linuxops

[linuxops]

aws ec2 频繁删除我的ssh公钥

aws 的 ec2 今天好几次ssh掉线， .ssh/authorized_keys 发生变化，把自己添加进去的key删除了

[linuxops]

1.1.1.1 dns did not work

今天发现上午很卡，是因为 /etc/resolv.conf 中配置的 primary dns 1.1.1.1 不能用，虽然我ping的时候响应时间很短，比其它国内的public dns 要短（所以我才选用它的），估计ping的响应是伪造的，如果1.1.1.1的服务器在国内，dns查询请求不会被 gfw 拦截的

[linuxops]

the PATH env was not hard coded in cron

I said it was hard coded before, I do not know why I said that, maybe now the behavior has changed

[linuxops]

apt repo url 中拼写错误导致 playbook 执行失败

'deb [arch=amd64] https://download.docker.com/linux/debbian buster stable'

debbian 应为 debian

[linuxops]

libssl_conf.so: cannot open shared object file: No such file or directory

There was an error when attempting to rsync a synced folder. Please inspect the error message below for more info. Host path: /home/hi/kiki/ Guest path: /vagrant Command: "rsync" "--verbose" "--archive" "--delete" "-z" "--copy-links" "--no-owner" "--no-group" "--rsync-path" "sudo rsync" "-e" "ssh -p 2222 -o LogLevel=FATAL -o ControlMaster=auto -o ControlPath=/tmp/vagrant-rsync-20200915-14497-90j7kg -o ControlPersist=10m -o IdentitiesOnly=yes -o StrictHostKeyChecking=no -o UserKnownHostsFile=/dev/null -i '/home/hi/kiki/.vagrant/machines/default/virtualbox/private_key'" "--exclude" ".vagrant/" "/home/hi/kiki/" "vagrant@127.0.0.1:/vagrant" Error: Auto configuration failed 139650847340248:error:25066067:DSO support routines:DLFCN_LOAD:could not load the shared library:dso_dlfcn.c:185:filename(libssl_conf.so): libssl_conf.so: cannot open shared object file: No such file or directory 139650847340248:error:25070067:DSO support routines:DSO_load:could not load the shared library:dso_lib.c:244: 139650847340248:error:0E07506E:configuration file routines:MODULE_LOAD_DSO:error loading dso:conf_mod.c:285:module=ssl_conf, path=ssl_conf 139650847340248:error:0E076071:configuration file routines:MODULE_RUN:unknown module name:conf_mod.c:222:module=ssl_conf rsync: connection unexpectedly closed (0 bytes received so far) [sender] rsync error: error in rsync protocol data stream (code 12) at io.c(235) [sender=3.1.3]

solution

export OPENSSL_CONF=/etc/ssl/

[linuxops]

卸载亿赛通客户端

公司的有一台windows7电脑安装了亿赛通客户端，我在管理控制台设置卸载了客户端，按理客户端会冒泡通知可以卸载，然后就可以成功卸载了（我在2台电脑分别为 windows 7 / windows 10 的系统上按照这样的办法成功地卸载了该客户端），谁知在卸载程序无法停掉正在运行的亿赛通客户端进程，也没有办法通过其它途径停掉该进程，尝试过好多办法都没能解决，最后想到通过 unlocker 把任务管理器中的相关进程的可执行文件删掉，这样卸载的时候就不会因为有进程在运行而无法成功卸载了。删除的过程中有一个主要程序文件删不掉，说可以设置重启后删除，结果重启后也并没有删除，但是我试了下可以用unlocker移动到其它位置（不知道不用unlocker 是否可以移动），然后注销重新登录之后就可以删除了。然后再尝试卸载，卡在卸载码这里了。于是一不做二不休，把整个程序目录干掉了，也不卸载了。反正只要程序不运行，就没有影响了。这个程序似乎会注入到系统进程中，这样做卸载不干净，但是只要以后不要无缘无故加密文件就好了。

[linuxops]

360 主动防御残留

360 安全卫生已经卸载了，主动防御却仍然保留着，试图卸载，却告知360安全卫生被破坏，是否要重新安装, fuck

[linuxops]

apt search

apt search is the front-end of apt-cache search, it support regex by default, to search from only names of the package, pass the option --names-only

apt search --names-only 'python.*pip'

[linuxops]

ansible: "bad symbolic permission for mode:"

- name: create user's directorys file: path: "{{ item.path }}" state: directory mode: "{{ item.mode | default(omit) }}" loop: - { path: "{{ ansible_env.HOME }}/.ssh", mode: 0700 }

ansible complained

"bad symbolic permission for mode: 448"

the reason is that pyyaml convert octal value 0700 to int 448, and then I put the mode's value in double quote, which make 448 be treat as a string, but the module assumes that string values are octals, which cause problems. So whenever you set value for unix permission mode, always quote it as a string

int('0700', 8) gets 448

[linuxops]

ansible import_tasks vs include_tasks

如下 ansible 配置

tasks: - include_tasks: tasks/install_pkgs.yml - include_tasks: tasks/config.yml tags: config

我想只运行 tasks/config.yml 中的 tasks

ansible-playbook -i hosts local.yml --tags config

结果console输出中显示只是执行了 include_tasks: tasks/config.yml, 但是 tasks/config.yml 中的task并没有执行， config 这个 tag 只是 apply to include_tasks: tasks/config.yml, 并没有 apply to tasks/config.yml 中包含的 tasks

using import_tasks instead of include_tasks will fix this problem

[linuxops]

mvn package 时报编码错误和非法字符

因为 svn repo 中的部分代码是加密的，他们开发人员本地有自动解密的配置，这个无法在jenkins上配置

[linuxops]

jenkins: mvn: 未找到命令

jenkins 执行 build 时，总是报错找不到 mvn 命令，原来是因为 The execute shell does not know about the Maven settings; only the maven job type or step does. 可以设置

export MAVEN_HOME=/opt/maven export PATH=$PATH:$MAVEN_HOME/bin

或者调用mvn命令时使用绝对路径

[linuxops]

cannot see what I am typing on terminal

I suddenly cannot see what I am typing on terminal, but it seems I can still run command on it, after some research on the internet, I tried reset, and this fix it

as someone in stackoverflow said

This can happen after a program dies leaving a terminal in an abnormal state. To fix it temporary you "reset" the terminal with:

[linuxops]

reset Jenkins to its factory settings?

edit /var/lib/jenkins/config.xml and change the following line to:

NEW then service jenkins restart

[linuxops]

公司变更ISP后没有修改DHCP中的DNS设置

公司搬迁，使用的宽带也由电信变成联通。最近发现一台服务器使用网络总是会出现问题，后来发现是其使用的DNS的问题，原来这个DNS是电信的，于是想到办公室网络有时候出现的问题也与这个有关，检查出口路由器的DHCP设置，分配个客户端的DNS果然也是这个电信的DNS。

[linuxops]

ssh -X X11 forwarding request failed

ssh server 是 centos 7 Resolution

Confirm xorg-x11-xauth package is installed.

Modify /etc/ssh/sshd_config and set AddressFamily inet and restart sshd.

If SSH client is authenticating with the public key and if there is parameter like no-agent-forwarding, no-port-forwarding, no-X11-forwarding before public key specified in ~/.ssh/authorized_keys of user's home directory on SSH server, these options need to be removed from the file.

Root Cause

By default sshd uses AddressFamily any, this can sometimes cause ssh forwarding to use IPV6 instead of IPV4.