)]}'
{"/COMMIT_MSG":[{"author":{"_account_id":15343,"name":"Tim Burke","email":"tburke@nvidia.com","username":"tburke"},"change_message_id":"ca9075b35a062aedeac4a12a3079064f50714450","unresolved":false,"context_lines":[{"line_number":20,"context_line":"If we see a 507 from a primary and we\u0027re not in the handoff list (we\u0027re"},{"line_number":21,"context_line":"an old primary post rebalance) it\u0027d probably be not so terrible to try"},{"line_number":22,"context_line":"to revert it to the first handoff if it\u0027s not already holding a part."},{"line_number":23,"context_line":"But that\u0027s more work and sounds more like lp bug #1510342"},{"line_number":24,"context_line":""},{"line_number":25,"context_line":"Closes-Bug: #1651530"},{"line_number":26,"context_line":""}],"source_content_type":"text/x-gerrit-commit-message","patch_set":2,"id":"3a461143_dc49c4b8","line":23,"updated":"2017-01-30 21:02:08.000000000","message":"Seems like we\u0027re harming availability if we don\u0027t do this part?","commit_id":"c8723f8bdc4f36dcd58e71de89c1c91633427b03"},{"author":{"_account_id":15343,"name":"Tim Burke","email":"tburke@nvidia.com","username":"tburke"},"change_message_id":"ca9075b35a062aedeac4a12a3079064f50714450","unresolved":false,"context_lines":[{"line_number":22,"context_line":"to revert it to the first handoff if it\u0027s not already holding a part."},{"line_number":23,"context_line":"But that\u0027s more work and sounds more like lp bug #1510342"},{"line_number":24,"context_line":""},{"line_number":25,"context_line":"Closes-Bug: #1651530"},{"line_number":26,"context_line":""},{"line_number":27,"context_line":"Change-Id: Ie351d8342fc8e589b143f981e95ce74e70e52784"}],"source_content_type":"text/x-gerrit-commit-message","patch_set":2,"id":"3a461143_9cd01cf8","line":25,"updated":"2017-01-30 21:02:08.000000000","message":"Is that really the right bug? Seems unrelated. At the very least, Closes-Bug is a bit much -- Related-Bug or Partial-Bug might be better?","commit_id":"c8723f8bdc4f36dcd58e71de89c1c91633427b03"},{"author":{"_account_id":1179,"name":"Clay Gerrard","email":"clay.gerrard@gmail.com","username":"clay-gerrard"},"change_message_id":"3ce7bbc8975c9fc217e46169fa7faebe3afffe0c","unresolved":false,"context_lines":[{"line_number":22,"context_line":"to revert it to the first handoff if it\u0027s not already holding a part."},{"line_number":23,"context_line":"But that\u0027s more work and sounds more like lp bug #1510342"},{"line_number":24,"context_line":""},{"line_number":25,"context_line":"Closes-Bug: #1651530"},{"line_number":26,"context_line":""},{"line_number":27,"context_line":"Change-Id: Ie351d8342fc8e589b143f981e95ce74e70e52784"}],"source_content_type":"text/x-gerrit-commit-message","patch_set":2,"id":"3a461143_121973d7","line":25,"updated":"2017-01-30 21:52:00.000000000","message":"it should be 1653169\n\nhttps://bugs.launchpad.net/swift/+bug/1653169","commit_id":"c8723f8bdc4f36dcd58e71de89c1c91633427b03"}],"swift/obj/reconstructor.py":[{"author":{"_account_id":15343,"name":"Tim Burke","email":"tburke@nvidia.com","username":"tburke"},"change_message_id":"ca9075b35a062aedeac4a12a3079064f50714450","unresolved":false,"context_lines":[{"line_number":645,"context_line":"            \u0027partition.delete.count.%s\u0027 % (job[\u0027local_dev\u0027][\u0027device\u0027],))"},{"line_number":646,"context_line":"        # we\u0027d desperately like to push this partition back to it\u0027s"},{"line_number":647,"context_line":"        # primary location, but if that node is down, the next best thing"},{"line_number":648,"context_line":"        # is one of the handoff locations - which *might* be us already!"},{"line_number":649,"context_line":"        dest_nodes \u003d itertools.chain("},{"line_number":650,"context_line":"            job[\u0027sync_to\u0027],"},{"line_number":651,"context_line":"            job[\u0027policy\u0027].object_ring.get_more_nodes(job[\u0027partition\u0027]),"}],"source_content_type":"text/x-python","patch_set":2,"id":"3a461143_39973a3b","side":"PARENT","line":648,"updated":"2017-01-30 21:02:08.000000000","message":"Well that\u0027s been there a while [1]. Anybody got insight on why we *ever* thought this was a good idea? Just looking for the comment, this seems to have come in somewhere between PS 5 and 6 [2] -- ring any bells for anyone?\n\nI could kinda see it being a good idea if *this* node isn\u0027t a handoff -- if an old handoff can at least get it to a current one (even if it can\u0027t contact a primary), that seems like an improvement -- at least a proxy can find it, so we\u0027ve helped availability? OTOH, if that means the (current) handoff now has doubled-up fragments, we still need to worry about durability. I guess that\u0027s the known issue you mentioned in the commit message.\n\n[1] https://github.com/openstack/swift/commit/647b66a\n[2] https://review.openstack.org/#/c/170339/5..6/swift/obj/reconstructor.py","commit_id":"2914e0449311482c91ccd61fc0f335a0579afa33"},{"author":{"_account_id":15343,"name":"Tim Burke","email":"tburke@nvidia.com","username":"tburke"},"change_message_id":"ca9075b35a062aedeac4a12a3079064f50714450","unresolved":false,"context_lines":[{"line_number":648,"context_line":"        for node in job[\u0027sync_to\u0027]:"},{"line_number":649,"context_line":"            if syncd_with \u003e\u003d len(job[\u0027sync_to\u0027]):"},{"line_number":650,"context_line":"                break"},{"line_number":651,"context_line":"            if node[\u0027id\u0027] \u003d\u003d job[\u0027local_dev\u0027][\u0027id\u0027]:"},{"line_number":652,"context_line":"                # this is as good a place as any for this data for now"},{"line_number":653,"context_line":"                break"},{"line_number":654,"context_line":"            success, in_sync_objs \u003d ssync_sender("}],"source_content_type":"text/x-python","patch_set":2,"id":"3a461143_fcb760d5","line":651,"updated":"2017-01-30 21:02:08.000000000","message":"Is this check still needed?","commit_id":"c8723f8bdc4f36dcd58e71de89c1c91633427b03"}]}