Deploy MediaWiki usgin Fiware Lab Facilities
Download as ODP, PDF0 likes157 views
How to use Fiware Lab Facilities in order to deploy an application. The presentation will lead us to deploy a full application taking advantage of Openstack Command Line Interfaces.
![32
# Retrieve an object from a container… In some scripts.
token=($(openstack token issue |
awk '/ id / || / project_id / {print $4}'))
TOKEN=${token[0]}
TENANT_ID=${token[1]}
URL=http://130.206.112.3:8080/v1
…
# Use the token, and the tenant.
swift --os-auth-token $TOKEN --os-storage-
url=$URL/AUTH_$TENANT_ID download summit wiki.tgz
Swift Commands (to automate)](https://image.slidesharecdn.com/fiwaremalaga-161223080815/85/Deploy-MediaWiki-usgin-Fiware-Lab-Facilities-32-320.jpg)
![Vagrant + Docker provider [+Puppet]](https://cdn.slidesharecdn.com/ss_thumbnails/npoggivagrant-docker-140725100147-phpapp02-thumbnail.jpg?width=560&fit=bounds)
Ad
More Related Content
What's hot (20)
Viewers also liked (20)
Ad
Similar to Deploy MediaWiki usgin Fiware Lab Facilities (20)
Ad
Recently uploaded (20)
![Download Wondershare Filmora Crack [2025] With Latest](https://cdn.slidesharecdn.com/ss_thumbnails/neo4j-howkgsareshapingthefutureofgenerativeaiatawssummitlondonapril2024-240426125209-2d9db05d-250419-250428115407-a04afffa-thumbnail.jpg?width=560&fit=bounds)
Deploy MediaWiki usgin Fiware Lab Facilities
- 1. Deploy Mediawiki using Fiware Lab facilities José Ignacio Carretero Guarde R&D Engineer at Telefonica I+D. In charge of Spain FIWARE Lab Node [email protected] @jicarreterogu
- 2. Mediawiki and FIWARE Lab APIs (Openstack) 2
- 3. 3 What Mediawiki has to do with FIWARE Lab (I) Nothing?
- 4. 4 ...But it could be good vehicle to show some things about FIWARE Lab IaaS reference Implementation GE. What Mediawiki has to do with FIWARE Lab (II)
- 5. 5 ● Why Wikimedia? – If it has nothing to do with FIWARE. ● It has a backend Database (Mysql) ● It has a Frontend Web Server (Apache) ● I Allows me to show some tools: IaaS GE. ● I Know how it Works, So I can automate it. ● Images and files are stored in Apache (…/wiki/images) ● Data in the database is stored in /var/lib/mysql Installing an Applicacion: Mediawiki
- 6. 6 ● https://catalogue.fiware.org/enablers/iaas-ge-fiware-reference-implement ● “We are using OpenStack Vanilla release for the IaaS GE.” ● Openstack APIs work: ● Compute APIs ● Block Storage APIs ● Image Service ● Object Storage ● There are CLIs writen in Python... IaaS GE – FIWARE Reference Implementation
- 7. 7 # Assuming Python 2.7, pip and virtualenv installed virtualenv osdemo source osdemo/bin/activate pip install python-openstackclient pip install python-swiftclient How to locally install the CLIs
- 8. A First Approach to the Installation Process 8
- 9. 9 Installing an Applicacion: Mediawiki (I)
- 10. 10 ● Virtual Machines – Instances ● Instances are managed by Nova. ● We need a few things before deploying an instance ● A Keypair ● In order to be Able to SSH our instances ● Security Groups/Rules ● Because OpenStack acts like a closed Firewall for every instance. Create a first VM
- 11. 11 # Creating the keypair nova keypair-add summitkp | tee summitkp && chmod 400 summitkp # Creating the security group and rules nova secgroup-create openthings opens_some_ports nova secgroup-add-rule openthings tcp 22 22 0/0 nova secgroup-add-rule openthings tcp 80 80 0/0 nova secgroup-add-rule openthings icmp -1 -1 0/0 # Creating the security group and rules (What I did!): nova secgroup-create allopen open_everything nova secgroup-add-rule allopen tcp 1 65535 0/0 nova secgroup-add-rule allopen ucp 1 65535 0/0 nova secgroup-add-rule allopen icmp -1 -1 0/0 Nova: Key Pairs and Security Groups
- 12. 12 ● Boot our first instance with some parameters ● (At least) One network ID to allow networking ● Flavor – The size of the Instance ● Our Security Group – our firewall configuration. ● A Keypair – To SSH our instance ● The base image – I’ll use an Ubuntu in the demos. ● Assign a floating IP to some of our instances ● A Floating IP is needed to SSH our instances. We might not be able to SSH private IPs ● 1 Floating IP is allowed per User. Create a first VM (ii)
- 13. 13 # Find the network ID NETUUID=$(openstack network list | awk '/ node-int-net-01 / {print $2}') # Boot Virtual Machine nova boot --nic net-id=$NETUUID --image base_ubuntu_14.04 --key-name summitkp --security-groups allopen --flavor m1.small mydatabase # Creating a floating IP: PUBLIC_EXT_NET=public-ext-net-01 openstack floating ip create $PUBLIC_EXT_NET # Associate the IP to our instance: nova list floating ip create $PUBLIC_EXT_NET openstack floating list nova floating-ip-associate mydatabase 130.206.112.0 Nova: Boot and add a floating IP
- 14. 14 # Get console URL nova get-vnc-console mydatabase novnc # Creating the security group and rules nova console-log mydatabase Nova: Consoles
- 15. 15 ● Compute is thought to Compute ● When Instances die, Ephemeral disks die ● Persistent Disk – Block Storage ● Cinder manages Persistent Disks ● The disks, once created can be attached to instances ● Instances must format (once) and mount the disks before use ● Persistent disk will survive Instances. ● A way to Keep data apart from computation. Cinder: Ephemeral Vs Persistent disks
- 16. 16 # Create a Persistent disk openstack volume create --size 1 myvolume openstack volume list VOLUMEID=$(openstack volume list | awk '/ myvolume / {print $2}') # Attaching the volume to the instance nova volume-attach mydatabase $VOLUMEID # Or attaching at boot time… nova boot --nic net-id=$NETUUID --image base_ubuntu_14.04 --key-name summitkp --security-groups allopen --block-device-mapping vdb=$VOLUMEID --flavor m1.small mydatabase Cinder: Persistent Disk Creation
- 17. Automate!!! 17
- 18. 18 ● Increase Productivity ● Automated processes reduce defects ● Reduce Human Errors ● Processes are run effortlessly ● Processes are more flexible => Changes are easier ● Increase satisfaction ● … You can write many other things here ● … And here. Automate: Why?
- 19. 19 Automate: Using GUI (Cloud Portal) (maybe... ) Too many clicks (maybe)... Too error prone (almost)... Impossible automation Easy to use
- 20. 20 ● Tools to automate ● Puppet/Chef ● An Agent is required in the instance ● Ansible ● Many features, even some for OpenStack ● Some OpenStack Tools ● Heat / Murano ● Ad Hoc Scripts ● This time it was my option, so I can show some CLI commands Automate: How?
- 21. 21 Installing an Applicacion: Mediawiki (II) /var/www/html/wiki/images /var/lib/mysql apache-vol kmysql-vol
- 22. 22 ● Create the Persistent Disk (if it doesn’t exists) ● Boot the VM attaching the Persistent Disk and injecting a Script ● The Script ● Formats the Persistent Disk and mounts it (using /etc/fstab) ● /var/lib/mysql ● Installs MySQL-Server software ● Creates Database and user for the Wiki. One Installation Process: Database
- 23. 23 ● Create the Persistent Disk (if it doesn’t exists) ● Boot the VM attaching the Persistent Disk and injecting a Script ● The Script ● Formats the Persistent Disk ● Installs Apache, php5, libs, etc ● Downloads Wikimedia and untars it /var/www/html ● Downloads SyntaxHighlight_plugin and untars it. ● Mounts the disk (/var/www/html/wiki/images) ● Creates a FloatingIP and associates it to the instance One Installation Process: Apache
- 24. 24 # Or attaching at boot time… nova boot --nic net-id=$NETUUID --image base_ubuntu_14.04 --key-name summitkp --security-groups allopen --block-device-mapping vdb=$VOLUMEID --user-data oneScript --flavor m1.small mydatabase The new boot command for Nova
- 26. 26 ● Mediawiki installation can’t be automated… but ● I can automate LocalSettings.php configuration ● The Script ● Gets the Public IP of our apache ● Sets the Logo, Extensions of Documents which can be uploaded ● Configures SyntaxHighlight extension ● Uploads LocalSettings.php and logo.jpg Can’t automate everything... but almost
- 27. Some other tools: Images and Object Storage 27
- 28. 28 ● I can take Snapshots of my instances ● So I can boot a preconfigured Instance ● Glance - The image service is used to create Snapshots ● I’ve taken an Snapshot of my Database Instance ● The process takes some time... ● It is a complicated process that involves many subprocesses Snapshots from instances
- 29. 29 # Create an Snapshot from an Instance nova image-create krtmysql krtmysql-snp # List images openstack image list # Delete images openstack image delete $IMAGE_ID Image Commands
- 30. 30 ● It stores static Objects ● We can retrieve the objects in the future ● Swift is the reference implementation of FIWARE’s Object Storage GE ● https://catalogue.fiware.org/enablers/object-storage-ge-fiware-imple ● Objects are Stored in containers ● I’ve uploaded my Preconfigured Mediawiki in a .tgz file to Object Store ● Let’s use this in next Mediawiki installations (as a demo) Object Storage
- 31. 31 # Create a new Container swift post summit # Upload an object to the Container swift upload summit wiki.tgz # List Containers, list objects from a container swift list swift list summit # Retrieve an object from a container swift download summit wiki.tgz Swift Commands
- 32. 32 # Retrieve an object from a container… In some scripts. token=($(openstack token issue | awk '/ id / || / project_id / {print $4}')) TOKEN=${token[0]} TENANT_ID=${token[1]} URL=http://130.206.112.3:8080/v1 … # Use the token, and the tenant. swift --os-auth-token $TOKEN --os-storage- url=$URL/AUTH_$TENANT_ID download summit wiki.tgz Swift Commands (to automate)