)]}'
{"swift/obj/updater.py":[{"author":{"_account_id":1179,"name":"Clay Gerrard","email":"clay.gerrard@gmail.com","username":"clay-gerrard"},"change_message_id":"9acc9434c36550621f4e3a877a95c50dc6e37efe","unresolved":false,"context_lines":[{"line_number":245,"context_line":"            self.failures +\u003d 1"},{"line_number":246,"context_line":"            self.logger.increment(\u0027failures\u0027)"},{"line_number":247,"context_line":"            self.logger.debug(\u0027Update failed for %(obj)s %(path)s\u0027,"},{"line_number":248,"context_line":"                              {\u0027obj\u0027: obj, \u0027path\u0027: update_path})"},{"line_number":249,"context_line":"            if new_successes:"},{"line_number":250,"context_line":"                update[\u0027successes\u0027] \u003d successes"},{"line_number":251,"context_line":"                write_pickle(update, update_path, os.path.join("}],"source_content_type":"text/x-python","patch_set":1,"id":"3a71b18c_f7b31698","line":248,"updated":"2016-12-05 18:09:31.000000000","message":"This is the pre-existing message that currently logs the result of all the requests, and it\u0027s a debug, and that\u0027s working fine, so I don\u0027t think the new message should be higher than this.\n\nPerhaps a compromise would be to make this message a warning?","commit_id":"765cc945d67deaf029e700885aa41ed2c72982eb"},{"author":{"_account_id":1179,"name":"Clay Gerrard","email":"clay.gerrard@gmail.com","username":"clay-gerrard"},"change_message_id":"de42967eb02eabfa597cd893d9bbf9ec3d6e45ee","unresolved":false,"context_lines":[{"line_number":245,"context_line":"            self.failures +\u003d 1"},{"line_number":246,"context_line":"            self.logger.increment(\u0027failures\u0027)"},{"line_number":247,"context_line":"            self.logger.debug(\u0027Update failed for %(obj)s %(path)s\u0027,"},{"line_number":248,"context_line":"                              {\u0027obj\u0027: obj, \u0027path\u0027: update_path})"},{"line_number":249,"context_line":"            if new_successes:"},{"line_number":250,"context_line":"                update[\u0027successes\u0027] \u003d successes"},{"line_number":251,"context_line":"                write_pickle(update, update_path, os.path.join("}],"source_content_type":"text/x-python","patch_set":1,"id":"3a71b18c_565c907c","line":248,"in_reply_to":"3a71b18c_94f80ec5","updated":"2016-12-06 20:35:27.000000000","message":"ZOMG - I like that diff *a lot* \n\nagree, separate change - lp bug #1647840","commit_id":"765cc945d67deaf029e700885aa41ed2c72982eb"},{"author":{"_account_id":4608,"name":"Kota Tsuyuzaki","email":"bloodeagle40234@gmail.com","username":"tsuyuzaki-kota"},"change_message_id":"477036bfd7c71565805f82307663f085d657d295","unresolved":false,"context_lines":[{"line_number":245,"context_line":"            self.failures +\u003d 1"},{"line_number":246,"context_line":"            self.logger.increment(\u0027failures\u0027)"},{"line_number":247,"context_line":"            self.logger.debug(\u0027Update failed for %(obj)s %(path)s\u0027,"},{"line_number":248,"context_line":"                              {\u0027obj\u0027: obj, \u0027path\u0027: update_path})"},{"line_number":249,"context_line":"            if new_successes:"},{"line_number":250,"context_line":"                update[\u0027successes\u0027] \u003d successes"},{"line_number":251,"context_line":"                write_pickle(update, update_path, os.path.join("}],"source_content_type":"text/x-python","patch_set":1,"id":"3a71b18c_94f80ec5","line":248,"in_reply_to":"3a71b18c_f7b31698","updated":"2016-12-06 02:52:43.000000000","message":"Fair point. And we could find the failure status via the incremental failures count in statsd?\n\nI think turning this one to warning may be a little problematic because the failure count will be incremented also Request Timeout cases so it may occur in some situation like the remote node restarted.\n\nHowever, I still want to set a warning line for the *related-change* case (i.e. all primaries return 404s) because obviously, it\u0027s in not healthy status.\n\nSo, I wrote up my PoC code how to improve this one here:\nhttps://gist.github.com/bloodeagle40234/2d747f44909d3d3f153a27a70afb7a39\n\nIn that patch, we log a warning line only when all (attempt) primaries return 404s (not yet consider 2nd-cycle if a part of primaries failed with Timeout at the first time, though)\n\nAnd thinking of that, I\u0027m realizing, it could be ok to move that to another thread (not one patch!).","commit_id":"765cc945d67deaf029e700885aa41ed2c72982eb"},{"author":{"_account_id":7847,"name":"Alistair Coles","email":"alistairncoles@gmail.com","username":"acoles"},"change_message_id":"fabb79d81d71e58908347dbeb1b6da5c46dfb5ff","unresolved":false,"context_lines":[{"line_number":270,"context_line":"                resp.read()"},{"line_number":271,"context_line":"                success \u003d is_success(resp.status)"},{"line_number":272,"context_line":"                if not success:"},{"line_number":273,"context_line":"                    self.logger.debug("},{"line_number":274,"context_line":"                        _(\u0027Error code %(status)d is returned from remote \u0027"},{"line_number":275,"context_line":"                          \u0027server %(ip)s: %(port)s / %(device)s\u0027),"},{"line_number":276,"context_line":"                        {\u0027status\u0027: resp.status, \u0027ip\u0027: node[\u0027ip\u0027],"}],"source_content_type":"text/x-python","patch_set":1,"id":"3a71b18c_e81630f0","line":273,"updated":"2016-12-02 12:21:02.000000000","message":"+1","commit_id":"765cc945d67deaf029e700885aa41ed2c72982eb"},{"author":{"_account_id":1179,"name":"Clay Gerrard","email":"clay.gerrard@gmail.com","username":"clay-gerrard"},"change_message_id":"de42967eb02eabfa597cd893d9bbf9ec3d6e45ee","unresolved":false,"context_lines":[{"line_number":270,"context_line":"                resp.read()"},{"line_number":271,"context_line":"                success \u003d is_success(resp.status)"},{"line_number":272,"context_line":"                if not success:"},{"line_number":273,"context_line":"                    self.logger.debug("},{"line_number":274,"context_line":"                        _(\u0027Error code %(status)d is returned from remote \u0027"},{"line_number":275,"context_line":"                          \u0027server %(ip)s: %(port)s / %(device)s\u0027),"},{"line_number":276,"context_line":"                        {\u0027status\u0027: resp.status, \u0027ip\u0027: node[\u0027ip\u0027],"}],"source_content_type":"text/x-python","patch_set":1,"id":"3a71b18c_f6c91c27","line":273,"in_reply_to":"3a71b18c_89d0fbf7","updated":"2016-12-06 20:35:27.000000000","message":"good point!  we log *a lot* already!  no need to add more here I think.","commit_id":"765cc945d67deaf029e700885aa41ed2c72982eb"},{"author":{"_account_id":1179,"name":"Clay Gerrard","email":"clay.gerrard@gmail.com","username":"clay-gerrard"},"change_message_id":"9acc9434c36550621f4e3a877a95c50dc6e37efe","unresolved":false,"context_lines":[{"line_number":270,"context_line":"                resp.read()"},{"line_number":271,"context_line":"                success \u003d is_success(resp.status)"},{"line_number":272,"context_line":"                if not success:"},{"line_number":273,"context_line":"                    self.logger.debug("},{"line_number":274,"context_line":"                        _(\u0027Error code %(status)d is returned from remote \u0027"},{"line_number":275,"context_line":"                          \u0027server %(ip)s: %(port)s / %(device)s\u0027),"},{"line_number":276,"context_line":"                        {\u0027status\u0027: resp.status, \u0027ip\u0027: node[\u0027ip\u0027],"}],"source_content_type":"text/x-python","patch_set":1,"id":"3a71b18c_9753c238","line":273,"in_reply_to":"3a71b18c_8e7f4c65","updated":"2016-12-05 18:09:31.000000000","message":"This is a new message added in the related change, and I\u0027m sure it was just for debugging, the issue in the related bug would not have been diagnosed or addressed any sooner because of a lack of operator insight.","commit_id":"765cc945d67deaf029e700885aa41ed2c72982eb"},{"author":{"_account_id":4608,"name":"Kota Tsuyuzaki","email":"bloodeagle40234@gmail.com","username":"tsuyuzaki-kota"},"change_message_id":"477036bfd7c71565805f82307663f085d657d295","unresolved":false,"context_lines":[{"line_number":270,"context_line":"                resp.read()"},{"line_number":271,"context_line":"                success \u003d is_success(resp.status)"},{"line_number":272,"context_line":"                if not success:"},{"line_number":273,"context_line":"                    self.logger.debug("},{"line_number":274,"context_line":"                        _(\u0027Error code %(status)d is returned from remote \u0027"},{"line_number":275,"context_line":"                          \u0027server %(ip)s: %(port)s / %(device)s\u0027),"},{"line_number":276,"context_line":"                        {\u0027status\u0027: resp.status, \u0027ip\u0027: node[\u0027ip\u0027],"}],"source_content_type":"text/x-python","patch_set":1,"id":"3a71b18c_89d0fbf7","line":273,"in_reply_to":"3a71b18c_9753c238","updated":"2016-12-06 02:52:43.000000000","message":"For now, we have error log lines if container updates failed[1]\n\n1: https://github.com/openstack/swift/blob/master/swift/obj/server.py#L251-L256","commit_id":"765cc945d67deaf029e700885aa41ed2c72982eb"},{"author":{"_account_id":4608,"name":"Kota Tsuyuzaki","email":"bloodeagle40234@gmail.com","username":"tsuyuzaki-kota"},"change_message_id":"40d6e0ba8c2c9940349a3afd2ff9437228b011f6","unresolved":false,"context_lines":[{"line_number":270,"context_line":"                resp.read()"},{"line_number":271,"context_line":"                success \u003d is_success(resp.status)"},{"line_number":272,"context_line":"                if not success:"},{"line_number":273,"context_line":"                    self.logger.debug("},{"line_number":274,"context_line":"                        _(\u0027Error code %(status)d is returned from remote \u0027"},{"line_number":275,"context_line":"                          \u0027server %(ip)s: %(port)s / %(device)s\u0027),"},{"line_number":276,"context_line":"                        {\u0027status\u0027: resp.status, \u0027ip\u0027: node[\u0027ip\u0027],"}],"source_content_type":"text/x-python","patch_set":1,"id":"3a71b18c_8e7f4c65","line":273,"in_reply_to":"3a71b18c_e81630f0","updated":"2016-12-05 08:00:54.000000000","message":"Hmm... looking at the related change, I\u0027d like to this as warning rather than debug if you don\u0027t like to make it as error.\n\nThat is because the case we got `not success` here is obviously something wrong health in the cluster (including all primary container dbs in handoffs) and some log level prior to debug could be helpful to find the state and consider the cluster management (e.g. check the clsuter health, rebalanece, etc...).\n\n\nOr if you don\u0027t like to log for indivisual requests, could we make it around [2] to summarize the update jobs? However, [2] is now logging as debug and perhaps we don\u0027t want to log on the Timeout case.\n\n1: https://review.openstack.org/#/c/99598/18/swift/obj/updater.py\n\n2: https://github.com/openstack/swift/blob/master/swift/obj/updater.py#L226-L253","commit_id":"765cc945d67deaf029e700885aa41ed2c72982eb"}],"test/probe/test_object_async_update.py":[{"author":{"_account_id":7847,"name":"Alistair Coles","email":"alistairncoles@gmail.com","username":"acoles"},"change_message_id":"fabb79d81d71e58908347dbeb1b6da5c46dfb5ff","unresolved":false,"context_lines":[{"line_number":110,"context_line":"                cnode, cpart, self.account, container)[1])"},{"line_number":111,"context_line":""},{"line_number":112,"context_line":"        # since the container is empty - we can delete it!"},{"line_number":113,"context_line":"        client.delete_container(self.url, self.token, container)"},{"line_number":114,"context_line":""},{"line_number":115,"context_line":"        # Re-run the object-updaters and now container replicas in primary"},{"line_number":116,"context_line":"        # container servers should get updated"}],"source_content_type":"text/x-python","patch_set":1,"id":"3a71b18c_2e42f13e","line":113,"updated":"2016-12-02 12:21:02.000000000","message":"ok, the subsequent async update PUTs will still succeed despite the container being marked as deleted, and the listing works because once the container has objects then it is not deleted any more!","commit_id":"765cc945d67deaf029e700885aa41ed2c72982eb"},{"author":{"_account_id":4608,"name":"Kota Tsuyuzaki","email":"bloodeagle40234@gmail.com","username":"tsuyuzaki-kota"},"change_message_id":"40d6e0ba8c2c9940349a3afd2ff9437228b011f6","unresolved":false,"context_lines":[{"line_number":110,"context_line":"                cnode, cpart, self.account, container)[1])"},{"line_number":111,"context_line":""},{"line_number":112,"context_line":"        # since the container is empty - we can delete it!"},{"line_number":113,"context_line":"        client.delete_container(self.url, self.token, container)"},{"line_number":114,"context_line":""},{"line_number":115,"context_line":"        # Re-run the object-updaters and now container replicas in primary"},{"line_number":116,"context_line":"        # container servers should get updated"}],"source_content_type":"text/x-python","patch_set":1,"id":"3a71b18c_8e866cd9","line":113,"in_reply_to":"3a71b18c_2e42f13e","updated":"2016-12-05 08:00:54.000000000","message":"Phew, exactly.","commit_id":"765cc945d67deaf029e700885aa41ed2c72982eb"}]}
