255.0.0

#hosts deny = *

# 允许的客户端最大连接数

max connections = 10

# 欢迎文件的路径，非必须

motd file = /etc/rsync/rsyncd.motd

# pid文件路径

pid file = /var/run/rsyncd.pid

# 记录传输文件日志

transfer logging = yes

# 日志文件格式

log format = %t %a %m %f %b

# 指定日志文件

log file = /var/log/rsync.log

# 剔除某些文件或目录，不同步

exclude = lost+found/

# 设置超时时间

timeout = 900

ignore nonreadable = yes

# 设置不需要压缩的文件

dont compress = *.gz *.tgz *.zip *.z *.Z *.rpm *.deb *.bz2

# 模块，可以配置多个，使用如: sate@172.18.50.125::125to110

[test]

path = /data/iuap_installer

# 是否允许列出模块内容

list = no

# 忽略错误

ignore errors

# 添加注释

comment = ftp export area

# 模块验证的用户名称，可使用空格或者逗号隔开多个用户名

auth users = root

# 模块验证密码文件可放在全局配置里

secrets file = /etc/rsyncd.secrets

# 剔除某些文件或目录，不同步

exclude = lost+found/ conf/ man/

cat > /etc/rsyncd.secrets <<EOF
mypassword
EOF

1

2

3

cat > /etc/rsyncd.secrets <<EOF

mypassword

EOF

具体使用方法如下：

rsync -avzP --password-file=/etc/rsyncd.secrets   root@192.168.1.3::test

1	rsync -avzP --password-file=/etc/rsyncd.secrets root@192.168.1.3::test

By sean, 5 years09/09/2020 ago

未分类

docker设置代理

docker pull某些镜像时速度特别慢，可以设置代理，设置方式是：编辑/usr/lib/systemd/system/docker.service，加入：

[Service]
Environment="PATH=/usr/local/bin:/bin:/sbin:/usr/bin:/usr/sbin HTTP_PROXY=myip:myport HTTPS_PROXY=myip:myport"

1 2	[Service] Environment="PATH=/usr/local/bin:/bin:/sbin:/usr/bin:/usr/sbin HTTP_PROXY=myip:myport HTTPS_PROXY=myip:myport"

By sean, 5 years07/30/2020 ago

未分类

git checkout 检出命令

检出命令（git checkout）是git最常用的命令之一，同时也是个很危险的命令，因为这条命令会重写工作区：用法一：git checkout [-q] [<commit>] [–] <paths>… 用法二：git checkout [<branch>] 用法三：git checkout [-m] [[-b|–orphan] <new_branch>] [<start_point>] 上面列出的第一种用法和第二种用法的区别在于，第一种用法在命令中包含路径<paths>.为了避免路径和引用（或者提交id）同名而发生冲突，可以在<paths>前面用两个连续的短线（减号作为分割）第一种用法的<commit>是可选项，如果省略则相当于从暂存区进行检出（检出的默认值是暂存区）。第一种用法（包含了路径<paths>的用法）不会改变HEAD头指针，主要是用于指定版本的文件覆盖工作区中对应的文件。如果省略<commit>，则会用暂存区的文件覆盖工作区的文件，否则用指定提交中的文件覆盖暂存区和工作区中对应的文件. 第二种用法则会改变HEAD头指针，主要用作切换到分支，如果省略<branch>则相当于对工作区进行状态检查。第三种用法主要是创建和切换到新的分支(<new_branch>),新的分支从<start_point>指定的提交开始创建。新的分支和master分支没有什么实质的不同，都是在refs/heads命名空间下的引用。具体示例： git checkout branch 检出branch分支，要完成上图的三个步骤，更新HEAD以指向branch分支，以及用branch指向的树更新暂存区和工作区 git checkout 汇总显示工作区，暂存区与HEAD的差异 git checkout HEAD 同上 git checkout — filename 用暂存区中filename文件来覆盖工作区中的filename文件。相当于git add filename的撤消，这个命令很危险，因为对于本地的修改会悄无声息的覆盖 git checkout — . 或写作 git checkout . Read more…

By sean, 5 years07/29/2020 ago

未分类

python中的subprocess.Popen()使用

python中的subprocess.Popen()使用从python2.4版本开始,可以用subprocess这个模块来产生子进程,并连接到子进程的标准输入/输出/错误中去，还可以得到子进程的返回值。 subprocess意在替代其他几个老的模块或者函数，比如：os.system os.spawn* os.popen* popen2.* commands.* 一、subprocess.Popen subprocess模块定义了一个类： Popen class subprocess.Popen( args, bufsize=0, executable=None, stdin=None, stdout=None, stderr=None, preexec_fn=None, close_fds=False, shell=False, cwd=None, env=None, universal_newlines=False, startupinfo=None, creationflags=0) 各参数含义如下： args: args参数。可以是一个字符串，可以是一个包含程序参数的列表。要执行的程序一般就是这个列表的第一项，或者是字符串本身。 subprocess.Popen([“cat”,”test.txt”]) subprocess.Popen(“cat test.txt”) 这两个之中，后者将不会工作。因为如果是一个字符串的话，必须是程序的路径才可以。(考虑unix的api函数exec，接受的是字符串列表) 但是下面的可以工作 subprocess.Popen(“cat test.txt”, shell=True) 这是因为它相当于 subprocess.Popen([“/bin/sh”, “-c”, “cat test.txt”]) 在*nix下，当shell=False（默认）时，Popen使用os.execvp()来执行子程序。args一般要是一个【列表】。如果args是个字符串的话，会被当做是可执行文件的路径，这样就不能传入任何参数了。注意： shlex.split()可以被用于序列化复杂的命令参数，比如： >>> shlex.split(‘ls ps Read more…

By sean, 5 years ago

未分类

docker和harbor启动失败解决

今天启动harbor时启动失败，有几个问题，记录一下：重启docker，报错： msg=”Error starting daemon: layer does not exist” 解决：只能是清空docker数据目录了，实际上官方还提供了一个略微安全的删除脚本：

#!/bin/sh
set -e

dir="$1"

if [ -z "$dir" ]; then
	{
		echo 'This script is for destroying old /var/lib/docker directories more safely than'
		echo '  "rm -rf", which can cause data loss or other serious issues.'
		echo
		echo "usage: $0 directory"
		echo "   ie: $0 /var/lib/docker"
	} >&2
	exit 1
fi

if [ "$(id -u)" != 0 ]; then
	echo >&2 "error: $0 must be run as root"
	exit 1
fi

if [ ! -d "$dir" ]; then
	echo >&2 "error: $dir is not a directory"
	exit 1
fi

dir="$(readlink -f "$dir")"

echo
echo "Nuking $dir ..."
echo '  (if this is wrong, press Ctrl+C NOW!)'
echo

( set -x; sleep 10 )
echo

dir_in_dir() {
	inner="$1"
	outer="$2"
	[ "${inner#$outer}" != "$inner" ]
}

# let's start by unmounting any submounts in $dir
#   (like -v /home:... for example - DON'T DELETE MY HOME DIRECTORY BRU!)
for mount in $(awk '{ print $5 }' /proc/self/mountinfo); do
	mount="$(readlink -f "$mount" || true)"
	if dir_in_dir "$mount" "$dir"; then
		( set -x; umount -f "$mount" )
	fi
done

# now, let's go destroy individual btrfs subvolumes, if any exist
if command -v btrfs > /dev/null 2>&1; then
	root="$(df "$dir" | awk 'NR>1 { print $NF }')"
	root="${root#/}" # if root is "/", we want it to become ""
	for subvol in $(btrfs subvolume list -o "$root/" 2>/dev/null | awk -F' path ' '{ print $2 }' | sort -r); do
		subvolDir="$root/$subvol"
		if dir_in_dir "$subvolDir" "$dir"; then
			( set -x; btrfs subvolume delete "$subvolDir" )
		fi
	done
fi

# finally, DESTROY ALL THINGS
( set -x; rm -rf "$dir" )

1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

20

21

22

23

24

25

26

27

28

29

30

31

32

33

34

35

36

37

38

39

40

41

42

43

44

45

46

47

48

49

50

51

52

53

54

55

56

57

58

59

60

61

62

63

64

65

#!/bin/sh

set -e

dir="$1"

if [ -z "$dir" ]; then

{

echo 'This script is for destroying old /var/lib/docker directories more safely than'

echo ' "rm -rf", which can cause data loss or other serious issues.'

echo

echo "usage: $0 directory"

echo " ie: $0 /var/lib/docker"

} >&2

exit 1

fi

if [ "$(id -u)" != 0 ]; then

echo >&2 "error: $0 must be run as root"

exit 1

fi

if [ ! -d "$dir" ]; then

echo >&2 "error: $dir is not a directory"

exit 1

fi

dir="$(readlink -f "$dir")"

echo

echo "Nuking $dir ..."

echo ' (if this is wrong, press Ctrl+C NOW!)'

echo

( set -x; sleep 10 )

echo

dir_in_dir() {

inner="$1"

outer="$2"

[ "${inner#$outer}" != "$inner" ]

}

# let's start by unmounting any submounts in $dir

# (like -v /home:... for example - DON'T DELETE MY HOME DIRECTORY BRU!)

for mount in $(awk '{ print $5 }' /proc/self/mountinfo); do

mount="$(readlink -f "$mount" || true)"

if dir_in_dir "$mount" "$dir"; then

( set -x; umount -f "$mount" )

fi

done

# now, let's go destroy individual btrfs subvolumes, if any exist

if command -v btrfs > /dev/null 2>&1; then

root="$(df "$dir" | awk 'NR>1 { print $NF }')"

root="${root#/}" # if root is "/", we want it to become ""

for subvol in $(btrfs subvolume list -o "$root/" 2>/dev/null | awk -F' path ' '{ print $2 }' | sort -r); do

subvolDir="$root/$subvol"

if dir_in_dir "$subvolDir" "$dir"; then

( set -x; btrfs subvolume delete "$subvolDir" )

fi

done

fi

# finally, DESTROY ALL THINGS

( set -x; rm -rf "$dir" )

将该脚本保存到本地后运行sh fix.sh /data/docker，清空/data/docker目录，重启docker服务 docker-compose启动harbor， harbor-log启动失败 Changing password for root. sudo: unable to change expired password: Authentication token manipulation error sudo: Account or password is expired, reset your password and try again 解决：参考：https://kb.vmware.com/s/article/79497 重新拉取镜像：docker pull goharbor/harbor-log:v1.10.3 修改配置文件，重启成功 Read more…

By sean, 5 years06/23/2020 ago