link to meetup event.
AwareInterface By Stev Leibelt
- reminder of dependency injection
- increases the readability of a class structure
- defined way of inject object/typ
- interface injection could lead to a project based rule to distinguish between mandatory and optional dependencies
- slides available hear.
Logstash and Graylog2 by Alexander Bernhardt
Simple log all.
questions to the log
- what happens on the other server?
- what happens when and on what circumstance?
- what are you logging?
how to solve this?
By using Logstash, Redis and ElasicSearch, aggregate all and push all to a graylog2.
- graylog can do some kind of monitoring, known as streams.
- you can add a alert method to each stream.
- nagios can listen to a graylog stream.
- could not replace statsd or graphite
Talk And Call For PHPUGHH Website By Stephan Vock And Ole Michaelis
- two Pages for one user group, which one to choose?
- the usergroup decides to use the new page.
- you need to have installed jekyll, rake or use the vagrant box
- directory structure
- each speaker should pull his own talk below "talks"
How To Contribute
- fork repository
- git checkout -b feature/foo
- do the changes
- git commit
- pull request to the main repository
- github.io builds new page on its own
- git remote add upstream firstname.lastname@example.org:phpughh/phpughh.github.com
- git fetch upstream
- git merge upstream/master
- git push
- issues are available
- enhancement issues is the roadmap
- open issues if needed
- links available as ticket
- take a look to the bephpug (github)
Next Level DevOps By Ole Michaelis
- infrastructure is created by feature (but without clean up or cleaning things up)
- did a practical training round in admin team for half a year
What Is DevOps?
- you build it, you run it (until it goes offline)
- monitor it, automate it, measure/metering it
- build small tools and services to form an api
How Became A DevOp?
- extend your development skills
- learn and do refactoring
- learn the shell
- learn the operation system
- pull deployment with deployment hooks
Softwarearchitecture Of A Modern Onlinegame By Tobias Rojahn
- crossplattform (mobile and web)
- realtime game (fast information about status changes)
- multiplayer (user can influence each other)
- user (request websocket from web server)
- NodeJS (handles websocket connection to redis and global redis)
- redis (message queue with push and pull)
- php worker (long living php process, asks webservices or database, are forkable by using php forking mechanism)
- if something is calculated by the php worker, the message is pushed in the global redis
The PHP Worker
- long living workers
- worker is observing the queue (and event handling) and is pushing the message to the dispatcher
- each message has a type to push to the right controller
- dispatcher is using the right controller (with a bunch of internal to external mapping and nearly zero logic)
- controller is pushing the message to the right domain service (using the right domain objects and/or repositories)
- controller is working with multiple db views
- domain service can also call the event dispatcher to trigger other systems (not core systems)
- only creates messages
- just time controlled game events
- socket (via socket io) replaces session
- add additional data (with indexes to gain speed) to the socket (via message queues from php)
- node to php: send all data
- php to node: a combination of data to all connected sockets
- forwarding of fitting messages (with removing of internal datas)
- ingoing message (test data)
- outgoing message (expectations)
- based on fitrues
- only php layer
- rollback on each test
This talk contains information about architecture.
When have done this for quite a while.
This works best in the past.
General solution was to add more hardware and that leads to more errors.
- monolithic architecture
- normalized data (in database)
- pull-principle (browser is build per request)
- full page cache is bad
- edge site include (for varnish/ESI) is also not the best solution
We are fast when:
- no real workload per request
- everything from memory
- denomalized data
How Often Does Data Change (From Rare To Often)?
Based on the example of a shop system.
- product information
Who Changes Data?
- product manager
- receipt of goods / outgoing goods
- images and static content is deliverd by static content webserver
- only dynamic part is done by php
- snippets are deliverd, not full pages
Benefit of key-value-store (for example redis) is, that the database itself can die and the only thing that happens is, that the entries are becoming obsolete.
But What About Filternavigation?
Use a searchengine that returns simple a collection of product keys. Use this product keys and ask the key value store to fetch product data.
But What About Personalization?
- use snippets with variables and default values
- fetch this snippet in the key value store
- thanks to "time to death" feature provided by many key-value-stores, you can easily define "special offers" per day and so on
Nice meetup, incredible how many people are attending already.