)]}'
{"ansible/roles/mariadb/tasks/lookup_cluster.yml":[{"author":{"_account_id":14826,"name":"Mark Goddard","email":"markgoddard86@gmail.com","username":"mgoddard"},"change_message_id":"8c0a54546472bc53c94cb84d1df2877114ca1b16","unresolved":false,"context_lines":[{"line_number":61,"context_line":""},{"line_number":62,"context_line":"    - name: Divide hosts by their MariaDB service WSREP sync status"},{"line_number":63,"context_line":"      group_by:"},{"line_number":64,"context_line":"        key: mariadb_sync_status_{{ mariadb_sync_status }}"},{"line_number":65,"context_line":""},{"line_number":66,"context_line":"    - name: Fail when MariaDB services are not synced across the whole cluster"},{"line_number":67,"context_line":"      fail:"}],"source_content_type":"text/x-yaml","patch_set":1,"id":"3fa7e38b_40abbad5","line":64,"range":{"start_line":64,"start_character":36,"end_line":64,"end_character":55},"updated":"2020-01-28 10:41:55.000000000","message":"| default(\u0027NA\u0027)","commit_id":"d2891c96ba0c714125d15b611b97b35934bd60ee"},{"author":{"_account_id":30491,"name":"Radosław Piliszek","display_name":"Radek","email":"radek@piliszek.it","username":"yoctozepto","status":"self-employed techologist, collaborating mostly with 7bulls.com"},"change_message_id":"69634ced8979c0b9f8b1cf0458aaa3f3fdef71e4","unresolved":false,"context_lines":[{"line_number":61,"context_line":""},{"line_number":62,"context_line":"    - name: Divide hosts by their MariaDB service WSREP sync status"},{"line_number":63,"context_line":"      group_by:"},{"line_number":64,"context_line":"        key: mariadb_sync_status_{{ mariadb_sync_status }}"},{"line_number":65,"context_line":""},{"line_number":66,"context_line":"    - name: Fail when MariaDB services are not synced across the whole cluster"},{"line_number":67,"context_line":"      fail:"}],"source_content_type":"text/x-yaml","patch_set":1,"id":"3fa7e38b_e05f46e1","line":64,"range":{"start_line":64,"start_character":36,"end_line":64,"end_character":55},"in_reply_to":"3fa7e38b_40abbad5","updated":"2020-01-28 10:48:30.000000000","message":"makes sense, thanks :-)","commit_id":"d2891c96ba0c714125d15b611b97b35934bd60ee"},{"author":{"_account_id":14826,"name":"Mark Goddard","email":"markgoddard86@gmail.com","username":"mgoddard"},"change_message_id":"8c0a54546472bc53c94cb84d1df2877114ca1b16","unresolved":false,"context_lines":[{"line_number":67,"context_line":"      fail:"},{"line_number":68,"context_line":"        msg: MariaDB cluster is not synced. Please wait for WSREP sync before proceeding."},{"line_number":69,"context_line":"      when:"},{"line_number":70,"context_line":"        - groups.mariadb_port_alive_True is defined"},{"line_number":71,"context_line":"        - groups.mariadb_sync_status_Synced is not defined or"},{"line_number":72,"context_line":"          groups.mariadb_port_alive_True | sort !\u003d groups.mariadb_sync_status_Synced | sort"},{"line_number":73,"context_line":"  when: not mariadb_recover | default(False)"}],"source_content_type":"text/x-yaml","patch_set":1,"id":"3fa7e38b_c0cecab5","line":70,"range":{"start_line":70,"start_character":10,"end_line":70,"end_character":51},"updated":"2020-01-28 10:41:55.000000000","message":"Do we mean mariadb_cluster_exists? That is what defines whether we bootstrap.\n\nThere is another failure case that I don\u0027t think we\u0027re catching here - a volume exists so mariadb_cluster_exists is True, but no services are up so groups.mariadb_port_alive_True is undefined.","commit_id":"d2891c96ba0c714125d15b611b97b35934bd60ee"},{"author":{"_account_id":14826,"name":"Mark Goddard","email":"markgoddard86@gmail.com","username":"mgoddard"},"change_message_id":"e418984d51428674814bd485f61e254bc3491876","unresolved":false,"context_lines":[{"line_number":67,"context_line":"      fail:"},{"line_number":68,"context_line":"        msg: MariaDB cluster is not synced. Please wait for WSREP sync before proceeding."},{"line_number":69,"context_line":"      when:"},{"line_number":70,"context_line":"        - groups.mariadb_port_alive_True is defined"},{"line_number":71,"context_line":"        - groups.mariadb_sync_status_Synced is not defined or"},{"line_number":72,"context_line":"          groups.mariadb_port_alive_True | sort !\u003d groups.mariadb_sync_status_Synced | sort"},{"line_number":73,"context_line":"  when: not mariadb_recover | default(False)"}],"source_content_type":"text/x-yaml","patch_set":1,"id":"3fa7e38b_c0100aae","line":70,"range":{"start_line":70,"start_character":10,"end_line":70,"end_character":51},"in_reply_to":"3fa7e38b_40f5faca","updated":"2020-01-28 11:07:32.000000000","message":"I don\u0027t think it\u0027s much of a corner case. Why wouldn\u0027t users assume that running deploy against a bunch of stopped containers would bring them back up? The least we could do is detect it and tell them to run recover.\n\nI\u0027m not sure about the case where there is only one host. Can we just start it up again? Otherwise how would you restart a single node \u0027cluster\u0027? Or maybe there is some membership timeout.","commit_id":"d2891c96ba0c714125d15b611b97b35934bd60ee"},{"author":{"_account_id":14826,"name":"Mark Goddard","email":"markgoddard86@gmail.com","username":"mgoddard"},"change_message_id":"ae30ea178a2d40db3a2c942e4a158cd896361bd2","unresolved":false,"context_lines":[{"line_number":67,"context_line":"      fail:"},{"line_number":68,"context_line":"        msg: MariaDB cluster is not synced. Please wait for WSREP sync before proceeding."},{"line_number":69,"context_line":"      when:"},{"line_number":70,"context_line":"        - groups.mariadb_port_alive_True is defined"},{"line_number":71,"context_line":"        - groups.mariadb_sync_status_Synced is not defined or"},{"line_number":72,"context_line":"          groups.mariadb_port_alive_True | sort !\u003d groups.mariadb_sync_status_Synced | sort"},{"line_number":73,"context_line":"  when: not mariadb_recover | default(False)"}],"source_content_type":"text/x-yaml","patch_set":1,"id":"3fa7e38b_83ae9c9e","line":70,"range":{"start_line":70,"start_character":10,"end_line":70,"end_character":51},"in_reply_to":"3fa7e38b_8339bca2","updated":"2020-01-28 11:48:15.000000000","message":"fail:\n  msg: MariaDB cluster is stopped. Please start it using kolla-ansible mariadb_recovery\nwhen:\n  - groups[\u0027mariadb\u0027] | length \u003e 1\n  - mariadb_cluster_exists\n  - groups.mariadb_port_alive_True is not defined","commit_id":"d2891c96ba0c714125d15b611b97b35934bd60ee"},{"author":{"_account_id":30491,"name":"Radosław Piliszek","display_name":"Radek","email":"radek@piliszek.it","username":"yoctozepto","status":"self-employed techologist, collaborating mostly with 7bulls.com"},"change_message_id":"cae324e7a83ef80f2f7b254de1878cfb092bf7d8","unresolved":false,"context_lines":[{"line_number":67,"context_line":"      fail:"},{"line_number":68,"context_line":"        msg: MariaDB cluster is not synced. Please wait for WSREP sync before proceeding."},{"line_number":69,"context_line":"      when:"},{"line_number":70,"context_line":"        - groups.mariadb_port_alive_True is defined"},{"line_number":71,"context_line":"        - groups.mariadb_sync_status_Synced is not defined or"},{"line_number":72,"context_line":"          groups.mariadb_port_alive_True | sort !\u003d groups.mariadb_sync_status_Synced | sort"},{"line_number":73,"context_line":"  when: not mariadb_recover | default(False)"}],"source_content_type":"text/x-yaml","patch_set":1,"id":"3fa7e38b_8339bca2","line":70,"range":{"start_line":70,"start_character":10,"end_line":70,"end_character":51},"in_reply_to":"3fa7e38b_c0100aae","updated":"2020-01-28 11:38:17.000000000","message":"Single-node will start. Multi-node not likely - you *could* luck out and start the last shutdown first but it\u0027s hard to fix this.\n\nWhat do you propose?","commit_id":"d2891c96ba0c714125d15b611b97b35934bd60ee"},{"author":{"_account_id":30491,"name":"Radosław Piliszek","display_name":"Radek","email":"radek@piliszek.it","username":"yoctozepto","status":"self-employed techologist, collaborating mostly with 7bulls.com"},"change_message_id":"69634ced8979c0b9f8b1cf0458aaa3f3fdef71e4","unresolved":false,"context_lines":[{"line_number":67,"context_line":"      fail:"},{"line_number":68,"context_line":"        msg: MariaDB cluster is not synced. Please wait for WSREP sync before proceeding."},{"line_number":69,"context_line":"      when:"},{"line_number":70,"context_line":"        - groups.mariadb_port_alive_True is defined"},{"line_number":71,"context_line":"        - groups.mariadb_sync_status_Synced is not defined or"},{"line_number":72,"context_line":"          groups.mariadb_port_alive_True | sort !\u003d groups.mariadb_sync_status_Synced | sort"},{"line_number":73,"context_line":"  when: not mariadb_recover | default(False)"}],"source_content_type":"text/x-yaml","patch_set":1,"id":"3fa7e38b_40f5faca","line":70,"range":{"start_line":70,"start_character":10,"end_line":70,"end_character":51},"in_reply_to":"3fa7e38b_c0cecab5","updated":"2020-01-28 10:48:30.000000000","message":"No, this is just prereq check for the other check. If we have a volume and no are up, then we try to start them in handlers. It will probably fail but we can\u0027t really protect k-a against all madness our users try to throw at it.","commit_id":"d2891c96ba0c714125d15b611b97b35934bd60ee"}]}