)]}'
{"/COMMIT_MSG":[{"author":{"_account_id":7847,"name":"Alistair Coles","email":"alistairncoles@gmail.com","username":"acoles"},"change_message_id":"54088d0b56617ca8c622f1690e21cee7e697e015","unresolved":false,"context_lines":[{"line_number":4,"context_line":"Commit:     Kota Tsuyuzaki \u003ctsuyuzaki.kota@lab.ntt.co.jp\u003e"},{"line_number":5,"context_line":"CommitDate: 2016-07-07 04:21:35 -0700"},{"line_number":6,"context_line":""},{"line_number":7,"context_line":"Add Pros/Cons table for global clusetr consideration"},{"line_number":8,"context_line":""},{"line_number":9,"context_line":"This comes from discussion in Bristol Hackathon (Feb 2016)."},{"line_number":10,"context_line":"Currently Swift has a couple of choices (Global Cluster and Container"}],"source_content_type":"text/x-gerrit-commit-message","patch_set":6,"id":"dada55a8_32876862","line":7,"updated":"2016-07-27 13:46:09.000000000","message":"s/clusetr/cluster/","commit_id":"bb627e7816c2e0cf2eba2f7105dba10995f531c1"},{"author":{"_account_id":4608,"name":"Kota Tsuyuzaki","email":"bloodeagle40234@gmail.com","username":"tsuyuzaki-kota"},"change_message_id":"6c9aa08e1c8345e4ad7adce3a15a3080f8a3e9f6","unresolved":false,"context_lines":[{"line_number":4,"context_line":"Commit:     Kota Tsuyuzaki \u003ctsuyuzaki.kota@lab.ntt.co.jp\u003e"},{"line_number":5,"context_line":"CommitDate: 2016-07-07 04:21:35 -0700"},{"line_number":6,"context_line":""},{"line_number":7,"context_line":"Add Pros/Cons table for global clusetr consideration"},{"line_number":8,"context_line":""},{"line_number":9,"context_line":"This comes from discussion in Bristol Hackathon (Feb 2016)."},{"line_number":10,"context_line":"Currently Swift has a couple of choices (Global Cluster and Container"}],"source_content_type":"text/x-gerrit-commit-message","patch_set":6,"id":"bacf61ea_2991df47","line":7,"in_reply_to":"dada55a8_32876862","updated":"2016-08-01 04:46:40.000000000","message":"Done","commit_id":"bb627e7816c2e0cf2eba2f7105dba10995f531c1"},{"author":{"_account_id":7847,"name":"Alistair Coles","email":"alistairncoles@gmail.com","username":"acoles"},"change_message_id":"54088d0b56617ca8c622f1690e21cee7e697e015","unresolved":false,"context_lines":[{"line_number":15,"context_line":"functionality fits their own use case."},{"line_number":16,"context_line":""},{"line_number":17,"context_line":"And, to be fairness with container-sync, this patch moves global"},{"line_number":18,"context_line":"cluster docs into overview_global_cluster.rst from admin_guide.rst."},{"line_number":19,"context_line":""},{"line_number":20,"context_line":"Change-Id: I624eb519503ae71dbc82245c33dab6e8637d0f8b"}],"source_content_type":"text/x-gerrit-commit-message","patch_set":6,"id":"dada55a8_dbff0032","line":18,"updated":"2016-07-27 13:46:09.000000000","message":"+1 that makes sense. That doc looks like it could be improved but that is not the scope of this change where you are just re-locating the text.","commit_id":"bb627e7816c2e0cf2eba2f7105dba10995f531c1"},{"author":{"_account_id":4608,"name":"Kota Tsuyuzaki","email":"bloodeagle40234@gmail.com","username":"tsuyuzaki-kota"},"change_message_id":"6c9aa08e1c8345e4ad7adce3a15a3080f8a3e9f6","unresolved":false,"context_lines":[{"line_number":15,"context_line":"functionality fits their own use case."},{"line_number":16,"context_line":""},{"line_number":17,"context_line":"And, to be fairness with container-sync, this patch moves global"},{"line_number":18,"context_line":"cluster docs into overview_global_cluster.rst from admin_guide.rst."},{"line_number":19,"context_line":""},{"line_number":20,"context_line":"Change-Id: I624eb519503ae71dbc82245c33dab6e8637d0f8b"}],"source_content_type":"text/x-gerrit-commit-message","patch_set":6,"id":"bacf61ea_499c135d","line":18,"in_reply_to":"dada55a8_dbff0032","updated":"2016-08-01 04:46:40.000000000","message":"So does that mean we should split up this patch to \"add docs\" and  \"relocating\"?","commit_id":"bb627e7816c2e0cf2eba2f7105dba10995f531c1"}],"doc/source/admin_guide.rst":[{"author":{"_account_id":1179,"name":"Clay Gerrard","email":"clay.gerrard@gmail.com","username":"clay-gerrard"},"change_message_id":"ba8441095d06124949bdf463179eccd0f7cf8c53","unresolved":false,"context_lines":[{"line_number":375,"context_line":"|      | - Cross container features/behaviors well known| - Allows to distribute the cluster management between teams              |"},{"line_number":376,"context_line":"|      | - Ops replication monitoring/throttling        |                                                                          |"},{"line_number":377,"context_line":"+------+------------------------------------------------+--------------------------------------------------------------------------+"},{"line_number":378,"context_line":"| Cons | - Region ring management shared ops            | - Current implementation sucks (is now working to fix)                   |"},{"line_number":379,"context_line":"|      | - Account/Container updates via WAN            | - User centric control may be a headache for the ops                     |"},{"line_number":380,"context_line":"|      | - All Accounts/Conttianers spans global spaces | - Cross container features isolated (need to be setup for each container)|"},{"line_number":381,"context_line":"|      | - No \"read repair/replicate\"                   | - Needs redundant replicas for each cluster                              |"}],"source_content_type":"text/x-rst","patch_set":2,"id":"5a18252c_671b6c1c","line":378,"updated":"2016-04-12 17:27:27.000000000","message":"lol, \"sucks\" is probably not the right word for the admin guide.\n\nMaybe we could try and quantify it?  \"Inefficient\" ?  I think the main hangup with container sync was that PUT/100 Continue support wasn\u0027t working - it\u0027s been worked around with a pre-flight HEAD I believe.","commit_id":"4f1796ebe8e0ed273425c72354127ce6252dcd3c"},{"author":{"_account_id":4608,"name":"Kota Tsuyuzaki","email":"bloodeagle40234@gmail.com","username":"tsuyuzaki-kota"},"change_message_id":"1806a965e55df57c803dcee4c596534cd03b48fb","unresolved":false,"context_lines":[{"line_number":375,"context_line":"|      | - Cross container features/behaviors well known| - Allows to distribute the cluster management between teams              |"},{"line_number":376,"context_line":"|      | - Ops replication monitoring/throttling        |                                                                          |"},{"line_number":377,"context_line":"+------+------------------------------------------------+--------------------------------------------------------------------------+"},{"line_number":378,"context_line":"| Cons | - Region ring management shared ops            | - Current implementation sucks (is now working to fix)                   |"},{"line_number":379,"context_line":"|      | - Account/Container updates via WAN            | - User centric control may be a headache for the ops                     |"},{"line_number":380,"context_line":"|      | - All Accounts/Conttianers spans global spaces | - Cross container features isolated (need to be setup for each container)|"},{"line_number":381,"context_line":"|      | - No \"read repair/replicate\"                   | - Needs redundant replicas for each cluster                              |"}],"source_content_type":"text/x-rst","patch_set":2,"id":"5a18252c_9d7d2bb4","line":378,"in_reply_to":"5a18252c_671b6c1c","updated":"2016-04-13 01:55:03.000000000","message":"bah, you are probably true. I just wrote down (almost copy and paste) from the photo of the whiteboard [1]. And I don\u0027t have the native (good) sense to translate to the public docs. \n\nIt might be just like \"still working in progress\"?\n\nAs you said in top comments, the progress is going to be changed absolutely. I think it\u0027s true that this docs might be already staled for items :/\n\n1: https://goo.gl/photos/ivzHDDhBGJmpVkc78","commit_id":"4f1796ebe8e0ed273425c72354127ce6252dcd3c"},{"author":{"_account_id":330,"name":"John Dickinson","email":"me@not.mn","username":"notmyname"},"change_message_id":"7af36b4e205f19d69ac6983768d8e6f2df89bb8b","unresolved":false,"context_lines":[{"line_number":410,"context_line":"Geographically Distributed Cluster Considerations"},{"line_number":411,"context_line":"-------------------------------------------------"},{"line_number":412,"context_line":""},{"line_number":413,"context_line":"When you are trying to deploy Swif cluster into geographically distributed"},{"line_number":414,"context_line":"location, you can choose a couple of options, :doc:`overview_global_cluster`"},{"line_number":415,"context_line":"(Global Cluster), :doc:`overview_container_sync` (Container Sync), to get"},{"line_number":416,"context_line":"the functionality."}],"source_content_type":"text/x-rst","patch_set":4,"id":"7aa08908_09a9e5df","line":413,"updated":"2016-06-06 16:29:34.000000000","message":"nit: Swif","commit_id":"6647a50a660cb83efff71bb86f0b26bb1b35b61a"},{"author":{"_account_id":4608,"name":"Kota Tsuyuzaki","email":"bloodeagle40234@gmail.com","username":"tsuyuzaki-kota"},"change_message_id":"6a13a9b03b871d2ce7d760bcae26becd04c5af8c","unresolved":false,"context_lines":[{"line_number":410,"context_line":"Geographically Distributed Cluster Considerations"},{"line_number":411,"context_line":"-------------------------------------------------"},{"line_number":412,"context_line":""},{"line_number":413,"context_line":"When you are trying to deploy Swif cluster into geographically distributed"},{"line_number":414,"context_line":"location, you can choose a couple of options, :doc:`overview_global_cluster`"},{"line_number":415,"context_line":"(Global Cluster), :doc:`overview_container_sync` (Container Sync), to get"},{"line_number":416,"context_line":"the functionality."}],"source_content_type":"text/x-rst","patch_set":4,"id":"7aa08908_4f21144d","line":413,"in_reply_to":"7aa08908_09a9e5df","updated":"2016-06-07 05:26:25.000000000","message":"Done","commit_id":"6647a50a660cb83efff71bb86f0b26bb1b35b61a"},{"author":{"_account_id":330,"name":"John Dickinson","email":"me@not.mn","username":"notmyname"},"change_message_id":"7af36b4e205f19d69ac6983768d8e6f2df89bb8b","unresolved":false,"context_lines":[{"line_number":430,"context_line":"| Cons | - Region ring management shared ops            | - User centric control may be a headache for the ops                     |"},{"line_number":431,"context_line":"|      | - Account/Container updates via WAN            | - Simple EC scheme                                                       |"},{"line_number":432,"context_line":"|      | - All Accounts/Conttianers spans global spaces | - Cross container features isolated (need to be setup for each container)|"},{"line_number":433,"context_line":"|      | - No \"read repair/replicate\"                   | - Needs redundant replicas for each cluster                              |"},{"line_number":434,"context_line":"+------+------------------------------------------------+--------------------------------------------------------------------------+"},{"line_number":435,"context_line":""},{"line_number":436,"context_line":"--------------------------------"}],"source_content_type":"text/x-rst","patch_set":4,"id":"7aa08908_a956d1cf","line":433,"updated":"2016-06-06 16:29:34.000000000","message":"what does \"No \u0027read repair/replicate\u0027\" mean?","commit_id":"6647a50a660cb83efff71bb86f0b26bb1b35b61a"},{"author":{"_account_id":4608,"name":"Kota Tsuyuzaki","email":"bloodeagle40234@gmail.com","username":"tsuyuzaki-kota"},"change_message_id":"6a13a9b03b871d2ce7d760bcae26becd04c5af8c","unresolved":false,"context_lines":[{"line_number":430,"context_line":"| Cons | - Region ring management shared ops            | - User centric control may be a headache for the ops                     |"},{"line_number":431,"context_line":"|      | - Account/Container updates via WAN            | - Simple EC scheme                                                       |"},{"line_number":432,"context_line":"|      | - All Accounts/Conttianers spans global spaces | - Cross container features isolated (need to be setup for each container)|"},{"line_number":433,"context_line":"|      | - No \"read repair/replicate\"                   | - Needs redundant replicas for each cluster                              |"},{"line_number":434,"context_line":"+------+------------------------------------------------+--------------------------------------------------------------------------+"},{"line_number":435,"context_line":""},{"line_number":436,"context_line":"--------------------------------"}],"source_content_type":"text/x-rst","patch_set":4,"id":"7aa08908_efe0c0fc","line":433,"in_reply_to":"7aa08908_a956d1cf","updated":"2016-06-07 05:26:25.000000000","message":"Good question but I don\u0027t have clear answer right now. That line was in the whiteboard[1] but the context was dropped from my memory.\n\nIt seems to mean Swift doesn\u0027t have read repair logic (i.e. when a misplaced data found at GET, trying to place it to correct location at the same time) but absolutely it\u0027s a design of *Swift* itself and container-sync also doesn\u0027t aim to do that.\n\nI think it\u0027s better to get rid of the line above for next patch, thanks for heading up.\n\n1: https://goo.gl/photos/ivzHDDhBGJmpVkc78","commit_id":"6647a50a660cb83efff71bb86f0b26bb1b35b61a"},{"author":{"_account_id":7847,"name":"Alistair Coles","email":"alistairncoles@gmail.com","username":"acoles"},"change_message_id":"54088d0b56617ca8c622f1690e21cee7e697e015","unresolved":false,"context_lines":[{"line_number":501,"context_line":"-------------------------------------------------"},{"line_number":502,"context_line":""},{"line_number":503,"context_line":"When you are trying to deploy Swift cluster into geographically distributed"},{"line_number":504,"context_line":"location, you can choose a couple of options, :doc:`overview_global_cluster`"},{"line_number":505,"context_line":"(Global Cluster), :doc:`overview_container_sync` (Container Sync), to get"},{"line_number":506,"context_line":"the functionality."},{"line_number":507,"context_line":""}],"source_content_type":"text/x-rst","patch_set":6,"id":"dada55a8_d59876c2","line":504,"range":{"start_line":504,"start_character":46,"end_line":504,"end_character":76},"updated":"2016-07-27 13:46:09.000000000","message":"(see comment in other file)\n\nThis renders as Geographically Distributed Clusters which is confusing w.r.t. the title of this section, and this section being about both features.\n\nIMHO the title of this section is correct and the title of the Global Cluster overview doc should be \"Global Clusters\"","commit_id":"bb627e7816c2e0cf2eba2f7105dba10995f531c1"},{"author":{"_account_id":4608,"name":"Kota Tsuyuzaki","email":"bloodeagle40234@gmail.com","username":"tsuyuzaki-kota"},"change_message_id":"6c9aa08e1c8345e4ad7adce3a15a3080f8a3e9f6","unresolved":false,"context_lines":[{"line_number":501,"context_line":"-------------------------------------------------"},{"line_number":502,"context_line":""},{"line_number":503,"context_line":"When you are trying to deploy Swift cluster into geographically distributed"},{"line_number":504,"context_line":"location, you can choose a couple of options, :doc:`overview_global_cluster`"},{"line_number":505,"context_line":"(Global Cluster), :doc:`overview_container_sync` (Container Sync), to get"},{"line_number":506,"context_line":"the functionality."},{"line_number":507,"context_line":""}],"source_content_type":"text/x-rst","patch_set":6,"id":"bacf61ea_c9bf63a8","line":504,"range":{"start_line":504,"start_character":46,"end_line":504,"end_character":76},"in_reply_to":"dada55a8_d59876c2","updated":"2016-08-01 04:46:40.000000000","message":"Fair enough. Will do that.","commit_id":"bb627e7816c2e0cf2eba2f7105dba10995f531c1"},{"author":{"_account_id":7847,"name":"Alistair Coles","email":"alistairncoles@gmail.com","username":"acoles"},"change_message_id":"54088d0b56617ca8c622f1690e21cee7e697e015","unresolved":false,"context_lines":[{"line_number":521,"context_line":"|      | - Account/Container updates via WAN            | - Simple EC scheme                                                       |"},{"line_number":522,"context_line":"|      | - All Accounts/Conttianers spans global spaces | - Cross container features isolated (need to be setup for each container)|"},{"line_number":523,"context_line":"|      |                                                | - Needs redundant replicas for each cluster                              |"},{"line_number":524,"context_line":"+------+------------------------------------------------+--------------------------------------------------------------------------+"},{"line_number":525,"context_line":""},{"line_number":526,"context_line":"--------------------------------"},{"line_number":527,"context_line":"Cluster Telemetry and Monitoring"}],"source_content_type":"text/x-rst","patch_set":6,"id":"dada55a8_b865f7e1","line":524,"updated":"2016-07-27 13:46:09.000000000","message":"I feel that the meaning and/or significance of some of the bullet points in this table may not be immediately obvious to a sys admin who is getting up top speed on these features.\n\ne.g. \"Optimistic WAN get\", \"Simple EC scheme\"\n\nI wonder if a less structured discussion may be more helpful than a table which by necessity requires brevity.","commit_id":"bb627e7816c2e0cf2eba2f7105dba10995f531c1"},{"author":{"_account_id":4608,"name":"Kota Tsuyuzaki","email":"bloodeagle40234@gmail.com","username":"tsuyuzaki-kota"},"change_message_id":"6c9aa08e1c8345e4ad7adce3a15a3080f8a3e9f6","unresolved":false,"context_lines":[{"line_number":521,"context_line":"|      | - Account/Container updates via WAN            | - Simple EC scheme                                                       |"},{"line_number":522,"context_line":"|      | - All Accounts/Conttianers spans global spaces | - Cross container features isolated (need to be setup for each container)|"},{"line_number":523,"context_line":"|      |                                                | - Needs redundant replicas for each cluster                              |"},{"line_number":524,"context_line":"+------+------------------------------------------------+--------------------------------------------------------------------------+"},{"line_number":525,"context_line":""},{"line_number":526,"context_line":"--------------------------------"},{"line_number":527,"context_line":"Cluster Telemetry and Monitoring"}],"source_content_type":"text/x-rst","patch_set":6,"id":"bacf61ea_e947c7b9","line":524,"in_reply_to":"dada55a8_b865f7e1","updated":"2016-08-01 04:46:40.000000000","message":"Good call. At fist, I thought the table could be easy to compare them quickly expecting the readers can refer for each docs if they needs more details but exactly it could be complicated for readers. However, if we add structured discussion here, I\u0027m wondering it may be hard to keep consistent (or no duplication) between here and each doc pages.\n\nIf i will try to add the structured ones here, it seems to be likely:\n\n Global Cluster\n \u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\n\n Pros\n ----\n sentences \n ...\n\n\nPerhaps, we should add them to each docs page (i.e. overview_global_cluster and overview_container_sync)? Is it easy to find?","commit_id":"bb627e7816c2e0cf2eba2f7105dba10995f531c1"},{"author":{"_account_id":7847,"name":"Alistair Coles","email":"alistairncoles@gmail.com","username":"acoles"},"change_message_id":"1ec2eeb292cb57a3eb4052e61a6d4724d1917559","unresolved":false,"context_lines":[{"line_number":501,"context_line":"-----------------------------------------------"},{"line_number":502,"context_line":""},{"line_number":503,"context_line":"Swift provides two features that may be used to distribute replicas of objects"},{"line_number":504,"context_line":"across multiple geographically distributed data-centres: with"},{"line_number":505,"context_line":":doc:`overview_global_cluster` object replicas may be dispersed across devices"},{"line_number":506,"context_line":"from different data-centres by using `regions` in ring device descriptors; with"},{"line_number":507,"context_line":":doc:`overview_container_sync` objects may be copied between independent Swift"}],"source_content_type":"text/x-rst","patch_set":8,"id":"3ac371cc_a107b7c8","line":504,"updated":"2016-08-17 11:51:57.000000000","message":"hmm, I guess an American might prefer a global s/centre/center/ :)","commit_id":"55405b04a49f7def8a3854a94e335d5e359e7025"},{"author":{"_account_id":7847,"name":"Alistair Coles","email":"alistairncoles@gmail.com","username":"acoles"},"change_message_id":"1ec2eeb292cb57a3eb4052e61a6d4724d1917559","unresolved":false,"context_lines":[{"line_number":528,"context_line":"     Container Sync requires the end user to ensure that all required"},{"line_number":529,"context_line":"     containers are sync\u0027d for these features to work in all data-centres."},{"line_number":530,"context_line":""},{"line_number":531,"context_line":"  #. Global Clusters makes object available for GET or HEAD requests in both"},{"line_number":532,"context_line":"     data-centre even if a replica of the object has not yet been"},{"line_number":533,"context_line":"     asynchronously migrated between data-centres, by forwarding requests"},{"line_number":534,"context_line":"     between data-centres. Container Sync is unable to serve requests for an"}],"source_content_type":"text/x-rst","patch_set":8,"id":"3ac371cc_c15bd3f4","line":531,"updated":"2016-08-17 11:51:57.000000000","message":"s/object/objects/","commit_id":"55405b04a49f7def8a3854a94e335d5e359e7025"},{"author":{"_account_id":7847,"name":"Alistair Coles","email":"alistairncoles@gmail.com","username":"acoles"},"change_message_id":"1ec2eeb292cb57a3eb4052e61a6d4724d1917559","unresolved":false,"context_lines":[{"line_number":529,"context_line":"     containers are sync\u0027d for these features to work in all data-centres."},{"line_number":530,"context_line":""},{"line_number":531,"context_line":"  #. Global Clusters makes object available for GET or HEAD requests in both"},{"line_number":532,"context_line":"     data-centre even if a replica of the object has not yet been"},{"line_number":533,"context_line":"     asynchronously migrated between data-centres, by forwarding requests"},{"line_number":534,"context_line":"     between data-centres. Container Sync is unable to serve requests for an"},{"line_number":535,"context_line":"     object in a particular data-centre until the asynchronous sync process has"}],"source_content_type":"text/x-rst","patch_set":8,"id":"3ac371cc_a12e9740","line":532,"updated":"2016-08-17 11:51:57.000000000","message":"s/centre/centres/","commit_id":"55405b04a49f7def8a3854a94e335d5e359e7025"},{"author":{"_account_id":7847,"name":"Alistair Coles","email":"alistairncoles@gmail.com","username":"acoles"},"change_message_id":"1ec2eeb292cb57a3eb4052e61a6d4724d1917559","unresolved":false,"context_lines":[{"line_number":545,"context_line":""},{"line_number":546,"context_line":"  #. Global Clusters execute all account/container metadata updates"},{"line_number":547,"context_line":"     synchronously to account/container replicas in all data-centres, which may"},{"line_number":548,"context_line":"     incur delays when making updates across WANs. Container Sync only copies"},{"line_number":549,"context_line":""},{"line_number":550,"context_line":""},{"line_number":551,"context_line":"~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~"}],"source_content_type":"text/x-rst","patch_set":8,"id":"3ac371cc_e1808ff9","line":548,"updated":"2016-08-17 11:51:57.000000000","message":"oops, looks like cut and paste fail. more lines from [1]:\n\n  +     objects between data-centres and all Swift internal traffic is\n+     confined to each data-centre.\n+\n+  #. Global Clusters does not yet guarantee the availability of objects stored\n+     in Erasure Coded policies when one data-centre is offline. With Container\n+     Sync the availability of objects in each data-centre is independent of the\n+     state of other data-centres. Container Sync also allows objects to be\n+     stored using different policy types in different data-centres.\n\n[1] https://gist.github.com/alistairncoles/7352f18736b270420fe6796168cfcda5","commit_id":"55405b04a49f7def8a3854a94e335d5e359e7025"}],"doc/source/overview_global_cluster.rst":[{"author":{"_account_id":7847,"name":"Alistair Coles","email":"alistairncoles@gmail.com","username":"acoles"},"change_message_id":"54088d0b56617ca8c622f1690e21cee7e697e015","unresolved":false,"context_lines":[{"line_number":1,"context_line":"\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d"},{"line_number":2,"context_line":"Geographically Distributed Clusters"},{"line_number":3,"context_line":"\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d"},{"line_number":4,"context_line":""},{"line_number":5,"context_line":"--------"}],"source_content_type":"text/x-rst","patch_set":6,"id":"dada55a8_951efe5c","line":2,"updated":"2016-07-27 13:46:09.000000000","message":"We need to choose one name for this feature and stick with it.\n\nIn admin guide we now have Geographically Distributed Clusters considerations which covers both global clusters and container sync.\n\nHere the title is \"Geographically Distributed Clusters\" but then we have a section \"Configuring Global Cluster\".\n\nSo should this document be titled Global Clusters ?","commit_id":"bb627e7816c2e0cf2eba2f7105dba10995f531c1"},{"author":{"_account_id":4608,"name":"Kota Tsuyuzaki","email":"bloodeagle40234@gmail.com","username":"tsuyuzaki-kota"},"change_message_id":"6c9aa08e1c8345e4ad7adce3a15a3080f8a3e9f6","unresolved":false,"context_lines":[{"line_number":1,"context_line":"\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d"},{"line_number":2,"context_line":"Geographically Distributed Clusters"},{"line_number":3,"context_line":"\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d"},{"line_number":4,"context_line":""},{"line_number":5,"context_line":"--------"}],"source_content_type":"text/x-rst","patch_set":6,"id":"bacf61ea_a9faef72","line":2,"in_reply_to":"dada55a8_951efe5c","updated":"2016-08-01 04:46:40.000000000","message":"Exactly, that inconsistency would be worse for readers to know what\u0027s addressed in this docs. I will change the title to Global Clusters here.","commit_id":"bb627e7816c2e0cf2eba2f7105dba10995f531c1"},{"author":{"_account_id":7233,"name":"Matthew Oliver","email":"matt@oliver.net.au","username":"mattoliverau"},"change_message_id":"320c9181526f55c65154ca4af6f4359e37741ebf","unresolved":false,"context_lines":[{"line_number":130,"context_line":"distribution will get worse and worse over time as objects pile up in SF."},{"line_number":131,"context_line":"If this happens, it is recommended to disable write_affinity and simply let"},{"line_number":132,"context_line":"object PUTs traverse the WAN link, as that will naturally limit the"},{"line_number":133,"context_line":"object growth rate to what your WAN link can handle."}],"source_content_type":"text/x-rst","patch_set":9,"id":"1ac06dbe_6b00b034","line":133,"updated":"2016-08-24 04:31:36.000000000","message":"Should we also talk about rsync_compress in the replicators here? as they are useful for multi-region global clusters?","commit_id":"dfa5523d8c729dd44f79015d8ad3d0260e7f86c3"}]}