Enterprise Hosting
Download as KEY, PDF5 likes824 views
The document discusses enterprise hosting solutions for Ruby and Python applications, highlighting the need for maximum availability and acceptable performance. It outlines the challenges of building a reliable cluster, including server redundancy and the complexity involved in managing multiple servers. Additionally, the document provides technical insights on components such as load balancers, application servers, and storage systems essential for optimal hosting efficiency.
Enterprise Hosting
- 1. Enterprise Hosting for Ruby and Python Julian Fischer fischer@enterprise-rails.de http://www.enterprise-rails.de
- 2. Introduction
- 3. Introduction About me
- 4. About me Julian Fischer ‣Twitter: http://www.twitter.com/railshoster ‣E-Mail: fischer@enterprise-rails.de
- 5. About me Julian Fischer ‣Twitter: http://www.twitter.com/railshoster ‣E-Mail: fischer@enterprise-rails.de ‣ CEO of Avarteq GmbH
- 6. About me Julian Fischer ‣Twitter: http://www.twitter.com/railshoster ‣E-Mail: fischer@enterprise-rails.de ‣ CEO of Avarteq GmbH ‣ Lecturer „Ruby on Rails“ @ HTWdS
- 7. About me Julian Fischer ‣Twitter: http://www.twitter.com/railshoster ‣E-Mail: fischer@enterprise-rails.de ‣ CEO of Avarteq GmbH ‣ Lecturer „Ruby on Rails“ @ HTWdS ‣ Ruby und Ruby on Rails programmer
- 8. About me Julian Fischer ‣Twitter: http://www.twitter.com/railshoster ‣E-Mail: fischer@enterprise-rails.de ‣ CEO of Avarteq GmbH ‣ Lecturer „Ruby on Rails“ @ HTWdS ‣ Ruby und Ruby on Rails programmer ‣ Entperise-Rails.de - Head of Hosting
- 10. About Avarteq GmbH http://www.twitter.com/railshoster
- 11. About Avarteq GmbH ‣ Founded in Nov. 2008 from two existing companies. http://www.twitter.com/railshoster
- 12. About Avarteq GmbH ‣ Founded in Nov. 2008 from two existing companies. ‣ Involvment of Key-Systems GmbH manages ~2,5 * 10^6 domains for customers of 200+ countries. http://www.twitter.com/railshoster
- 13. About Avarteq GmbH ‣ Founded in Nov. 2008 from two existing companies. ‣ Involvment of Key-Systems GmbH manages ~2,5 * 10^6 domains for customers of 200+ countries. ‣ Team size: 14 people 8 full-time, 6 part-time/freelancer http://www.twitter.com/railshoster
- 15. About Avarteq GmbH http://www.twitter.com/railshoster
- 16. About Avarteq GmbH ‣ Covers all stages of web development http://www.twitter.com/railshoster
- 17. About Avarteq GmbH ‣ Covers all stages of web development ‣ Consulting http://www.twitter.com/railshoster
- 18. About Avarteq GmbH ‣ Covers all stages of web development ‣ Consulting ‣ Conceptual ~ and screen design http://www.twitter.com/railshoster
- 19. About Avarteq GmbH ‣ Covers all stages of web development ‣ Consulting ‣ Conceptual ~ and screen design ‣ Ruby&onplace. development. In house in Rails http://www.twitter.com/railshoster
- 20. About Avarteq GmbH ‣ Covers all stages of web development ‣ Consulting ‣ Conceptual ~ and screen design ‣ Ruby&onplace. development. In house in Rails ‣ Ruby on Rails hosting/servers/clusters RailsHoster.de - Enterprise-Rails.de http://www.twitter.com/railshoster
- 22. Enterprise Hosting Situation
- 23. EH - Basic Idea http://www.twitter.com/railshoster
- 24. EH - Basic Idea ‣ Given http://www.twitter.com/railshoster
- 25. EH - Basic Idea ‣ Given ‣ Python / Ruby / Rails App http://www.twitter.com/railshoster
- 26. EH - Basic Idea ‣ Given ‣ Python / Ruby / Rails App ‣ Needed http://www.twitter.com/railshoster
- 27. EH - Basic Idea ‣ Given ‣ Python / Ruby / Rails App ‣ Needed ‣ Max. availability ideal: 100% uptime http://www.twitter.com/railshoster
- 28. EH - Basic Idea ‣ Given ‣ Python / Ruby / Rails App ‣ Needed ‣ Max. availability ideal: 100% uptime ‣ Acceptable performance even for peaks, e.g. <= 200 ms response time / request http://www.twitter.com/railshoster
- 29. Enterprise Hosting Obstacles
- 30. EH - Obstacles http://www.twitter.com/railshoster
- 31. EH - Obstacles ‣ Single server = a lot of SPOFs http://www.twitter.com/railshoster
- 32. EH - Obstacles ‣ Single server = a lot of SPOFs ‣ Multiple servers = a lot of work http://www.twitter.com/railshoster
- 33. EH - Obstacles ‣ Single server = a lot of SPOFs ‣ Multiple servers = a lot of work ‣ Eliminating SPOFs is not easy/possible for every service http://www.twitter.com/railshoster
- 34. How to build a fancy & rockin‘ cluster?
- 35. Cluster Building Blocks Monitoring DBApp Backups Storage MA/SL LB Enterprise Cluster
- 36. Cluster Building Blocks Monitoring DB MA/SL Backups Storage LB App Enterprise Cluster
- 37. Cluster Building Blocks Monitoring DB MA/SL Backups Storage App LB Enterprise Cluster
- 38. Cluster Building Blocks Monitoring DB MA/SL Backups Storage App LB Enterprise Cluster
- 39. Cluster Building Blocks Monitoring Backups DBApp Storage MA/SL LB Enterprise Cluster
- 40. Cluster Building Blocks Monitoring DBApp Backups Storage MA/SL LB Enterprise Cluster
- 41. Cluster Building Blocks Monitoring DBApp Backups Storage MA/SL LB Enterprise Cluster
- 42. Cluster Building Blocks Monitoring DBApp Backups Storage MA/SL LB Enterprise Cluster ✓
- 43. maybe it‘s not that easy :)
- 44. How to build a cluster Requirements Analysis
- 45. How to build a cluster http://www.twitter.com/railshoster
- 46. How to build a cluster ‣ desired response time which again depends on several factors http://www.twitter.com/railshoster
- 47. How to build a cluster ‣ desired response time which again depends on several factors ‣ average load e.g. on a cloudy sunday afternoon http://www.twitter.com/railshoster
- 48. How to build a cluster ‣ desired response time which again depends on several factors ‣ average load e.g. on a cloudy sunday afternoon ‣ peak load e.g. after being a new trend on twitter http://www.twitter.com/railshoster
- 49. How to build a cluster ‣ desired response time which again depends on several factors ‣ average load e.g. on a cloudy sunday afternoon ‣ peak load e.g. after being a new trend on twitter ‣ ... http://www.twitter.com/railshoster
- 50. How to build a cluster http://www.twitter.com/railshoster
- 51. How to build a cluster ‣ which components are present? search server, background processing, key/value store, replicating rdbms, ... http://www.twitter.com/railshoster
- 52. How to build a cluster ‣ which components are present? search server, background processing, key/value store, replicating rdbms, ... ‣ apptired programmers? & quality coded by super geeks or a lot of performance http://www.twitter.com/railshoster
- 53. How to build a cluster http://www.twitter.com/railshoster
- 54. How to build a cluster ‣ Do it yourself or full management? http://www.twitter.com/railshoster
- 55. How to build a cluster ‣ Do it yourself or full management? ‣ admin know how present? Is your team able to care of things like http://www.twitter.com/railshoster
- 56. How to build a cluster ‣ Do it yourself or full management? ‣ admin know how present? Is your team able to care of things like ‣ os updates http://www.twitter.com/railshoster
- 57. How to build a cluster ‣ Do it yourself or full management? ‣ admin know how present? Is your team able to care of things like ‣ os updates ‣ web server & load balancer setup http://www.twitter.com/railshoster
- 58. How to build a cluster ‣ Do it yourself or full management? ‣ admin know how present? Is your team able to care of things like ‣ os updates ‣ web server & load balancer setup ‣ db configuration & tuning http://www.twitter.com/railshoster
- 59. How to build a cluster ‣ Do it yourself or full management? ‣ admin know how present? Is your team able to care of things like ‣ os updates ‣ web server & load balancer setup ‣ db configuration & tuning ‣ monitoring, ... http://www.twitter.com/railshoster
- 60. How to build a cluster http://www.twitter.com/railshoster
- 61. How to build a cluster ‣ Willing to fight with a large number of servers a lot of programmers don‘t want to do both: programming & hosting http://www.twitter.com/railshoster
- 62. How to build a cluster ‣ Willing to fight with a large number of servers a lot of programmers don‘t want to do both: programming & hosting ‣ Available budget http://www.twitter.com/railshoster
- 63. How to build a cluster ‣ Willing to fight with a large number of servers a lot of programmers don‘t want to do both: programming & hosting ‣ Available budget ‣ more servers ... more reliability, more performance http://www.twitter.com/railshoster
- 64. How to build a cluster ‣ Willing to fight with a large number of servers a lot of programmers don‘t want to do both: programming & hosting ‣ Available budget ‣ more servers ... more reliability, more performance ‣ Many other questions ... http://www.twitter.com/railshoster
- 65. How to build a cluster Choose your HW platform
- 66. Choose HW plattform http://www.twitter.com/railshoster
- 67. Choose HW plattform ‣ Physical HW low budget or real server HW? http://www.twitter.com/railshoster
- 68. Choose HW plattform ‣ Physical HW low budget or real server HW? ‣ Virtual HW Amazon EC2, Joyent, EY, ... http://www.twitter.com/railshoster
- 69. Choose HW plattform ‣ Physical HW low budget or real server HW? ‣ Virtual HW Amazon EC2, Joyent, EY, ... ‣ Assume we use physical HW ... http://www.twitter.com/railshoster
- 70. Choose HW plattform http://www.twitter.com/railshoster
- 71. Choose HW plattform ‣ Compare TCO and performance/€ http://www.twitter.com/railshoster
- 72. Choose HW plattform ‣ Compare TCO and performance/€ ‣ When is your next scale out? http://www.twitter.com/railshoster
- 73. Choose HW plattform ‣ Compare TCO and performance/€ ‣ When is your next scale out? ‣ Do you really need to scale within hours? ... and is it worth to pay a lot of extra money? Estimate your resource consumption grow rate e.g. using Munin. http://www.twitter.com/railshoster
- 74. How to build a cluster Cluster dimensions
- 75. Cluster dimensions http://www.twitter.com/railshoster
- 76. Cluster dimensions ‣ How many servers needed? 2, 4, 6, 10, ... http://www.twitter.com/railshoster
- 77. Cluster dimensions ‣ How many servers needed? 2, 4, 6, 10, ... ‣ Which systems components will be there? App, MA DB, SL DB, BG-Jobs, Search-Server, Memcached, Storage, ... http://www.twitter.com/railshoster
- 78. Cluster dimensions ‣ How many servers needed? 2, 4, 6, 10, ... ‣ Which systems components will be there? App, MA DB, SL DB, BG-Jobs, Search-Server, Memcached, Storage, ... ‣ Which HW will be most suitable for each individual role? DB Server = a lot of fast hdds, BG-Kob Server = a lot of cpus, ... http://www.twitter.com/railshoster
- 79. How to build a cluster Cluster design
- 80. Cluster design http://www.twitter.com/railshoster
- 81. Cluster design ‣ How to distribute components to servers? http://www.twitter.com/railshoster
- 82. Cluster design ‣ How to distribute components to servers? ‣ App Server & DB Server together? http://www.twitter.com/railshoster
- 83. Cluster design ‣ How to distribute components to servers? ‣ App Server & DB Server together? ‣ Dedicated DB Server? http://www.twitter.com/railshoster
- 84. Cluster design http://www.twitter.com/railshoster
- 85. Cluster design ‣ Where to put the search server? Sphinx, Solr, ... http://www.twitter.com/railshoster
- 86. Cluster design ‣ Where to put the search server? Sphinx, Solr, ... ‣ Where to run background jobs? http://www.twitter.com/railshoster
- 87. Cluster design ‣ Where to put the search server? Sphinx, Solr, ... ‣ Where to run background jobs? ‣ Where to run message queues? See talk of Paolo Negri. http://www.twitter.com/railshoster
- 88. How to build a cluster Cluster building blocks
- 89. Cluster building blocks Network
- 90. Network http://www.twitter.com/railshoster
- 91. Network ‣ Connection between cluster nodes http://www.twitter.com/railshoster
- 92. Network ‣ Connection between cluster nodes ‣ Use a private (!) physical network http://www.twitter.com/railshoster
- 93. Network ‣ Connection between cluster nodes ‣ Use a private (!) physical network ‣ Minimize communication latency distribution might cost performance for low load scenarios http://www.twitter.com/railshoster
- 94. Network ‣ Connection between cluster nodes ‣ Use a private (!) physical network ‣ Minimize communication latency distribution might cost performance for low load scenarios ‣ >= 1even within the same datacenter hosters limit bandwith between servers GBit/s be aware that some http://www.twitter.com/railshoster
- 95. Cluster building blocks Application Server
- 96. Application Server http://www.twitter.com/railshoster
- 97. Application Server ‣ Ruby / Ruby on Rails http://www.twitter.com/railshoster
- 98. Application Server ‣ Ruby / Ruby on Rails ‣ Passenger our favorite. http://www.twitter.com/railshoster
- 99. Application Server ‣ Ruby / Ruby on Rails ‣ Passenger our favorite. ‣ LB &&interesting. Thin and mongrel do their job. Unicorn is (Unicorn || Thin || Mongrel) http://www.twitter.com/railshoster
- 100. Application Server ‣ Ruby / Ruby on Rails ‣ Passenger our favorite. ‣ LB &&interesting. Thin and mongrel do their job. Unicorn is (Unicorn || Thin || Mongrel) ‣ JRuby with Glassfish-Gem or servlet containers like Glassfish-Server, Jetty, Tomacat, ... http://www.twitter.com/railshoster
- 101. Application Server ‣ Ruby / Ruby on Rails ‣ Passenger our favorite. ‣ LB &&interesting. Thin and mongrel do their job. Unicorn is (Unicorn || Thin || Mongrel) ‣ JRuby with Glassfish-Gem or servlet containers like Glassfish-Server, Jetty, Tomacat, ... ‣ others CGI, fastCGI, SCGI, ... http://www.twitter.com/railshoster
- 102. Application Server http://www.twitter.com/railshoster
- 103. Application Server ‣ Python http://www.twitter.com/railshoster
- 104. Application Server ‣ Python ‣ mod_python run Python interpreter inside apache. http://www.twitter.com/railshoster
- 105. Application Server ‣ Python ‣ mod_python run Python interpreter inside apache. ‣ mod_wsgi better speed & memory usage than mod_python. http://www.twitter.com/railshoster
- 106. Application Server ‣ Python ‣ mod_python run Python interpreter inside apache. ‣ mod_wsgi better speed & memory usage than mod_python. ‣ others CGI, fastCGI, SCGI, ... http://www.twitter.com/railshoster
- 107. Cluster building blocks Load Balancer
- 108. Load Balancer http://www.twitter.com/railshoster
- 109. Load Balancer ‣ Most likely not the bottleneck http://www.twitter.com/railshoster
- 110. Load Balancer ‣ Most likely not the bottleneck ‣ Often a single point of failure (SPOF) http://www.twitter.com/railshoster
- 111. Load Balancer ‣ Most likely not the bottleneck ‣ Often a single point of failure (SPOF) ‣ Two are better than one! http://www.twitter.com/railshoster
- 112. Load Balancer ‣ Most likely not the bottleneck ‣ Often a single point of failure (SPOF) ‣ Two are better than one! ‣ What to do if one fails? http://www.twitter.com/railshoster
- 113. Load Balancer ‣ Most likely not the bottleneck ‣ Often a single point of failure (SPOF) ‣ Two are better than one! ‣ What to do if one fails? ‣ Setup automatic IP failover not possible in every datacenter http://www.twitter.com/railshoster
- 114. Load Balancer http://www.twitter.com/railshoster
- 115. Load Balancer ‣ Open Source Software LBs http://www.twitter.com/railshoster
- 116. Load Balancer ‣ Open Source Software LBs ‣ Apache, NGINX, Pound, HA Proxy, ... http://www.twitter.com/railshoster
- 117. Load Balancer ‣ Open Source Software LBs ‣ Apache, NGINX, Pound, HA Proxy, ... ‣ LB Appliances BalanceNG, ... http://www.twitter.com/railshoster
- 118. Cluster building blocks Storage
- 119. Storage http://www.twitter.com/railshoster
- 120. Storage ‣ Use a shared filesystem to store user uploads such as profile pictures, etc. http://www.twitter.com/railshoster
- 121. Storage ‣ Use a shared filesystem to store user uploads such as profile pictures, etc. ‣ Some storage technologies http://www.twitter.com/railshoster
- 122. Storage ‣ Use a shared filesystem to store user uploads such as profile pictures, etc. ‣ Some storage technologies ‣ NFS, sync or async, often a SPOF, lot of overhead http://www.twitter.com/railshoster
- 123. Storage ‣ Use a shared filesystem to store user uploads such as profile pictures, etc. ‣ Some storage technologies ‣ NFS, sync or async, often a SPOF, lot of overhead ‣ RSync, async, replication lag http://www.twitter.com/railshoster
- 124. Storage ‣ Use a shared filesystem to store user uploads such as profile pictures, etc. ‣ Some storage technologies ‣ NFS, sync or async, often a SPOF, lot of overhead ‣ RSync, async, replication lag ‣ iSCSI && (GFS || OCFS2), incredibly fast http://www.twitter.com/railshoster
- 125. Storage iSCSI
- 126. iSCSI Storage http://www.twitter.com/railshoster
- 127. iSCSI Storage ‣ iSCSI = SCSI over TCP http://www.twitter.com/railshoster
- 128. iSCSI Storage ‣ iSCSI = SCSI over TCP ‣ Exports a blockdevice http://www.twitter.com/railshoster
- 129. iSCSI Storage ‣ iSCSI = SCSI over TCP ‣ Exports a blockdevice ‣ iSCSI = less overhead than NFS http://www.twitter.com/railshoster
- 130. iSCSI Storage ‣ iSCSI = SCSI over TCP ‣ Exports a blockdevice ‣ iSCSI = less overhead than NFS ‣ Need to use a distributed FS like http://www.twitter.com/railshoster
- 131. iSCSI Storage ‣ iSCSI = SCSI over TCP ‣ Exports a blockdevice ‣ iSCSI = less overhead than NFS ‣ Need to use a distributed FS like ‣ GFS http://www.twitter.com/railshoster
- 132. iSCSI Storage ‣ iSCSI = SCSI over TCP ‣ Exports a blockdevice ‣ iSCSI = less overhead than NFS ‣ Need to use a distributed FS like ‣ GFS ‣ OCFS2 http://www.twitter.com/railshoster
- 134. iSCSI Storage Appliance http://www.twitter.com/railshoster
- 135. iSCSI Storage Appliance ‣ Think about using a storage appliance, e.g. NetApp http://www.twitter.com/railshoster
- 136. iSCSI Storage Appliance ‣ Think about using a storage appliance, e.g. NetApp ‣ Has redundant HW http://www.twitter.com/railshoster
- 137. iSCSI Storage Appliance ‣ Think about using a storage appliance, e.g. NetApp ‣ Has redundant HW ‣ Out of the box solution http://www.twitter.com/railshoster
- 138. iSCSI Storage Appliance ‣ Think about using a storage appliance, e.g. NetApp ‣ Has redundant HW ‣ Out of the box solution ‣ Very low maintenance http://www.twitter.com/railshoster
- 139. iSCSI Storage Appliance ‣ Think about using a storage appliance, e.g. NetApp ‣ Has redundant HW ‣ Out of the box solution ‣ Very low maintenance ‣ Connect each app server via dedicated GBit LAN connections http://www.twitter.com/railshoster
- 140. Cluster building blocks Database
- 141. Database http://www.twitter.com/railshoster
- 142. Database ‣ RDBMS? There are alternatives CouchDB, MongoDB, ... http://www.twitter.com/railshoster
- 143. Database ‣ RDBMS? There are alternatives CouchDB, MongoDB, ... ‣ Most likely MySQL, Postgres, ... http://www.twitter.com/railshoster
- 144. Database ‣ RDBMS? There are alternatives CouchDB, MongoDB, ... ‣ Most likely MySQL, Postgres, ... ‣ Horizontal DB scale out is ugly http://www.twitter.com/railshoster
- 145. Database ‣ RDBMS? There are alternatives CouchDB, MongoDB, ... ‣ Most likely MySQL, Postgres, ... ‣ Horizontal DB scale out is ugly ‣ Scaleout vertically, first http://www.twitter.com/railshoster
- 146. Database ‣ RDBMS? There are alternatives CouchDB, MongoDB, ... ‣ Most likely MySQL, Postgres, ... ‣ Horizontal DB scale out is ugly ‣ Scaleout vertically, first ‣ CPU++, RAM++, a lot of fast HDDs http://www.twitter.com/railshoster
- 147. Horizontal MySQL scaling?
- 149. Database http://www.twitter.com/railshoster
- 150. Database ‣ NDB MySQL Cluster http://www.twitter.com/railshoster
- 151. Database ‣ NDB MySQL Cluster ‣ Features look performance, ... automatic failover, transactions, good write very nice http://www.twitter.com/railshoster
- 152. Database ‣ NDB MySQL Cluster ‣ Features look performance, ... automatic failover, transactions, good write very nice ‣ At failover advantages. gain least 3 physical maschines needed to http://www.twitter.com/railshoster
- 153. Database ‣ NDB MySQL Cluster ‣ Features look performance, ... automatic failover, transactions, good write very nice ‣ At failover advantages. gain least 3 physical maschines needed to ‣ Certain joins. such as large statements are incredibly slow http://www.twitter.com/railshoster
- 154. Database ‣ NDB MySQL Cluster ‣ Features look performance, ... automatic failover, transactions, good write very nice ‣ At failover advantages. gain least 3 physical maschines needed to ‣ Certain joins. such as large statements are incredibly slow ‣ DB sizestored in memory. RAM size since indexed columns are limited by http://www.twitter.com/railshoster
- 155. Database http://www.twitter.com/railshoster
- 156. Database ‣ NDB MySQL Cluster http://www.twitter.com/railshoster
- 157. Database ‣ NDB MySQL Cluster ‣ Only suitable for certain situations http://www.twitter.com/railshoster
- 158. Database ‣ NDB MySQL Cluster ‣ Only suitable for certain situations ‣ Joins avoided by using a search server solr, sphinx, ... http://www.twitter.com/railshoster
- 159. Database ‣ NDB MySQL Cluster ‣ Only suitable for certain situations ‣ Joins avoided by using a search server solr, sphinx, ... ‣ Many write transactions but slow data grow rate http://www.twitter.com/railshoster
- 160. Database ‣ NDB MySQL Cluster ‣ Only suitable for certain situations ‣ Joins avoided by using a search server solr, sphinx, ... ‣ Many write transactions but slow data grow rate ‣ Maybe replication is an alternative ... http://www.twitter.com/railshoster
- 161. Database Replication
- 162. Database http://www.twitter.com/railshoster
- 163. Database ‣ MA/MA replication http://www.twitter.com/railshoster
- 164. Database ‣ MA/MA replication ‣ Possible but http://www.twitter.com/railshoster
- 165. Database ‣ MA/MA replication ‣ Possible but ‣ Not conflicts record recommendable except your app resolves db http://www.twitter.com/railshoster
- 166. Database http://www.twitter.com/railshoster
- 167. Database ‣ MA/SL replication http://www.twitter.com/railshoster
- 168. Database ‣ MA/SL replication ‣ Good starting point http://www.twitter.com/railshoster
- 169. Database ‣ MA/SL replication ‣ Good starting point ‣ SL = realtime backup of MA DB but http://www.twitter.com/railshoster
- 170. Database ‣ MA/SL replication ‣ Good starting point ‣ SL = realtime backup of MA DB but ‣ Can be used read-only http://www.twitter.com/railshoster
- 171. Database ‣ MA/SL replication ‣ Good starting point ‣ SL = realtime backup of MA DB but ‣ Can be used read-only ‣ No automatica failover but manual failoverfor the faster than reinstalling single server. But calculate time is still master reintegration! http://www.twitter.com/railshoster
- 172. Software Deployment
- 173. Deployment
- 175. Deployment ‣ Capistrano ‣ Adapt deployment recipe to server and server roles
- 176. Deployment ‣ Capistrano ‣ Adapt deployment recipe to server and server roles ‣ Initial deployment painful
- 177. Deployment ‣ Capistrano ‣ Adapt deployment recipe to server and server roles ‣ Initial deployment painful ‣ Further deployments more or less painless
- 178. Backups
- 179. Backups
- 180. Backups ‣ Create automated backups!
- 181. Backups ‣ Create automated backups! ‣ Cover filesystem and database
- 182. Backups ‣ Create automated backups! ‣ Cover filesystem and database ‣ Create them, pack them, encrypt them and upload them to a safe place. e.g. use duplicity!
- 183. Backups ‣ Create automated backups! ‣ Cover filesystem and database ‣ Create them, pack them, encrypt them and upload them to a safe place. e.g. use duplicity! ‣ Test backups regularily. Broken backups = no backups.
- 184. Monitoring
- 185. „Just because you’re paranoid, doesn’t mean they’re not out to get you.“
- 186. How does your admin know that something is broken?
- 187. Monitoring Nagios
- 188. Nagios
- 189. Nagios ‣ Nagios monitors key resources of a system. At least it that how it should be configured.
- 190. Nagios ‣ Nagios monitors key resources of a system. At least it that how it should be configured. ‣ Is some resource not responding or hits a threshold: notify! Again, that‘s how it should be.
- 191. Nagios ‣ Nagios monitors key resources of a system. At least it that how it should be configured. ‣ Is some resource not responding or hits a threshold: notify! Again, that‘s how it should be. ‣ Also consider using monit for local proactive monitoring issues.
- 192. Nagios
- 193. Nagios ‣ What should be checked? List is incomplete.
- 194. Nagios ‣ What should be checked? List is incomplete. ‣ Swap. Lot of swap usage = not enough RAM = Sloooooow.
- 195. Nagios ‣ What should be checked? List is incomplete. ‣ Swap. Lot of swap usage = not enough RAM = Sloooooow. ‣ LOAD. More sysload than cores? Processes need to wait.
- 196. Nagios ‣ What should be checked? List is incomplete. ‣ Swap. Lot of swap usage = not enough RAM = Sloooooow. ‣ LOAD. More sysload than cores? Processes need to wait. ‣ HTTP. Webserver, are you still there?
- 197. Nagios ‣ What should be checked? List is incomplete. ‣ Swap. Lot of swap usage = not enough RAM = Sloooooow. ‣ LOAD. More sysload than cores? Processes need to wait. ‣ HTTP. Webserver, are you still there? ‣ APP. Does my app still serve my control string XY?
- 198. Nagios ‣ What should be checked? List is incomplete. ‣ Swap. Lot of swap usage = not enough RAM = Sloooooow. ‣ LOAD. More sysload than cores? Processes need to wait. ‣ HTTP. Webserver, are you still there? ‣ APP. Does my app still serve my control string XY? ‣ and a lot more!
- 200. Resource History
- 202. Resource History
- 203. Resource History ‣ Stores the consumption of individual resources over time.
- 204. Resource History ‣ Stores the consumption of individual resources over time. ‣ Trend analysis. What happened after the last puplicity ploy on our servers?
- 205. Resource History ‣ Stores the consumption of individual resources over time. ‣ Trend analysis. What happened after the last puplicity ploy on our servers? ‣ Scale out needed? Or maybe in a month?
- 206. Resource History ‣ Stores the consumption of individual resources over time. ‣ Trend analysis. What happened after the last puplicity ploy on our servers? ‣ Scale out needed? Or maybe in a month? ‣ Does my app leak?
- 207. Maintenance
- 208. Maintenance
- 209. Maintenance ‣ OS updates. updates. Minor-, major-, distro
- 210. Maintenance ‣ OS updates. updates. Minor-, major-, distro ‣ Take a server out of the cluster to perform bigger updates. Reintegrate if afterwards. No or minor service impact.
- 211. Maintenance ‣ OS updates. updates. Minor-, major-, distro ‣ Take a server out of the cluster to perform bigger updates. Reintegrate if afterwards. No or minor service impact. ‣ Ruby Updates. Ruby, Rails, Gems, ...
- 212. Maintenance ‣ OS updates. updates. Minor-, major-, distro ‣ Take a server out of the cluster to perform bigger updates. Reintegrate if afterwards. No or minor service impact. ‣ Ruby Updates. Ruby, Rails, Gems, ... ‣ Rotation and monitoring of logs. Bad things going on?
- 213. Maintenance ‣ OS updates. updates. Minor-, major-, distro ‣ Take a server out of the cluster to perform bigger updates. Reintegrate if afterwards. No or minor service impact. ‣ Ruby Updates. Ruby, Rails, Gems, ... ‣ Rotation and monitoring of logs. Bad things going on? ‣ Check your backups!
- 214. Maintenance ‣ OS updates. updates. Minor-, major-, distro ‣ Take a server out of the cluster to perform bigger updates. Reintegrate if afterwards. No or minor service impact. ‣ Ruby Updates. Ruby, Rails, Gems, ... ‣ Rotation and monitoring of logs. Bad things going on? ‣ Check your backups! ‣ ...
- 215. Fazit Ok, nice to know but ...
- 216. Fazit ... who the hell should to do all those things and additionally develop an application?
- 218. Thank you for your attention! Headquarter: Rails Enterprise Hosting: http://www.avarteq.de http://www.enterprise- rails.de Blog: http://ww.treibstofff.de Rails Hosting: http://www.railshoster.de
- 219. Resources http://www.twitter.com/railshoster
- 220. Resources ‣ Python Web Application Hosting http://hostingfu.com/article/python-web-application-hosting http://www.twitter.com/railshoster
- 221. Resources ‣ Python Web Application Hosting http://hostingfu.com/article/python-web-application-hosting ‣ Apache and mod_wsgi http://djangoinproduction.com/wiki/Apache_and_mod_wsgi http://www.twitter.com/railshoster
- 222. Resources ‣ Python Web Application Hosting http://hostingfu.com/article/python-web-application-hosting ‣ Apache and mod_wsgi http://djangoinproduction.com/wiki/Apache_and_mod_wsgi ‣ MySQL Cluster http://www.mysql.com/products/database/cluster/ http://www.twitter.com/railshoster
- 223. Resources ‣ Python Web Application Hosting http://hostingfu.com/article/python-web-application-hosting ‣ Apache and mod_wsgi http://djangoinproduction.com/wiki/Apache_and_mod_wsgi ‣ MySQL Cluster http://www.mysql.com/products/database/cluster/ ‣ Duplicity Backup Tool http://duplicity.nongnu.org/ http://www.twitter.com/railshoster
Editor's Notes
- #20: diff to non enterprise hosting
- #44: components -> influence on how to distribute your components to servers && how many && which servers to choose
- #45: components -> influence on how to distribute your components to servers && how many && which servers to choose
- #60: lets assume physical HW
- #61: lets assume physical HW
- #62: lets assume physical HW
- #76: We provide storage space on shared NetApps
- #77: We provide storage space on shared NetApps
- #78: We provide storage space on shared NetApps
- #79: We provide storage space on shared NetApps
- #81: We provide storage space on shared NetApps
- #82: We provide storage space on shared NetApps
- #83: We provide storage space on shared NetApps
- #84: We provide storage space on shared NetApps
- #85: We provide storage space on shared NetApps
- #86: wsgi = the way to go, updated frequently mod_python
- #87: wsgi = the way to go, updated frequently mod_python
- #88: wsgi = the way to go, updated frequently mod_python
- #89: wsgi = the way to go, updated frequently mod_python
- #106: distributed FS because multiple nodes write to a single block device.
- #107: distributed FS because multiple nodes write to a single block device.
- #108: distributed FS because multiple nodes write to a single block device.
- #109: distributed FS because multiple nodes write to a single block device.
- #110: distributed FS because multiple nodes write to a single block device.
- #111: distributed FS because multiple nodes write to a single block device.
- #113: We provide storage space on shared NetApps
- #114: We provide storage space on shared NetApps
- #115: We provide storage space on shared NetApps
- #116: We provide storage space on shared NetApps
- #117: We provide storage space on shared NetApps
- #119: Further focus on MySQL
- #120: Further focus on MySQL
- #121: Further focus on MySQL
- #122: Further focus on MySQL
- #123: Further focus on MySQL
- #126: Further focus on MySQL
- #127: Further focus on MySQL
- #128: Further focus on MySQL
- #129: Further focus on MySQL
- #130: Further focus on MySQL
- #131: Further focus on MySQL
- #132: Further focus on MySQL
- #133: Further focus on MySQL
- #134: Further focus on MySQL
- #135: Further focus on MySQL
- #137: Further focus on MySQL
- #138: Further focus on MySQL
- #139: Further focus on MySQL
- #140: Further focus on MySQL
- #141: Further focus on MySQL
- #142: Further focus on MySQL
- #143: Further focus on MySQL
- #144: Further focus on MySQL
- #146: Kurzer R&#xFC;ckblick -> Anwendung l&#xE4;uft jetzt.
- #147: Kurzer R&#xFC;ckblick -> Anwendung l&#xE4;uft jetzt.
- #148: Kurzer R&#xFC;ckblick -> Anwendung l&#xE4;uft jetzt.
- #149: Kurzer R&#xFC;ckblick -> Anwendung l&#xE4;uft jetzt.
- #150: Nicht immer ist es Server-Versagen. Hacker, Anwendungsfehler --> Datenverlust
- #162: HTTP+APP = Differentialdiagnose. Webserver OK, Anwendung nicht. Webserver down + APP Down. Gut einger. Monitoring = Gesundheitszustand des Clusters. Sonst &#xFC;bersicht verlieren.
- #163: HTTP+APP = Differentialdiagnose. Webserver OK, Anwendung nicht. Webserver down + APP Down. Gut einger. Monitoring = Gesundheitszustand des Clusters. Sonst &#xFC;bersicht verlieren.
- #164: HTTP+APP = Differentialdiagnose. Webserver OK, Anwendung nicht. Webserver down + APP Down. Gut einger. Monitoring = Gesundheitszustand des Clusters. Sonst &#xFC;bersicht verlieren.
- #165: HTTP+APP = Differentialdiagnose. Webserver OK, Anwendung nicht. Webserver down + APP Down. Gut einger. Monitoring = Gesundheitszustand des Clusters. Sonst &#xFC;bersicht verlieren.
- #166: HTTP+APP = Differentialdiagnose. Webserver OK, Anwendung nicht. Webserver down + APP Down. Gut einger. Monitoring = Gesundheitszustand des Clusters. Sonst &#xFC;bersicht verlieren.
- #167: HTTP+APP = Differentialdiagnose. Webserver OK, Anwendung nicht. Webserver down + APP Down. Gut einger. Monitoring = Gesundheitszustand des Clusters. Sonst &#xFC;bersicht verlieren.
- #170: Useraufuchs + Kurze Zusammenfassung: app l&#xE4;uft, backup + monitoring. soll auch so bleiben!
- #171: Useraufuchs + Kurze Zusammenfassung: app l&#xE4;uft, backup + monitoring. soll auch so bleiben!
- #172: Useraufuchs + Kurze Zusammenfassung: app l&#xE4;uft, backup + monitoring. soll auch so bleiben!
- #173: Useraufuchs + Kurze Zusammenfassung: app l&#xE4;uft, backup + monitoring. soll auch so bleiben!
- #175: Kernelupdates -> Downtime, Dis Update weil Support ausl&#xE4;uft Ruby/Rails in Abstimmung mit dem Kunden, Verantwortlichen definieren, der sich darum k&#xFC;mmert
- #176: Kernelupdates -> Downtime, Dis Update weil Support ausl&#xE4;uft Ruby/Rails in Abstimmung mit dem Kunden, Verantwortlichen definieren, der sich darum k&#xFC;mmert
- #177: Kernelupdates -> Downtime, Dis Update weil Support ausl&#xE4;uft Ruby/Rails in Abstimmung mit dem Kunden, Verantwortlichen definieren, der sich darum k&#xFC;mmert
- #178: Kernelupdates -> Downtime, Dis Update weil Support ausl&#xE4;uft Ruby/Rails in Abstimmung mit dem Kunden, Verantwortlichen definieren, der sich darum k&#xFC;mmert
- #179: Kernelupdates -> Downtime, Dis Update weil Support ausl&#xE4;uft Ruby/Rails in Abstimmung mit dem Kunden, Verantwortlichen definieren, der sich darum k&#xFC;mmert
- #180: Kernelupdates -> Downtime, Dis Update weil Support ausl&#xE4;uft Ruby/Rails in Abstimmung mit dem Kunden, Verantwortlichen definieren, der sich darum k&#xFC;mmert
- #183: Trennung der Verantwortlichkeiten, Hosting Team, Dev-Team, h&#xF6;here Produktivit&#xE4;t.