Linux Tips 10

TurnToJPG -->

1. samba mount

mount samba via:

# mount -t cifs // /mnt -o username=uuuu,password=ffff

2. CentOS ansible

On vps we start a docker instance via:

# docker run -it centos:7 /bin/bash
# yum update -y
# yum install -y epel-release
# yum update -y
# yum install -y python-pip
# mkdir ~/ansible && cd ~/ansible
# pip download ansible
# tar czvf ansible.tar.gz ansible

Download the ansible.tar.gz to local, and transfer them into the centos offline machines, install ansible via:

# tar xzvf ansible.tar.gz
# cd ansible
# pip install --no-index --find-links . ansible
# which ansible
# ansible --version
ansible 2.7.10

Also you have to download the jinja2, upgrade to 2.10.1 version:

# pip download jinja2
# pip install --no-index --find-links . jinja2 --upgrade

3. Fast Kubespray

Download the tar.gz, untar it, then modify the Vagrantfile, then rm -f inventory/sample/hosts.ini, then vagrant up you could get all of the packages and images downloaded to your vm.

Be sure to use firewall-less networking, and set the vm’s resolv.conf to your firewall-less dns server.

# rm -f /etc/resolv.conf
# echo "nameserver">/etc/resolv.conf

4. kubeadm git tree state

Modify the file hack/lib/

  if [[ -n ${KUBE_GIT_COMMIT-} ]] || KUBE_GIT_COMMIT=$("${git[@]}" rev-parse "HEAD^{commit}" 2>/dev/null); then
    if [[ -z ${KUBE_GIT_TREE_STATE-} ]]; then
      # Check if the tree is dirty.  default to dirty
      if git_status=$("${git[@]}" status --porcelain 2>/dev/null) && [[ -z ${git_status} ]]; then

5. pandoc template

For generating pdf:

# wget
# mkdir -p ~/.pandoc/templates
# tar xzvf Eisvogel-1.2.2.tar.gz
# cp eisvogenl.tex ~/.pandoc/templates/eisvogel.latex
# cd ~/.pandoc/templates
# wget

But this doesn’t work at all.

6. kubernetes leader election

Refers to:

7. reinstall rpm with dependencies

via following command:

# yum reinstall $(repoquery --requires --recursive --resolve packagename)

8. dmesg warning

After upgrading to newest kernel, I got some error message during dmesg:

[118383.485389] e1000e 0000:00:19.0 enp0s25: Detected Hardware Unit Hang:
                  TDH                  <0>
                  TDT                  <5>
                  next_to_use          <5>
                  next_to_clean        <0>
                  time_stamp           <102176c2b>
                  next_to_watch        <0>
                  jiffies              <1021c8e80>
                  next_to_watch.status <0>
                MAC Status             <80000>
                PHY Status             <7949>
                PHY 1000BASE-T Status  <0>
                PHY Extended Status    <3000>
                PCI Status             <10>

Solution is disable the TCP checksome offloading:

$ sudo ethtool -K enp0s25 tx off rx off

9. Proxmox iotop

For installing iotop on Proxmox, do following:

root@ks:~# cat /etc/issue


Welcome to the Proxmox Virtual Environment. Please use your web browser to 
configure this server - connect to:



root@ks:~# cat /etc/debian_version 

Find the debian version 9.4 is stretch, then we could find the package using google, and download iotop package then uploading to server and install it.

10. bash debugging

Enable debugging for:

# bash -x ./

10. vncviewer disable send key

Via following commands:

$ vncviewer -FullscreenSystemKeys=0

11. kubectl run

Avoid pulling images always, specify following parameter:


12. ssh tunnel

Using a tunnel for forwarding remote ssh port to local via:

alias sshtunnel='ssh -L dash@'

After you activated ssh tunnel, use ssh -p 10022 root@localhost for accessing.

13. ifupdown

When encounting following error in vagrant:

/sbin/ifup 'eth1'

Stdout from the command:

Stderr from the command:

bash: line 4: /sbin/ifdown: No such file or directory
bash: line 20: /sbin/ifup: No such file or directory

Install apt-get install -y ifupdown, you could fix your problem.

14. usb networking issue

After upgrading to 5.1.6 kernel, my asix ethernet card won’t working, install following packages for making it worked.

$ yaourt asix-dkms

15. delegate to issue

When using kubespray you got delegate to issue, do following:

# vagrant up --provider=libvirt

16. usb ethernet issue

via lsusb -t you could view the usb device tree and its speed.

17. view nvidia gpu temperature

via :

nvidia-smi -q -d temperature

18. pip network error

Install pip again via:

# curl | python

19. consola issues

If you use gnome-terminal, then it’s hard to choose yahei consola, so we have to use xfce4-terminal.

20. Use cdrom as repository

Following steps will use ubuntu iso for installation:

sudo mkdir /aptoncd-mountpoint
sudo mount /media/USB/aptoncd.iso ~/aptoncd-mountpoint -oloop
sudo apt-cdrom -d=/aptoncd-mountpoint add

21. apt-get down

download the packages into /var/cache via:

# apt-get -d install xxxxxxxYourPackageName

22. ansible warning

How to resolve this warning.

 [WARNING]: flush_handlers task does not support when conditional

23. build goharbor

with chartmuseum support, do following:

# make package_offline GOBUILDIMAGE=golang:1.9.2 COMPILETAG=compile_golangimage  NOTARYFLAG=true

Make on arm64 architecture:

/media/sda/harbor/harbor-arm64-develop# make package_offline GOBUILDIMAGE=golang:1.9.2 COMPILETAG=compile_golangimage VERSIONTAG=1.7.0-arm64 PKGVERSIONTAG=1.7.0-arm64 CLAIRFLAG=true NOTARYFLAG=true CHARTFLAG=true

24. Get rpi temperature

via following command:

cpu=$(</sys/class/thermal/thermal_zone0/temp);echo "$((cpu/1000)) c"

25. Force dns query using tcp

Add following options into the /etc/resolv.conf:

options use-vc

26. tips on Friday

working progress:

1. python-pip should be installed and docker-compose needed to be compile. 
2. some packages are located in 128, also libssl/libssl-dev have to be added into the repository
3. secure registry server not stable, why? 
4. package dependency problem should be solved. 
5. docker push is ok, now we could push to the registry. 
6. harbor need to be verified. 

27. VncServer

not only listening on localhost, try following:

# vncserver -localhost no

28. delegate_to

Newest ansible version(v2.8.1) has changed the feature, so we have to use following commands:

# ansible-playbook -i xxxx xxxx.cluster.yml

29. aarch64 vagrant issue

Encounter following:

Error while creating domain: Error saving the server: Call to virDomainDefineXML failed: unsupported configuration: ACPI requires UEFI on this architecture

pflash vs rom.

30. Copy only Packages.gz included

via following commands:

for i in `cat /root/Packages  | grep '^Package:' | awk {'print $2'}`
        echo cp $i"_*.deb" /root/pure/ | bash -

        #cp $i_* /root/pure

31. kubespray openssl issue

Change the openssl signature for v3_ext definition.

[ v3_ext ]
root@arm02:/media/md0/Rong1907/roles/etcd# grep -i "v3_ext" ./ -r
./templates/openssl.conf.j2:[ v3_ext ]
./templates/        openssl x509 -req -in member-${host}.csr -CA ca.pem -CAkey ca-key.pem -CAcreateserial -out member-${host}.pem -days {{certificates_duration}} -extensions v3_ext -extfile ${CONFIG} > /dev/null 2>&1
./templates/        openssl x509 -req -in admin-${host}.csr -CA ca.pem -CAkey ca-key.pem -CAcreateserial -out admin-${host}.pem -days {{certificates_duration}} -extensions v3_ext  -extfile ${CONFIG} > /dev/null 2>&1
./templates/        openssl x509 -req -in node-${host}.csr -CA ca.pem -CAkey ca-key.pem -CAcreateserial -out node-${host}.pem -days {{certificates_duration}} -extensions v3_ext  -extfile ${CONFIG} > /dev/null 2>&1

32. run commands in term

In linux via:

# xterm -hold -e 'apropos editor' & 

33. netdata for ubuntu xenial

Using ppa version:

sudo add-apt-repository ppa:sdeziel/ppa
sudo apt-get update
sudo apt-get install -y netdata

34. minikube ssh/password

Default ssh username and password are docker and tcuser.

35. disable ipv6 on ubuntu

Edit the /etc/default/grub, add following definition:


disable ipv6 will cause vagrant working, so I remove this line again, and fall back to ipv4/ipv6 co-exists.

36. rsync specify ssh ports

sync the remote with local directory:

rsync -a -e 'ssh -p 1xxxx' --progress kubespray-2.11.0

37. ansible docker for kubespray

For lacking the netaddr in ansible docker(lexauw/ansible-alpine:v2.7.9), do following steps:

$ sudo docker run -it lexauw/ansible-alpine:v2.7.9 /bin/sh
/ # pip3 install netaddr

$ sudo docker commit d15d9a5910b4 core/ansible:v2.7.9

Now using the core/ansible:v2.7.9 for deployment, you will get through all of the playbooks.

38. videocutter

Install on ubuntu:

sudo add-apt-repository ppa:ozmartian/apps
sudo apt-get update
sudo apt-get install vidcutter

39. Making video


$ ffmpeg -y -i gongjian.mp4 -vf myoutput.mp4

40. wget quite and overwrite

via following parameters:

# wget -q http://xxxxxx -O /opt/bin/xxxxx

41. Enable bbr on Ubuntu18.04

Via following commands:

# echo "net.core.default_qdisc=fq" >> /etc/sysctl.conf
# echo "net.ipv4.tcp_congestion_control=bbr" >> /etc/sysctl.conf
# sysctl -p
net.core.default_qdisc = fq
net.ipv4.tcp_congestion_control = bbr
# sysctl net.ipv4.tcp_available_congestion_control
net.ipv4.tcp_available_congestion_control = reno cubic bbr
# lsmod | grep bbr
tcp_bbr                20480  4
# sudo sysctl net.core.default_qdisc

当输出中 net.core.default_qdisc 为 fq 且 net.ipv4.tcp_available_congestion_control 包含 bbr 即表示内核已启用 BBR 算法。

42. arm64 Installation


# apt-get install -y virt-manager xubuntu-desktop qemu-efi-aarch64 vnc4server chromium-browser
# systemctl start libvirtd
# systemctl enable libvirtd
# vim ~/.vnc/xstartup
xrdb $HOME/.Xresources
startxfce4 &
# vncserver

Thus you could get a vnc based virt-manager runnable environment.

43. pip install via requirements

Download via:

apt-get update -y                                                                                                        
apt-get install -y python-pip python3-pip                                                                                
mkdir pip                                                                                                                
cd pip/                                                                                                                  
vi requirements.txt                                                                                                      
pip install --download=`pwd` -r requirements.txt                                                                         
pip3 install --download=`pwd` ruamel.yaml                                                                                
tar czvf pip.tar.gz pip                                 

Install offlinely via:

pip install --no-index --find-links=`pwd` -r requirements.txt 
pip3 install --no-index --find-links=`pwd` ruamel.yaml

Examine via:

ansible --version

44. download 163 music

via a pip3 installed packages:

pip3 install pymusic-dl
music-dl -u

45. ffmpeg transform to mp4


ffmpeg -i IMG_3125.MOV -vcodec h264 -acodec mp2  video1.mp4

46. Get repository of ubuntu

switch back to official repository mode:

$ cd /etc/apt/
$ sudo wget

47. rdesktop to different port


$ rdesktop

48. promtail debug


# curl -H "Content-Type: application/json" -XPOST -s "http://localhost:3100/api/prom/push" --data-raw '{"streams": [{ "labels": "{foo=\"bar\"}", "entries": [{ "ts": "2019-10-21T08:28:06.801064-04:00", "line": "fizzbuzz" }] }]}'
 # curl "http://localhost:3100/ready"

49. run es in docker

Run following command:

sudo docker run -d --name elasticsearch -p 9200:9200 -p 9300:9300 -e "discovery.type=single-node" -v `pwd`:/usr/share/elasticsearch/data elasticsearch:6.8.4

50. alerta quickstart

Via docker-compose:

version: '2.1'
    image: alerta/alerta-web
      - "8080:8080"
      - db
      - DEBUG=1  # remove this line to turn DEBUG off
      - DATABASE_URL=postgres://postgres:postgres@db:5432/monitoring
      - AUTH_REQUIRED=True
      - ADMIN_USERS=admin,, #default password: alerta
      - PLUGINS=reject,blackout,normalise,enhance
    restart: always
    image: postgres
      - ./pg-data:/var/lib/postgresql/data
      POSTGRES_DB: monitoring
      POSTGRES_USER: postgres
      POSTGRES_PASSWORD: postgres
    restart: always

Then visiti localhost:8080, to get the api key, write the configuration file:

# vim ~/.alerta.conf 
endpoint = http://localhost:8080/api
key = gaoweugowuegouVkJjzSbNwnod3wiRuywdrYfmoyB40GyJmk
[profile production]
endpoint = http://localhost:8080/api
key = gaoweugowuegouVkJjzSbNwnod3wiRuywdrYfmoyB40GyJmk

Then pip install the cli tools:

# sudo pip3 install alerta
# alerta send -r web01 -e HttpError -g Web -s major --attributes region="EU" --environment Production -S fuck

Now in the website you will see following alert available:


Next step I will setup the alerta together with netdata and prometheus.

51. setuid issue

Someone changed the priviledge of /usr/bin/sudo on server, thus every user in sudo group could not switch to root user via sudo bash, the solution is via:

# su root
# chmod u+s /usr/bin/sudo
# chmod 4755 /usr/bin/sudo

52. awesome focus highlight

Via changing the awesome theme color:

# sudo vim /usr/share/awesome/themes/default/theme.lua
--theme.bg_focus      = "#535d6c"
theme.bg_focus      = "#14EEEE"
# echo 'awesome.restart()' | awesome-client

53. minikube items

Notice the metrics-scraper:

NAMESPACE              NAME                                         READY     STATUS    RESTARTS   AGE
kube-system            coredns-5644d7b6d9-lqpv2                     1/1       Running   0          3m
kube-system            coredns-5644d7b6d9-nrbcp                     1/1       Running   0          3m
kube-system            etcd-minikube                                1/1       Running   0          2m
kube-system            kube-addon-manager-minikube                  1/1       Running   0          2m
kube-system            kube-apiserver-minikube                      1/1       Running   0          2m
kube-system            kube-controller-manager-minikube             1/1       Running   0          2m
kube-system            kube-proxy-2fx6d                             1/1       Running   0          3m
kube-system            kube-scheduler-minikube                      1/1       Running   0          2m
kube-system            nginx-ingress-controller-57bf9855c8-b5w68    1/1       Running   0          3m
kube-system            storage-provisioner                          1/1       Running   0          3m
kubernetes-dashboard   dashboard-metrics-scraper-76585494d8-bn2cb   1/1       Running   0          3m
kubernetes-dashboard   kubernetes-dashboard-57f4cb4545-jgpqz        1/1       Running   0          3m

54. boomaga

Boomaga could combine several print jobs into one pdf file.

55. xz with multiple context

Via xz -T4 big.tar, then you will get 400% speed of the single xz file.

56. react.js quickstart

With the npm version 6 we could start using react.js:

$  curl -sL -o
$ npm -v
$ npm init react-app my-app

57. Quickly redsock libvirt

Via following command:

$ sudo sysctl -w net.ipv4.conf.all.route_localnet=1
$ sudo iptables -t nat -A PREROUTING -p tcp -s -j DNAT --to-destination

Then the subnet of your libvirtd will get to internet via redsocks.

58. etchosts issue

kubespray when encountering etchosts issue, by setting from dhcp to static ip and setup the route could solve the issue.

59. 1604 to 1804 kubespray items

Manually resolve the package dependencies:

# apt-get install -y openssh-server update-motd parted build-essential telnet tcpdump python ebtables libgeoip1

60. destroy all running libvirt items

via :

virsh list | sed -n '1,2!p' | head -n -1 | awk {'print $1'}

61. find the modified timestamp

via find options:

find . -printf "%T@ %Tc %p\n" | sort -n

62. CentOS vagrant

vagrant for centos7 added following items:

# visudo
Defaults:vagrant !requiretty

63. helm/chart nfs-client

via following command:

$ helm install nfs-client-provisioner-1.2.8.tgz --set nfs.server= --set nfs.path=/media/sdb/k8snfs --set storageClass.defaultClass=true

64. nfsd issue

When using nfsd, it defaultly reject the connections from NATed vm’s request, cause the port is over 1024, so you have to add following parameters into the nfsd configuration:

# vim /etc/exports
/var/nfs *(rw,sync,no_root_squash,no_subtree_check,insecure)
/media/md0/nfs *(rw,sync,no_root_squash,no_subtree_check,insecure)

insecure let you accept connections >1024, thus your nfs client could connect to nfsd server.

65. git ignore big files

Get the whole size:

$ find . -size +2M  | tr '\n' ' '
$ du -ch xxxx

Add the file list into gitignore file:

$ find . -size +2M | cat >> .gitignore

66. trouble shooting on metrics-server

With k8s 1.13.5, metrics-server could not startup with tls error,

 1 customresource_discovery_controller.go:203] Starting DiscoveryController
I0920 16:35:36.878076       1 log.go:172] http: TLS handshake error from EOF
I0920 16:35:36.878360       1 log.go:172] http: TLS handshake error from EOF

Copy the right one from kubespray 2.11.0, the newer configuration for metrics-server work properly.


# kubectl get apiservice | grep metrics                  kube-system/metrics-server   True        29m

67. Reinstall netdata

When reinstalling, encounter dpkg-statoverride issue. solved via following:

# bash <(curl -Ss
# dpkg-statoverride --remove /var/lib/netdata/registry
# bash <(curl -Ss

Now re-install and we could got the netdata installed again.

68. display ps’s hirieachy

via ps auxf you could see a process’s father/son, etc.

69. git ignore last commit

via following command:

$ git reset --soft HEAD~1

70. forward to vm

via following command:

$ iptables -I FORWARD -o virbr1 -d -j ACCEPT
$ iptables -t nat -I PREROUTING -p tcp --dport 9867 -j DNAT --to

71. coredns issue


[root@k8s-node-1 ~]# kubectl logs -f -n kube-system coredns-9d85fb698-tnrgn
2019-04-29T12:26:42.180Z [INFO] plugin/reload: Running configuration MD5 =
2019-04-29T12:26:42.180Z [INFO] CoreDNS-1.5.0
2019-04-29T12:26:42.180Z [INFO] linux/amd64, go1.12.2, e3f9a80
linux/amd64, go1.12.2, e3f9a80
2019-04-29T12:26:48.181Z [ERROR] plugin/errors: 2
8373768935828175380.8715076686105595443. HINFO: read udp> i/o timeout
2019-04-29T12:26:51.181Z [ERROR] plugin/errors: 2
8373768935828175380.8715076686105595443. HINFO: read udp> i/o timeout
2019-04-29T12:26:52.181Z [ERROR] plugin/errors: 2
8373768935828175380.8715076686105595443. HINFO: read udp> i/o timeout
2019-04-29T12:26:53.181Z [ERROR] plugin/errors: 2
8373768935828175380.8715076686105595443. HINFO: read udp> i/o timeout
2019-04-29T12:26:56.181Z [ERROR] plugin/errors: 2
8373768935828175380.8715076686105595443. HINFO: read udp> i/o timeout
2019-04-29T12:26:59.182Z [ERROR] plugin/errors: 2
8373768935828175380.8715076686105595443. HINFO: read udp> i/o timeout
2019-04-29T12:27:02.182Z [ERROR] plugin/errors: 2
8373768935828175380.8715076686105595443. HINFO: read udp> i/o timeout
2019-04-29T12:27:05.182Z [ERROR] plugin/errors: 2
8373768935828175380.8715076686105595443. HINFO: read udp> i/o timeout
2019-04-29T12:27:08.182Z [ERROR] plugin/errors: 2
8373768935828175380.8715076686105595443. HINFO: read udp> i/o timeout
2019-04-29T12:27:11.183Z [ERROR] plugin/errors: 2
8373768935828175380.8715076686105595443. HINFO: read udp> i/o timeout

Solved via:

WARNING: IPtables FORWARD policy is DROP. Consider enabling traffic forwarding
with: sudo iptables -P FORWARD ACCEPT

fixing it worked for me

72. virt-manager issue

After upgrading to newest virt-manager(archlinux), vm could not boot for no bootable devices, solved via:

# virsh edit xxxxx

      <source file='/media/sda/ovirt/node1.qcow2' index='2'/>
+      <backingStore type='file' index='3'>
+        <format type='qcow2'/>
+        <source file='/media/sda/ovirt/Base/ovirtBase.qcow2'/>
+        <backingStore/>
+      </backingStore>

Then normally bootup the vm, the vm works properly.

73. vagrant plugin install speedup

via following command:

# vagrant plugin install vagrant-scp --plugin-clean-sources --plugin-source

74. vagrant with debug

via following command:


75. Tips for building wopi

Bypass skip tests:

mvn clean package -DskipTests

76. pip mirror in china


mkdir ~/.pip
cat <<EOF > ~/.pip/pip.conf
 trusted-host =
 index-url =

77. using bridge in libvirtd

Create br and set br’s parameter:

$ sudo brctl addbr libvirt0
$ sudo brctl setfd libvirt0 0

Then specify the shared device to libvirt0.

78. xrandr for rotating screen

rotate left and reback to normal:

# xrandr --output HDMI-0   --rotate left --mode 2560x1440  --right-of eDP-1-1
# xrandr --output HDMI-0   --rotate normal --mode 2560x1440  --right-of eDP-1-1