)]}'
{"doc/source/test_plans/openstack_wan/plan.rst":[{"author":{"_account_id":5950,"name":"Ilya Shakhat","email":"shakhat@gmail.com","username":"shakhat"},"change_message_id":"97b2d250a4a7b129a11ab00eb52c58af1d4794b8","unresolved":false,"context_lines":[{"line_number":10,"context_line":":Abstract:"},{"line_number":11,"context_line":""},{"line_number":12,"context_line":"   This document describes a test plan for analysing how OpenStack"},{"line_number":13,"context_line":"   behaves under Wide Are Network (WAN) conditions. In particular,"},{"line_number":14,"context_line":"   measuring the effect of high latency, low bandwidth, packet loss"},{"line_number":15,"context_line":"   and link failure."},{"line_number":16,"context_line":""}],"source_content_type":"text/x-rst","patch_set":2,"id":"df140735_f9a1bc47","line":13,"updated":"2017-06-02 11:52:55.000000000","message":"Are -\u003e Area","commit_id":"9eca27eab3e813a37013bce6353550b237a149e8"},{"author":{"_account_id":23936,"name":"Ronan-Alexandre Cherrueau","email":"Ronan-Alexandre.Cherrueau@inria.fr","username":"rcherrueau"},"change_message_id":"7aaba19965c759d6cb38c686d0d4fe3783c9dc34","unresolved":false,"context_lines":[{"line_number":10,"context_line":":Abstract:"},{"line_number":11,"context_line":""},{"line_number":12,"context_line":"   This document describes a test plan for analysing how OpenStack"},{"line_number":13,"context_line":"   behaves under Wide Are Network (WAN) conditions. In particular,"},{"line_number":14,"context_line":"   measuring the effect of high latency, low bandwidth, packet loss"},{"line_number":15,"context_line":"   and link failure."},{"line_number":16,"context_line":""}],"source_content_type":"text/x-rst","patch_set":2,"id":"df140735_85e48f3d","line":13,"in_reply_to":"df140735_f9a1bc47","updated":"2017-06-02 13:46:15.000000000","message":"Done","commit_id":"9eca27eab3e813a37013bce6353550b237a149e8"},{"author":{"_account_id":5950,"name":"Ilya Shakhat","email":"shakhat@gmail.com","username":"shakhat"},"change_message_id":"97b2d250a4a7b129a11ab00eb52c58af1d4794b8","unresolved":false,"context_lines":[{"line_number":155,"context_line":"  - ``NovaSecGroup.boot_server_and_add_secgroups``"},{"line_number":156,"context_line":""},{"line_number":157,"context_line":"Neutron scenarios"},{"line_number":158,"context_line":"  - ``NeutronNetworks.create_and_delete_networks``"},{"line_number":159,"context_line":"  - ``NeutronNetworks.create_and_delete_ports``"},{"line_number":160,"context_line":"  - ``NeutronNetworks.create_and_delete_routers``"},{"line_number":161,"context_line":"  - ``NeutronNetworks.create_and_delete_subnets``"}],"source_content_type":"text/x-rst","patch_set":2,"id":"df140735_f4ad439b","line":158,"updated":"2017-06-02 11:52:55.000000000","message":"These Neutron scenarios should not be affected by WAN conditions (they create DB objects on Controller and namespaces on Network node). However it deserves to be verified :)","commit_id":"9eca27eab3e813a37013bce6353550b237a149e8"},{"author":{"_account_id":5950,"name":"Ilya Shakhat","email":"shakhat@gmail.com","username":"shakhat"},"change_message_id":"af13b8539d39f1ec557fdfcb238bc0fc5342e4f5","unresolved":false,"context_lines":[{"line_number":155,"context_line":"  - ``NovaSecGroup.boot_server_and_add_secgroups``"},{"line_number":156,"context_line":""},{"line_number":157,"context_line":"Neutron scenarios"},{"line_number":158,"context_line":"  - ``NeutronNetworks.create_and_delete_networks``"},{"line_number":159,"context_line":"  - ``NeutronNetworks.create_and_delete_ports``"},{"line_number":160,"context_line":"  - ``NeutronNetworks.create_and_delete_routers``"},{"line_number":161,"context_line":"  - ``NeutronNetworks.create_and_delete_subnets``"}],"source_content_type":"text/x-rst","patch_set":2,"id":"df140735_886143ac","line":158,"in_reply_to":"df140735_65997b77","updated":"2017-06-02 14:20:25.000000000","message":"Right, agree.","commit_id":"9eca27eab3e813a37013bce6353550b237a149e8"},{"author":{"_account_id":23936,"name":"Ronan-Alexandre Cherrueau","email":"Ronan-Alexandre.Cherrueau@inria.fr","username":"rcherrueau"},"change_message_id":"7aaba19965c759d6cb38c686d0d4fe3783c9dc34","unresolved":false,"context_lines":[{"line_number":155,"context_line":"  - ``NovaSecGroup.boot_server_and_add_secgroups``"},{"line_number":156,"context_line":""},{"line_number":157,"context_line":"Neutron scenarios"},{"line_number":158,"context_line":"  - ``NeutronNetworks.create_and_delete_networks``"},{"line_number":159,"context_line":"  - ``NeutronNetworks.create_and_delete_ports``"},{"line_number":160,"context_line":"  - ``NeutronNetworks.create_and_delete_routers``"},{"line_number":161,"context_line":"  - ``NeutronNetworks.create_and_delete_subnets``"}],"source_content_type":"text/x-rst","patch_set":2,"id":"df140735_65997b77","line":158,"in_reply_to":"df140735_f4ad439b","updated":"2017-06-02 13:46:15.000000000","message":"Should be tested, but pretty sure they could, depending on Neutron configuration. For instance, activating the DVR[1] pushes neutron l3 agents on computes. These agents take part of the process at control plane level, so scenarios should be affected by WAN condition.\n\n[1] https://wiki.openstack.org/wiki/Neutron/DVR","commit_id":"9eca27eab3e813a37013bce6353550b237a149e8"},{"author":{"_account_id":5950,"name":"Ilya Shakhat","email":"shakhat@gmail.com","username":"shakhat"},"change_message_id":"97b2d250a4a7b129a11ab00eb52c58af1d4794b8","unresolved":false,"context_lines":[{"line_number":212,"context_line":"parameter. However, to the best of our knowledge, there are no tools"},{"line_number":213,"context_line":"to automatically inject such failure at a specific point of an"},{"line_number":214,"context_line":"OpenStack workflow (something more convenient than modifying the"},{"line_number":215,"context_line":"OpenStack source code)."},{"line_number":216,"context_line":""},{"line_number":217,"context_line":".. _RPC call: https://github.com/openstack/nova/blob/bf34d5f25ca6f4650777a5eb4e3316c5af8ba54d/nova/compute/manager.py#L1720"},{"line_number":218,"context_line":""}],"source_content_type":"text/x-rst","patch_set":2,"id":"df140735_341bbbd0","line":215,"updated":"2017-06-02 11:52:55.000000000","message":"right, failures cannot be injected based on internal OpenStack state, but random approach may be helpful too (https://docs.openstack.org/developer/performance-docs/test_plans/reliability/version_2/plan.html)","commit_id":"9eca27eab3e813a37013bce6353550b237a149e8"},{"author":{"_account_id":23936,"name":"Ronan-Alexandre Cherrueau","email":"Ronan-Alexandre.Cherrueau@inria.fr","username":"rcherrueau"},"change_message_id":"7aaba19965c759d6cb38c686d0d4fe3783c9dc34","unresolved":false,"context_lines":[{"line_number":212,"context_line":"parameter. However, to the best of our knowledge, there are no tools"},{"line_number":213,"context_line":"to automatically inject such failure at a specific point of an"},{"line_number":214,"context_line":"OpenStack workflow (something more convenient than modifying the"},{"line_number":215,"context_line":"OpenStack source code)."},{"line_number":216,"context_line":""},{"line_number":217,"context_line":".. _RPC call: https://github.com/openstack/nova/blob/bf34d5f25ca6f4650777a5eb4e3316c5af8ba54d/nova/compute/manager.py#L1720"},{"line_number":218,"context_line":""}],"source_content_type":"text/x-rst","patch_set":2,"id":"df140735_c5bcc607","line":215,"in_reply_to":"df140735_341bbbd0","updated":"2017-06-02 13:46:15.000000000","message":"Oh! I don\u0027t know about os-faults. It seems a useful tool for fault injections. Do you think it\u0027s a good idea to mention it here, even if there is no mean to perform a fault injection at a specific point of the workflow?","commit_id":"9eca27eab3e813a37013bce6353550b237a149e8"},{"author":{"_account_id":5950,"name":"Ilya Shakhat","email":"shakhat@gmail.com","username":"shakhat"},"change_message_id":"af13b8539d39f1ec557fdfcb238bc0fc5342e4f5","unresolved":false,"context_lines":[{"line_number":212,"context_line":"parameter. However, to the best of our knowledge, there are no tools"},{"line_number":213,"context_line":"to automatically inject such failure at a specific point of an"},{"line_number":214,"context_line":"OpenStack workflow (something more convenient than modifying the"},{"line_number":215,"context_line":"OpenStack source code)."},{"line_number":216,"context_line":""},{"line_number":217,"context_line":".. _RPC call: https://github.com/openstack/nova/blob/bf34d5f25ca6f4650777a5eb4e3316c5af8ba54d/nova/compute/manager.py#L1720"},{"line_number":218,"context_line":""}],"source_content_type":"text/x-rst","patch_set":2,"id":"df140735_08cd3365","line":215,"in_reply_to":"df140735_c5bcc607","updated":"2017-06-02 14:20:25.000000000","message":"I think we can skip it here (I just wanted to let you know that there is a tool to simplify fault injection)","commit_id":"9eca27eab3e813a37013bce6353550b237a149e8"}]}