)]}'
{"/COMMIT_MSG":[{"author":{"_account_id":748,"name":"Armando Migliaccio","email":"armamig@gmail.com","username":"armando-migliaccio"},"change_message_id":"4645fd14ef9fa58b8d25c6c8d88baa5cba2fbbbd","unresolved":false,"context_lines":[{"line_number":4,"context_line":"Commit:     Neil Jerram \u003cNeil.Jerram@metaswitch.com\u003e"},{"line_number":5,"context_line":"CommitDate: 2015-09-06 01:09:16 +0100"},{"line_number":6,"context_line":""},{"line_number":7,"context_line":"Handle process disappearing before we ask for its PPID"},{"line_number":8,"context_line":""},{"line_number":9,"context_line":"Change-Id: I573aba8e11dca16f8a6565f7e9704be18e938566"},{"line_number":10,"context_line":"Closes-Bug: #1478190"}],"source_content_type":"text/x-gerrit-commit-message","patch_set":1,"id":"da20952f_2855c5f8","line":7,"updated":"2015-09-07 00:51:39.000000000","message":"I haven\u0027t looked enough in the bug report to fully understand the issue, and although this fix seems sensible I wonder whether the process is expected to be there or not. Just providing a safe get_pid as you did suffices to mask the underlying issue, but I wonder if it does not going to lead to some other weird failure mode. Please check with Elena Ezhova (if not done already), as she was the original assignee of the bug.","commit_id":"42f80682d3eff58af60199f817ac402f457491a0"},{"author":{"_account_id":9656,"name":"Ihar Hrachyshka","email":"ihrachys@redhat.com","username":"ihrachys","status":"Red Hat Networking Systems Engineer"},"change_message_id":"203b5a214cef2303fd068259e972bc319712b3a5","unresolved":false,"context_lines":[{"line_number":4,"context_line":"Commit:     Neil Jerram \u003cNeil.Jerram@metaswitch.com\u003e"},{"line_number":5,"context_line":"CommitDate: 2015-09-06 01:09:16 +0100"},{"line_number":6,"context_line":""},{"line_number":7,"context_line":"Handle process disappearing before we ask for its PPID"},{"line_number":8,"context_line":""},{"line_number":9,"context_line":"Change-Id: I573aba8e11dca16f8a6565f7e9704be18e938566"},{"line_number":10,"context_line":"Closes-Bug: #1478190"}],"source_content_type":"text/x-gerrit-commit-message","patch_set":1,"id":"da20952f_f926ae65","line":7,"in_reply_to":"da20952f_2855c5f8","updated":"2015-09-07 14:27:21.000000000","message":"process_iter() returns all pids on the system, even not test managed. So we should be prepared for any of those external processes to die while iteration is ongoing.","commit_id":"42f80682d3eff58af60199f817ac402f457491a0"},{"author":{"_account_id":13734,"name":"Nell Jerram","email":"nell@tigera.io","username":"neiljerram"},"change_message_id":"39826b4c864d336c22914a174ba544705325cf36","unresolved":false,"context_lines":[{"line_number":4,"context_line":"Commit:     Neil Jerram \u003cNeil.Jerram@metaswitch.com\u003e"},{"line_number":5,"context_line":"CommitDate: 2015-09-06 01:09:16 +0100"},{"line_number":6,"context_line":""},{"line_number":7,"context_line":"Handle process disappearing before we ask for its PPID"},{"line_number":8,"context_line":""},{"line_number":9,"context_line":"Change-Id: I573aba8e11dca16f8a6565f7e9704be18e938566"},{"line_number":10,"context_line":"Closes-Bug: #1478190"}],"source_content_type":"text/x-gerrit-commit-message","patch_set":1,"id":"da20952f_2c3142a2","line":7,"in_reply_to":"da20952f_f926ae65","updated":"2015-09-07 14:51:58.000000000","message":"Ihar - Yes, that is my understanding too; thanks for writing this.\n\nArmando - I\u0027ll add Elena as a reviewer and also try to catch her on IRC later...","commit_id":"42f80682d3eff58af60199f817ac402f457491a0"}],"neutron/tests/functional/test_server.py":[{"author":{"_account_id":2874,"name":"yong sheng gong","email":"gong.yongsheng@99cloud.net","username":"gongysh"},"change_message_id":"4903a1e25baa74539baf49fdb8896026958b20db","unresolved":false,"context_lines":[{"line_number":116,"context_line":"                    if safe_ppid(proc) \u003d\u003d self.service_pid]"},{"line_number":117,"context_line":"        else:"},{"line_number":118,"context_line":"            return [proc.pid for proc in psutil.process_iter()"},{"line_number":119,"context_line":"                    if proc.pid \u003d\u003d self.service_pid]"},{"line_number":120,"context_line":""},{"line_number":121,"context_line":"    def _fake_reset(self):"},{"line_number":122,"context_line":"        \"\"\"Writes FAKE_RESET_MSG to temporary file on each call.\"\"\""}],"source_content_type":"text/x-python","patch_set":1,"id":"da20952f_8b36594f","line":119,"updated":"2015-09-06 02:09:27.000000000","message":"what about this proc.pid?","commit_id":"42f80682d3eff58af60199f817ac402f457491a0"},{"author":{"_account_id":13734,"name":"Nell Jerram","email":"nell@tigera.io","username":"neiljerram"},"change_message_id":"f5151485998c6b3db008c295febc873a46da850b","unresolved":false,"context_lines":[{"line_number":116,"context_line":"                    if safe_ppid(proc) \u003d\u003d self.service_pid]"},{"line_number":117,"context_line":"        else:"},{"line_number":118,"context_line":"            return [proc.pid for proc in psutil.process_iter()"},{"line_number":119,"context_line":"                    if proc.pid \u003d\u003d self.service_pid]"},{"line_number":120,"context_line":""},{"line_number":121,"context_line":"    def _fake_reset(self):"},{"line_number":122,"context_line":"        \"\"\"Writes FAKE_RESET_MSG to temporary file on each call.\"\"\""}],"source_content_type":"text/x-python","patch_set":1,"id":"da20952f_f3c6eb66","line":119,"in_reply_to":"da20952f_8b36594f","updated":"2015-09-06 09:59:56.000000000","message":"Indeed, interesting question.  However:\n\n- In practice we\u0027re not seeing any test failures (AFAIK) involving the proc.pid case, and I\u0027d prefer not to \u0027fix\u0027 a problem that we don\u0027t know we have.\n\n- I would guess - without going back to read the psutil code in detail - that proc.pid is already set up by psutil.process_iter(), and doesn\u0027t involve any new processing when proc.pid is referenced (as here).\n\nSo, I guess that the .pid case is genuinely different and doesn\u0027t need a similar fix as .ppid does.","commit_id":"42f80682d3eff58af60199f817ac402f457491a0"},{"author":{"_account_id":13734,"name":"Nell Jerram","email":"nell@tigera.io","username":"neiljerram"},"change_message_id":"39826b4c864d336c22914a174ba544705325cf36","unresolved":false,"context_lines":[{"line_number":116,"context_line":"                    if safe_ppid(proc) \u003d\u003d self.service_pid]"},{"line_number":117,"context_line":"        else:"},{"line_number":118,"context_line":"            return [proc.pid for proc in psutil.process_iter()"},{"line_number":119,"context_line":"                    if proc.pid \u003d\u003d self.service_pid]"},{"line_number":120,"context_line":""},{"line_number":121,"context_line":"    def _fake_reset(self):"},{"line_number":122,"context_line":"        \"\"\"Writes FAKE_RESET_MSG to temporary file on each call.\"\"\""}],"source_content_type":"text/x-python","patch_set":1,"id":"da20952f_ccb1a602","line":119,"in_reply_to":"da20952f_b9c11682","updated":"2015-09-07 14:51:58.000000000","message":"I\u0027m wary about complicating the code for a problem that we don\u0027t and may never have, though...  I personally favour not changing the code when we don\u0027t have to, although these situations are tricky to judge.","commit_id":"42f80682d3eff58af60199f817ac402f457491a0"},{"author":{"_account_id":9656,"name":"Ihar Hrachyshka","email":"ihrachys@redhat.com","username":"ihrachys","status":"Red Hat Networking Systems Engineer"},"change_message_id":"a480af4234d0ea05f24689ccb5216d749b11ec63","unresolved":false,"context_lines":[{"line_number":116,"context_line":"                    if safe_ppid(proc) \u003d\u003d self.service_pid]"},{"line_number":117,"context_line":"        else:"},{"line_number":118,"context_line":"            return [proc.pid for proc in psutil.process_iter()"},{"line_number":119,"context_line":"                    if proc.pid \u003d\u003d self.service_pid]"},{"line_number":120,"context_line":""},{"line_number":121,"context_line":"    def _fake_reset(self):"},{"line_number":122,"context_line":"        \"\"\"Writes FAKE_RESET_MSG to temporary file on each call.\"\"\""}],"source_content_type":"text/x-python","patch_set":1,"id":"da20952f_b9c11682","line":119,"in_reply_to":"da20952f_f3c6eb66","updated":"2015-09-07 14:26:08.000000000","message":"Yes, .pid is cached in Process class: https://code.google.com/p/psutil/source/browse/psutil/__init__.py#464 Though I think it would be more fail proof if we guard against it too since it\u0027s not part of psutil public API.","commit_id":"42f80682d3eff58af60199f817ac402f457491a0"}]}
