)]}'
{"/PATCHSET_LEVEL":[{"author":{"_account_id":7118,"name":"Ian Wienand","email":"iwienand@redhat.com","username":"iwienand"},"change_message_id":"e8feea649026ef252eac3b08737614772d99ee6a","unresolved":false,"context_lines":[],"source_content_type":"","patch_set":2,"id":"c9f9f5fe_be23728f","updated":"2021-10-26 20:59:16.000000000","message":"LGTM.  I feel like we have good patterns in place for managing db containers and backups ","commit_id":"9bace208224c05ebb324b49a10e9ec75932fb04d"}],"specs/mailman3.rst":[{"author":{"_account_id":4146,"name":"Clark Boylan","email":"cboylan@sapwetik.org","username":"cboylan"},"change_message_id":"416307967a377351e27868cc51f507f78e193e5c","unresolved":true,"context_lines":[{"line_number":82,"context_line":""},{"line_number":83,"context_line":"We\u0027ll write Ansible playbooks and Zuul jobs to deploy the three"},{"line_number":84,"context_line":"Mailman service containers along with an accompanying MySQLd or"},{"line_number":85,"context_line":"MariaDB container, and then configure system packages of Apache and"},{"line_number":86,"context_line":"Exim to support these. The Mailman container documentation provides"},{"line_number":87,"context_line":"examples for configuring `Exim"},{"line_number":88,"context_line":"\u003chttps://asynchronous.in/docker-mailman/#exim4\u003e`_ and `Nginx"}],"source_content_type":"text/x-rst","patch_set":1,"id":"7c61fd86_d367c7cf","line":85,"range":{"start_line":85,"start_character":0,"end_line":85,"end_character":17},"updated":"2021-10-08 18:49:40.000000000","message":"I think we should just go ahead and say we\u0027ll use mariadb since all of our other containerized dbs are maria I think (gerrit and gitea at least).","commit_id":"f05d84ac32d34a748b2c3a1c830ced0490cfb507"},{"author":{"_account_id":5263,"name":"Jeremy Stanley","display_name":"fungi","email":"fungi@yuggoth.org","username":"fungi","status":"missing, presumed fed"},"change_message_id":"3ed7a8847d55f8de64a57311adaf7d691656d4c5","unresolved":false,"context_lines":[{"line_number":82,"context_line":""},{"line_number":83,"context_line":"We\u0027ll write Ansible playbooks and Zuul jobs to deploy the three"},{"line_number":84,"context_line":"Mailman service containers along with an accompanying MySQLd or"},{"line_number":85,"context_line":"MariaDB container, and then configure system packages of Apache and"},{"line_number":86,"context_line":"Exim to support these. The Mailman container documentation provides"},{"line_number":87,"context_line":"examples for configuring `Exim"},{"line_number":88,"context_line":"\u003chttps://asynchronous.in/docker-mailman/#exim4\u003e`_ and `Nginx"}],"source_content_type":"text/x-rst","patch_set":1,"id":"90337efd_12b63e12","line":85,"updated":"2021-10-10 17:46:12.000000000","message":"Sure, I just couldn\u0027t remember which we\u0027d been using. I\u0027ll firm this up.","commit_id":"f05d84ac32d34a748b2c3a1c830ced0490cfb507"},{"author":{"_account_id":5263,"name":"Jeremy Stanley","display_name":"fungi","email":"fungi@yuggoth.org","username":"fungi","status":"missing, presumed fed"},"change_message_id":"3ed7a8847d55f8de64a57311adaf7d691656d4c5","unresolved":false,"context_lines":[{"line_number":110,"context_line":"any existing bookmarks or historical hyperlinks. If necessary, we"},{"line_number":111,"context_line":"can perform this migration one domain at a time, since the official"},{"line_number":112,"context_line":"cut over happens with a corresponding change for that virtual"},{"line_number":113,"context_line":"domain\u0027s hostname in the public DNS."},{"line_number":114,"context_line":""},{"line_number":115,"context_line":"After all migrations are completed and any remaining issues sorted,"},{"line_number":116,"context_line":"the old mailing list servers will be imaged for posterity and then"}],"source_content_type":"text/x-rst","patch_set":1,"id":"a1f78cb3_bca00aa6","line":113,"updated":"2021-10-10 17:46:12.000000000","message":"Sounds good, I\u0027ll make it that.","commit_id":"f05d84ac32d34a748b2c3a1c830ced0490cfb507"},{"author":{"_account_id":4146,"name":"Clark Boylan","email":"cboylan@sapwetik.org","username":"cboylan"},"change_message_id":"416307967a377351e27868cc51f507f78e193e5c","unresolved":true,"context_lines":[{"line_number":110,"context_line":"any existing bookmarks or historical hyperlinks. If necessary, we"},{"line_number":111,"context_line":"can perform this migration one domain at a time, since the official"},{"line_number":112,"context_line":"cut over happens with a corresponding change for that virtual"},{"line_number":113,"context_line":"domain\u0027s hostname in the public DNS."},{"line_number":114,"context_line":""},{"line_number":115,"context_line":"After all migrations are completed and any remaining issues sorted,"},{"line_number":116,"context_line":"the old mailing list servers will be imaged for posterity and then"}],"source_content_type":"text/x-rst","patch_set":1,"id":"e8b61b4d_2ccacf58","line":113,"updated":"2021-10-08 18:49:40.000000000","message":"We might want to go ahead and say opendev will convert first with its lists? Then migrate everyone else in a little while later when we are happy with the results?","commit_id":"f05d84ac32d34a748b2c3a1c830ced0490cfb507"},{"author":{"_account_id":5263,"name":"Jeremy Stanley","display_name":"fungi","email":"fungi@yuggoth.org","username":"fungi","status":"missing, presumed fed"},"change_message_id":"3ed7a8847d55f8de64a57311adaf7d691656d4c5","unresolved":false,"context_lines":[{"line_number":148,"context_line":""},{"line_number":149,"context_line":"Primary assignee:"},{"line_number":150,"context_line":""},{"line_number":151,"context_line":"- fungi"},{"line_number":152,"context_line":""},{"line_number":153,"context_line":"Gerrit Topic"},{"line_number":154,"context_line":"------------"}],"source_content_type":"text/x-rst","patch_set":1,"id":"ebcfb236_0647169e","line":151,"updated":"2021-10-10 17:46:12.000000000","message":"Can do! The more the merrier.","commit_id":"f05d84ac32d34a748b2c3a1c830ced0490cfb507"},{"author":{"_account_id":4146,"name":"Clark Boylan","email":"cboylan@sapwetik.org","username":"cboylan"},"change_message_id":"416307967a377351e27868cc51f507f78e193e5c","unresolved":true,"context_lines":[{"line_number":148,"context_line":""},{"line_number":149,"context_line":"Primary assignee:"},{"line_number":150,"context_line":""},{"line_number":151,"context_line":"- fungi"},{"line_number":152,"context_line":""},{"line_number":153,"context_line":"Gerrit Topic"},{"line_number":154,"context_line":"------------"}],"source_content_type":"text/x-rst","patch_set":1,"id":"8416f361_e46d921c","line":151,"updated":"2021-10-08 18:49:40.000000000","message":"Feel free to add clarkb to this list too.","commit_id":"f05d84ac32d34a748b2c3a1c830ced0490cfb507"},{"author":{"_account_id":13252,"name":"Dr. Jens Harbott","display_name":"Jens Harbott (frickler)","email":"frickler@offenerstapel.de","username":"jrosenboom"},"change_message_id":"f737b48ae823aaecf33edb61a230142ebfd83e28","unresolved":true,"context_lines":[{"line_number":35,"context_line":"- Mailman 3, which was experimental for a very long time but has"},{"line_number":36,"context_line":"  become more stable and essentially reached feature parity with its"},{"line_number":37,"context_line":"  predecessor some years ago now, is functionally the same"},{"line_number":38,"context_line":"  internally as far as exhcanging E-mail messages is concerned, but"},{"line_number":39,"context_line":"  has expanded substantially on its Web interface with a new"},{"line_number":40,"context_line":"  Postorius list management interface and Hyperkitty interactive Web"},{"line_number":41,"context_line":"  archive."}],"source_content_type":"text/x-rst","patch_set":2,"id":"d6554f43_774783c9","line":38,"range":{"start_line":38,"start_character":23,"end_line":38,"end_character":33},"updated":"2021-11-03 07:15:36.000000000","message":"Nit: exchanging","commit_id":"9bace208224c05ebb324b49a10e9ec75932fb04d"},{"author":{"_account_id":5263,"name":"Jeremy Stanley","display_name":"fungi","email":"fungi@yuggoth.org","username":"fungi","status":"missing, presumed fed"},"change_message_id":"fb2e434b6e5b623c8da0def1b6291868693587ef","unresolved":false,"context_lines":[{"line_number":35,"context_line":"- Mailman 3, which was experimental for a very long time but has"},{"line_number":36,"context_line":"  become more stable and essentially reached feature parity with its"},{"line_number":37,"context_line":"  predecessor some years ago now, is functionally the same"},{"line_number":38,"context_line":"  internally as far as exhcanging E-mail messages is concerned, but"},{"line_number":39,"context_line":"  has expanded substantially on its Web interface with a new"},{"line_number":40,"context_line":"  Postorius list management interface and Hyperkitty interactive Web"},{"line_number":41,"context_line":"  archive."}],"source_content_type":"text/x-rst","patch_set":2,"id":"89cc32cd_1b56c413","line":38,"updated":"2021-11-03 15:19:23.000000000","message":"Oops, thanks! I\u0027ll fix that.","commit_id":"9bace208224c05ebb324b49a10e9ec75932fb04d"},{"author":{"_account_id":5263,"name":"Jeremy Stanley","display_name":"fungi","email":"fungi@yuggoth.org","username":"fungi","status":"missing, presumed fed"},"change_message_id":"fb2e434b6e5b623c8da0def1b6291868693587ef","unresolved":false,"context_lines":[{"line_number":44,"context_line":"have occurred in the ways people interface with E-mail and hold"},{"line_number":45,"context_line":"group discussions. The mass freemail providers have become"},{"line_number":46,"context_line":"entrenched, ubiquitous, and emboldened to begin breaking"},{"line_number":47,"context_line":"long-established protocol standards. Portable handleld devices are"},{"line_number":48,"context_line":"becoming the primary way many people interact. Web forums have"},{"line_number":49,"context_line":"morphed into social media platforms, and their popularity has"},{"line_number":50,"context_line":"further altered expectations of the next generation of computer"}],"source_content_type":"text/x-rst","patch_set":2,"id":"fdc5ef74_9b1907bd","line":47,"updated":"2021-11-03 15:19:23.000000000","message":"Fixing that too, thanks again.","commit_id":"9bace208224c05ebb324b49a10e9ec75932fb04d"},{"author":{"_account_id":13252,"name":"Dr. Jens Harbott","display_name":"Jens Harbott (frickler)","email":"frickler@offenerstapel.de","username":"jrosenboom"},"change_message_id":"f737b48ae823aaecf33edb61a230142ebfd83e28","unresolved":true,"context_lines":[{"line_number":44,"context_line":"have occurred in the ways people interface with E-mail and hold"},{"line_number":45,"context_line":"group discussions. The mass freemail providers have become"},{"line_number":46,"context_line":"entrenched, ubiquitous, and emboldened to begin breaking"},{"line_number":47,"context_line":"long-established protocol standards. Portable handleld devices are"},{"line_number":48,"context_line":"becoming the primary way many people interact. Web forums have"},{"line_number":49,"context_line":"morphed into social media platforms, and their popularity has"},{"line_number":50,"context_line":"further altered expectations of the next generation of computer"}],"source_content_type":"text/x-rst","patch_set":2,"id":"4994bb58_79e7fa70","line":47,"range":{"start_line":47,"start_character":46,"end_line":47,"end_character":54},"updated":"2021-11-03 07:15:36.000000000","message":"handheld","commit_id":"9bace208224c05ebb324b49a10e9ec75932fb04d"},{"author":{"_account_id":7118,"name":"Ian Wienand","email":"iwienand@redhat.com","username":"iwienand"},"change_message_id":"e8feea649026ef252eac3b08737614772d99ee6a","unresolved":true,"context_lines":[{"line_number":51,"context_line":"users. At the same time, these shifts in behavior have driven"},{"line_number":52,"context_line":"changes in how Web applications are developed and delivered."},{"line_number":53,"context_line":""},{"line_number":54,"context_line":"Mailman 3\u0027s Hyperkitty archive interface is essentially a bolt-on"},{"line_number":55,"context_line":"Web forum frontend, with more advanced authentication and some"},{"line_number":56,"context_line":"(mis?)features borrowed from popular social media. Its Postorius"},{"line_number":57,"context_line":"list management interface uses the same new authentication"},{"line_number":58,"context_line":"subsystems, having distinct per-user accounts and role-based access"},{"line_number":59,"context_line":"controls instead of the shared password model used by MM2. Both of"}],"source_content_type":"text/x-rst","patch_set":2,"id":"227c9459_5f6f3bda","line":56,"range":{"start_line":54,"start_character":0,"end_line":56,"end_character":50},"updated":"2021-10-26 20:59:16.000000000","message":"Having an account here is an optional thing, right?\n\nSo once this is up, everyone will be subscribed to their old lists; but if they want to make use of the hyperkitty interface they need to do a separate sign-up step in that application?  and then they can manage their subscriptions  (which we have migrated for them) interactively from hyperkitty because it\u0027s all keyed together by email-address?","commit_id":"9bace208224c05ebb324b49a10e9ec75932fb04d"},{"author":{"_account_id":5263,"name":"Jeremy Stanley","display_name":"fungi","email":"fungi@yuggoth.org","username":"fungi","status":"missing, presumed fed"},"change_message_id":"fb2e434b6e5b623c8da0def1b6291868693587ef","unresolved":false,"context_lines":[{"line_number":53,"context_line":""},{"line_number":54,"context_line":"Mailman 3\u0027s Hyperkitty archive interface is essentially a bolt-on"},{"line_number":55,"context_line":"Web forum frontend, with more advanced authentication and some"},{"line_number":56,"context_line":"(mis?)features borrowed from popular social media. Its Postorius"},{"line_number":57,"context_line":"list management interface uses the same new authentication"},{"line_number":58,"context_line":"subsystems, having distinct per-user accounts and role-based access"},{"line_number":59,"context_line":"controls instead of the shared password model used by MM2. Both of"}],"source_content_type":"text/x-rst","patch_set":2,"id":"8fcb1376_46ad44dd","line":56,"updated":"2021-11-03 15:19:23.000000000","message":"It relies on Django, so can use Django plugins for other authentication mechanisms and external account identifiers. I would consider that out of scope for the initial implementation since we don\u0027t have it in MM2 and lack any means of easily mapping existing users to external IDs (but they could in theory do it themselves once we add such a feature). I\u0027ll add a note about it.","commit_id":"9bace208224c05ebb324b49a10e9ec75932fb04d"},{"author":{"_account_id":5263,"name":"Jeremy Stanley","display_name":"fungi","email":"fungi@yuggoth.org","username":"fungi","status":"missing, presumed fed"},"change_message_id":"2c5824a77569020276bb4f3485022e60285282dc","unresolved":false,"context_lines":[{"line_number":53,"context_line":""},{"line_number":54,"context_line":"Mailman 3\u0027s Hyperkitty archive interface is essentially a bolt-on"},{"line_number":55,"context_line":"Web forum frontend, with more advanced authentication and some"},{"line_number":56,"context_line":"(mis?)features borrowed from popular social media. Its Postorius"},{"line_number":57,"context_line":"list management interface uses the same new authentication"},{"line_number":58,"context_line":"subsystems, having distinct per-user accounts and role-based access"},{"line_number":59,"context_line":"controls instead of the shared password model used by MM2. Both of"}],"source_content_type":"text/x-rst","patch_set":2,"id":"37201c6f_ca1951a3","line":56,"updated":"2021-10-26 21:50:17.000000000","message":"Yes, accounts are for managing preferences and also for using the Web UI to post to the lists instead of E-mail if that\u0027s your bag (and for things like list moderation and the admin configuration interface). Technically all subscribers have \"accounts\" I think (just like in the version we\u0027re using) but you don\u0027t need to authenticate to send to the lists via SMTP.\n\nIt\u0027ll definitely need some migration testing, but I expect existing subscribers will use the password reset feature if they need to get access to their account, while new subscribers will create an account when subscribing. However unlike in our version, your account spans all the lists on the server (across multiple domains even).","commit_id":"9bace208224c05ebb324b49a10e9ec75932fb04d"},{"author":{"_account_id":13252,"name":"Dr. Jens Harbott","display_name":"Jens Harbott (frickler)","email":"frickler@offenerstapel.de","username":"jrosenboom"},"change_message_id":"f737b48ae823aaecf33edb61a230142ebfd83e28","unresolved":false,"context_lines":[{"line_number":53,"context_line":""},{"line_number":54,"context_line":"Mailman 3\u0027s Hyperkitty archive interface is essentially a bolt-on"},{"line_number":55,"context_line":"Web forum frontend, with more advanced authentication and some"},{"line_number":56,"context_line":"(mis?)features borrowed from popular social media. Its Postorius"},{"line_number":57,"context_line":"list management interface uses the same new authentication"},{"line_number":58,"context_line":"subsystems, having distinct per-user accounts and role-based access"},{"line_number":59,"context_line":"controls instead of the shared password model used by MM2. Both of"}],"source_content_type":"text/x-rst","patch_set":2,"id":"0726a767_73e16f9d","line":56,"in_reply_to":"37201c6f_ca1951a3","updated":"2021-11-03 07:15:36.000000000","message":"Maybe not relevant for the initial migration, but is there an option to rely on some external authentication that might unify our different account setups or is this necessarily a world of its own just for mailman?","commit_id":"9bace208224c05ebb324b49a10e9ec75932fb04d"},{"author":{"_account_id":13252,"name":"Dr. Jens Harbott","display_name":"Jens Harbott (frickler)","email":"frickler@offenerstapel.de","username":"jrosenboom"},"change_message_id":"f737b48ae823aaecf33edb61a230142ebfd83e28","unresolved":true,"context_lines":[{"line_number":82,"context_line":""},{"line_number":83,"context_line":"We\u0027ll write Ansible playbooks and Zuul jobs to deploy the three"},{"line_number":84,"context_line":"Mailman service containers along with an accompanying MariaDB"},{"line_number":85,"context_line":"container, and then configure system packages of Apache and Exim to"},{"line_number":86,"context_line":"support these. The Mailman container documentation provides examples"},{"line_number":87,"context_line":"for configuring `Exim"},{"line_number":88,"context_line":"\u003chttps://asynchronous.in/docker-mailman/#exim4\u003e`_ and `Nginx"}],"source_content_type":"text/x-rst","patch_set":2,"id":"13541c60_5cec265d","line":85,"updated":"2021-11-03 07:15:36.000000000","message":"Did you consider running the within containers, too? Would be nice to at least have a short sentence about why to prefer using system pkgs here.\n\nAlso, will the base OS be Ubuntu 20.04 for now or might it be feasible to develop on 22.04 prerelease directly in order to skip one update cycle?","commit_id":"9bace208224c05ebb324b49a10e9ec75932fb04d"},{"author":{"_account_id":5263,"name":"Jeremy Stanley","display_name":"fungi","email":"fungi@yuggoth.org","username":"fungi","status":"missing, presumed fed"},"change_message_id":"fb2e434b6e5b623c8da0def1b6291868693587ef","unresolved":false,"context_lines":[{"line_number":82,"context_line":""},{"line_number":83,"context_line":"We\u0027ll write Ansible playbooks and Zuul jobs to deploy the three"},{"line_number":84,"context_line":"Mailman service containers along with an accompanying MariaDB"},{"line_number":85,"context_line":"container, and then configure system packages of Apache and Exim to"},{"line_number":86,"context_line":"support these. The Mailman container documentation provides examples"},{"line_number":87,"context_line":"for configuring `Exim"},{"line_number":88,"context_line":"\u003chttps://asynchronous.in/docker-mailman/#exim4\u003e`_ and `Nginx"}],"source_content_type":"text/x-rst","patch_set":2,"id":"d70fa724_bd0a76bb","line":85,"updated":"2021-11-03 15:19:23.000000000","message":"I picked Apache and Exim from distro packages because that\u0027s consistent with how we\u0027re running our other services (e.g. Gerrit), and don\u0027t personally see the benefit of running them from containers when they don\u0027t really change significantly from one distro version to the next anyway. I\u0027ll update to mention it.\n\nI expect we\u0027ll build on whatever our default distribution is when work begins, which is likely to be Ubuntu 20.04 LTS if we\u0027re starting on it in the next few weeks (trying to chase 20.04 LTS 6 months before it releases could be painful). I\u0027ll mention 20.04 LTS or current default in the servers section below.","commit_id":"9bace208224c05ebb324b49a10e9ec75932fb04d"},{"author":{"_account_id":5263,"name":"Jeremy Stanley","display_name":"fungi","email":"fungi@yuggoth.org","username":"fungi","status":"missing, presumed fed"},"change_message_id":"fb2e434b6e5b623c8da0def1b6291868693587ef","unresolved":false,"context_lines":[{"line_number":89,"context_line":"\u003chttps://asynchronous.in/docker-mailman/#setting-up-your-web-server\u003e`_"},{"line_number":90,"context_line":"but the latter can be easily adapted to Apache and the Mailman"},{"line_number":91,"context_line":"documentation also has some `tips for this"},{"line_number":92,"context_line":"\u003chttps://docs.list.org/projects/hyperkitty/en/latest/install.html?highlight\u003dapache#running-hyperkitty-on-apache-with-mod-wsgi\u003e`_."},{"line_number":93,"context_line":""},{"line_number":94,"context_line":"Following our usual test-driven development methodology, once basic"},{"line_number":95,"context_line":"functionality is working and tested successfully, we\u0027ll hold a test"}],"source_content_type":"text/x-rst","patch_set":2,"id":"8b1dd946_3667e6d4","line":92,"updated":"2021-11-03 15:19:23.000000000","message":"I don\u0027t get the impression Nginx is the \"default in that ecosystem\" it\u0027s just what happens to be documented on one of the examples. Sticking with Apache for consistency with our other systems is likely to be a greater benefit to us. I\u0027ll note that as the reason I suggested Apache.","commit_id":"9bace208224c05ebb324b49a10e9ec75932fb04d"},{"author":{"_account_id":13252,"name":"Dr. Jens Harbott","display_name":"Jens Harbott (frickler)","email":"frickler@offenerstapel.de","username":"jrosenboom"},"change_message_id":"f737b48ae823aaecf33edb61a230142ebfd83e28","unresolved":true,"context_lines":[{"line_number":89,"context_line":"\u003chttps://asynchronous.in/docker-mailman/#setting-up-your-web-server\u003e`_"},{"line_number":90,"context_line":"but the latter can be easily adapted to Apache and the Mailman"},{"line_number":91,"context_line":"documentation also has some `tips for this"},{"line_number":92,"context_line":"\u003chttps://docs.list.org/projects/hyperkitty/en/latest/install.html?highlight\u003dapache#running-hyperkitty-on-apache-with-mod-wsgi\u003e`_."},{"line_number":93,"context_line":""},{"line_number":94,"context_line":"Following our usual test-driven development methodology, once basic"},{"line_number":95,"context_line":"functionality is working and tested successfully, we\u0027ll hold a test"}],"source_content_type":"text/x-rst","patch_set":2,"id":"2b2be15a_10fd682d","line":92,"updated":"2021-11-03 07:15:36.000000000","message":"So this sounds like you would opt for apache because that\u0027s what we have in a lot of other places already? I won\u0027t object to that, just questioning whether instead going with what seems to be the default in that ecosystem (nginx) might mean less adaption work to do in the long run.","commit_id":"9bace208224c05ebb324b49a10e9ec75932fb04d"},{"author":{"_account_id":5263,"name":"Jeremy Stanley","display_name":"fungi","email":"fungi@yuggoth.org","username":"fungi","status":"missing, presumed fed"},"change_message_id":"fb2e434b6e5b623c8da0def1b6291868693587ef","unresolved":false,"context_lines":[{"line_number":93,"context_line":""},{"line_number":94,"context_line":"Following our usual test-driven development methodology, once basic"},{"line_number":95,"context_line":"functionality is working and tested successfully, we\u0027ll hold a test"},{"line_number":96,"context_line":"nodes and perform more direct interactive testing, first with sample"},{"line_number":97,"context_line":"mailing lists, but later with imported production configuration and"},{"line_number":98,"context_line":"archives (taking sufficient measures to avoid annoying people with"},{"line_number":99,"context_line":"unwanted tests or duplicate messages). The Mailman developers"}],"source_content_type":"text/x-rst","patch_set":2,"id":"dc789658_d8fba879","line":96,"updated":"2021-11-03 15:19:23.000000000","message":"Thanks, fixing.","commit_id":"9bace208224c05ebb324b49a10e9ec75932fb04d"},{"author":{"_account_id":13252,"name":"Dr. Jens Harbott","display_name":"Jens Harbott (frickler)","email":"frickler@offenerstapel.de","username":"jrosenboom"},"change_message_id":"f737b48ae823aaecf33edb61a230142ebfd83e28","unresolved":true,"context_lines":[{"line_number":93,"context_line":""},{"line_number":94,"context_line":"Following our usual test-driven development methodology, once basic"},{"line_number":95,"context_line":"functionality is working and tested successfully, we\u0027ll hold a test"},{"line_number":96,"context_line":"nodes and perform more direct interactive testing, first with sample"},{"line_number":97,"context_line":"mailing lists, but later with imported production configuration and"},{"line_number":98,"context_line":"archives (taking sufficient measures to avoid annoying people with"},{"line_number":99,"context_line":"unwanted tests or duplicate messages). The Mailman developers"}],"source_content_type":"text/x-rst","patch_set":2,"id":"37959786_fa5a2b75","line":96,"range":{"start_line":96,"start_character":0,"end_line":96,"end_character":5},"updated":"2021-11-03 07:15:36.000000000","message":"node","commit_id":"9bace208224c05ebb324b49a10e9ec75932fb04d"},{"author":{"_account_id":5263,"name":"Jeremy Stanley","display_name":"fungi","email":"fungi@yuggoth.org","username":"fungi","status":"missing, presumed fed"},"change_message_id":"fb2e434b6e5b623c8da0def1b6291868693587ef","unresolved":false,"context_lines":[{"line_number":115,"context_line":"After all migrations are completed and any remaining issues sorted,"},{"line_number":116,"context_line":"the old mailing list servers will be imaged for posterity and then"},{"line_number":117,"context_line":"deleted."},{"line_number":118,"context_line":""},{"line_number":119,"context_line":"Alternatives"},{"line_number":120,"context_line":"------------"},{"line_number":121,"context_line":""}],"source_content_type":"text/x-rst","patch_set":2,"id":"98851ad6_a2c12e8a","line":118,"updated":"2021-11-03 15:19:23.000000000","message":"Unknown until we test, so yes we should check such things. Some headers will almost certainly change or be added, though hopefully the ones people tend to filter on like List-Id will stay the same. I\u0027ll note to check these, thanks.","commit_id":"9bace208224c05ebb324b49a10e9ec75932fb04d"},{"author":{"_account_id":13252,"name":"Dr. Jens Harbott","display_name":"Jens Harbott (frickler)","email":"frickler@offenerstapel.de","username":"jrosenboom"},"change_message_id":"f737b48ae823aaecf33edb61a230142ebfd83e28","unresolved":true,"context_lines":[{"line_number":115,"context_line":"After all migrations are completed and any remaining issues sorted,"},{"line_number":116,"context_line":"the old mailing list servers will be imaged for posterity and then"},{"line_number":117,"context_line":"deleted."},{"line_number":118,"context_line":""},{"line_number":119,"context_line":"Alternatives"},{"line_number":120,"context_line":"------------"},{"line_number":121,"context_line":""}],"source_content_type":"text/x-rst","patch_set":2,"id":"8312c729_5a81d6c1","line":118,"updated":"2021-11-03 07:15:36.000000000","message":"Will handling of bounces and messages from non-subscribers change? Might be worth mentioning to check that. Also changes to list-related mail headers that people might be using for filtering.","commit_id":"9bace208224c05ebb324b49a10e9ec75932fb04d"},{"author":{"_account_id":5263,"name":"Jeremy Stanley","display_name":"fungi","email":"fungi@yuggoth.org","username":"fungi","status":"missing, presumed fed"},"change_message_id":"fb2e434b6e5b623c8da0def1b6291868693587ef","unresolved":false,"context_lines":[{"line_number":193,"context_line":"records will need updating to the addresses of the new server during"},{"line_number":194,"context_line":"migration, as will the DNS records for any other virtual domains"},{"line_number":195,"context_line":"added to the old servers prior to migration. New lists01.opendev.org"},{"line_number":196,"context_line":"records will be added, as will ACME records to support HTTPS for all"},{"line_number":197,"context_line":"the virtual domains."},{"line_number":198,"context_line":""},{"line_number":199,"context_line":"Backups"}],"source_content_type":"text/x-rst","patch_set":2,"id":"a48e5e07_25d6e4f2","line":196,"updated":"2021-11-03 15:19:23.000000000","message":"We previously didn\u0027t add HTTPS because of the way the MM2 doesn\u0027t bother to safeguard its user authentication over other channels, so we didn\u0027t want to give our users the impression that the credentials they were supplying to the service were secure in any way. The more modern authentication in MM3 is really the fix here, so switching from HTTP to HTTPS is merely a side effect (HTTPS isn\u0027t a *feature* in and of itself, but rather a necessary implementation detail).","commit_id":"9bace208224c05ebb324b49a10e9ec75932fb04d"},{"author":{"_account_id":5263,"name":"Jeremy Stanley","display_name":"fungi","email":"fungi@yuggoth.org","username":"fungi","status":"missing, presumed fed"},"change_message_id":"252edc69c42008b08cd8cd21bee0485953726ba6","unresolved":false,"context_lines":[{"line_number":193,"context_line":"records will need updating to the addresses of the new server during"},{"line_number":194,"context_line":"migration, as will the DNS records for any other virtual domains"},{"line_number":195,"context_line":"added to the old servers prior to migration. New lists01.opendev.org"},{"line_number":196,"context_line":"records will be added, as will ACME records to support HTTPS for all"},{"line_number":197,"context_line":"the virtual domains."},{"line_number":198,"context_line":""},{"line_number":199,"context_line":"Backups"}],"source_content_type":"text/x-rst","patch_set":2,"id":"de24c1cc_338ef779","line":196,"updated":"2021-11-03 17:56:54.000000000","message":"Yeah, I mean there are a number of incentives for upgrading listed here, a majority of which I didn\u0027t bother to enumerate either: https://docs.list.org/en/latest/migration.html#why-upgrade\n\nSimilarly we could add HTTPS to the existing MM2 deployment fairly trivially if we wanted, but we *need* it for MM3.","commit_id":"9bace208224c05ebb324b49a10e9ec75932fb04d"},{"author":{"_account_id":13252,"name":"Dr. Jens Harbott","display_name":"Jens Harbott (frickler)","email":"frickler@offenerstapel.de","username":"jrosenboom"},"change_message_id":"f737b48ae823aaecf33edb61a230142ebfd83e28","unresolved":true,"context_lines":[{"line_number":193,"context_line":"records will need updating to the addresses of the new server during"},{"line_number":194,"context_line":"migration, as will the DNS records for any other virtual domains"},{"line_number":195,"context_line":"added to the old servers prior to migration. New lists01.opendev.org"},{"line_number":196,"context_line":"records will be added, as will ACME records to support HTTPS for all"},{"line_number":197,"context_line":"the virtual domains."},{"line_number":198,"context_line":""},{"line_number":199,"context_line":"Backups"}],"source_content_type":"text/x-rst","patch_set":2,"id":"59cc12f7_00ae60e2","line":196,"updated":"2021-11-03 07:15:36.000000000","message":"afaict the current servers don\u0027t do https for the web servers, but that would be a very welcomed and important addition.\n\nYou even could add lack of https to the problem description, as it is regularly popping up when browsing in https-only mode.","commit_id":"9bace208224c05ebb324b49a10e9ec75932fb04d"},{"author":{"_account_id":13252,"name":"Dr. Jens Harbott","display_name":"Jens Harbott (frickler)","email":"frickler@offenerstapel.de","username":"jrosenboom"},"change_message_id":"1efb17ae82bdf1fbe98a267188abcbfa793a3875","unresolved":false,"context_lines":[{"line_number":193,"context_line":"records will need updating to the addresses of the new server during"},{"line_number":194,"context_line":"migration, as will the DNS records for any other virtual domains"},{"line_number":195,"context_line":"added to the old servers prior to migration. New lists01.opendev.org"},{"line_number":196,"context_line":"records will be added, as will ACME records to support HTTPS for all"},{"line_number":197,"context_line":"the virtual domains."},{"line_number":198,"context_line":""},{"line_number":199,"context_line":"Backups"}],"source_content_type":"text/x-rst","patch_set":2,"id":"de6870ef_9d73f8b2","line":196,"in_reply_to":"a48e5e07_25d6e4f2","updated":"2021-11-03 17:46:32.000000000","message":"I was mainly thinking about browsing the archives and providing privacy against eavesdropping while doing so, which I do consider a feature. But if you argue that this feature isn\u0027t directly tied to the upgrade and thus shouldn\u0027t have focus here, I\u0027m fine with that.","commit_id":"9bace208224c05ebb324b49a10e9ec75932fb04d"}]}