)]}'
{"lib/ceph":[{"author":{"_account_id":7118,"name":"Ian Wienand","email":"iwienand@redhat.com","username":"iwienand"},"change_message_id":"a4d4857adb0f65866272a7a2cd7edbd46377b061","unresolved":false,"context_lines":[{"line_number":92,"context_line":"function cleanup_ceph {"},{"line_number":93,"context_line":"    for proc in ceph-mon ceph-osd; do"},{"line_number":94,"context_line":"        sudo pkill -f $proc"},{"line_number":95,"context_line":"        if ! timeout 15 sh -c \"while sudo pkill --signal 0 $proc; do sleep 1; done\"; then"},{"line_number":96,"context_line":"            err $LINENO \"$proc has not shutdown in the expected timeframe.\""},{"line_number":97,"context_line":"        fi"},{"line_number":98,"context_line":"    done"}],"source_content_type":"application/octet-stream","patch_set":1,"id":"fa98f980_e205dd15","line":95,"updated":"2014-09-19 01:02:45.000000000","message":"In general i\u0027d prefer not to spawn shells; things like set -e don\u0027t go down to them and we\u0027ve had cases where we miss things because they\u0027re in subshells.\n\nwe already have is_running (although unused, it seems).  can we just use that and a counting loop?","commit_id":"81e8ba21b5ba5d3a502a8e249c2efea0b426949a"},{"author":{"_account_id":7175,"name":"Robbie Harwood","email":"rharwood@redhat.com","username":"frozencemetery"},"change_message_id":"97073a064140a371dfbb8321aaf42cfac8556d98","unresolved":false,"context_lines":[{"line_number":93,"context_line":"    for proc in ceph-mon ceph-osd; do"},{"line_number":94,"context_line":"        sudo pkill -f $proc"},{"line_number":95,"context_line":"        if ! timeout 15 sh -c \"while sudo pkill --signal 0 $proc; do sleep 1; done\"; then"},{"line_number":96,"context_line":"            err $LINENO \"$proc has not shutdown in the expected timeframe.\""},{"line_number":97,"context_line":"        fi"},{"line_number":98,"context_line":"    done"},{"line_number":99,"context_line":""}],"source_content_type":"application/octet-stream","patch_set":1,"id":"fa98f980_9aa4634c","line":96,"updated":"2014-09-19 20:28:28.000000000","message":"I think it would be good to explicitly indicate that we\u0027re continuing here, though it seems to me that since this is cleanup we ought to continue.","commit_id":"81e8ba21b5ba5d3a502a8e249c2efea0b426949a"},{"author":{"_account_id":7118,"name":"Ian Wienand","email":"iwienand@redhat.com","username":"iwienand"},"change_message_id":"a4d4857adb0f65866272a7a2cd7edbd46377b061","unresolved":false,"context_lines":[{"line_number":93,"context_line":"    for proc in ceph-mon ceph-osd; do"},{"line_number":94,"context_line":"        sudo pkill -f $proc"},{"line_number":95,"context_line":"        if ! timeout 15 sh -c \"while sudo pkill --signal 0 $proc; do sleep 1; done\"; then"},{"line_number":96,"context_line":"            err $LINENO \"$proc has not shutdown in the expected timeframe.\""},{"line_number":97,"context_line":"        fi"},{"line_number":98,"context_line":"    done"},{"line_number":99,"context_line":""}],"source_content_type":"application/octet-stream","patch_set":1,"id":"fa98f980_021f41b8","line":96,"updated":"2014-09-19 01:02:45.000000000","message":"shouldn\u0027t we die() here?","commit_id":"81e8ba21b5ba5d3a502a8e249c2efea0b426949a"},{"author":{"_account_id":4523,"name":"Eric Harney","email":"eharney@redhat.com","username":"eharney"},"change_message_id":"d7b8867ad984646aa17a01a11e2ab7faee2b55af","unresolved":false,"context_lines":[{"line_number":93,"context_line":"    for proc in ceph-mon ceph-osd; do"},{"line_number":94,"context_line":"        sudo pkill -f $proc"},{"line_number":95,"context_line":"        if ! timeout 15 sh -c \"while sudo pkill --signal 0 $proc; do sleep 1; done\"; then"},{"line_number":96,"context_line":"            err $LINENO \"$proc has not shutdown in the expected timeframe.\""},{"line_number":97,"context_line":"        fi"},{"line_number":98,"context_line":"    done"},{"line_number":99,"context_line":""}],"source_content_type":"application/octet-stream","patch_set":1,"id":"da9df570_9eab5374","line":96,"in_reply_to":"da9df570_458d1c7b","updated":"2014-09-25 20:57:23.000000000","message":"Sébastien: I think it\u0027s still a race even doing that.  (Not as likely of one, but still.)","commit_id":"81e8ba21b5ba5d3a502a8e249c2efea0b426949a"},{"author":{"_account_id":4523,"name":"Eric Harney","email":"eharney@redhat.com","username":"eharney"},"change_message_id":"81f85119f3541a242f8399fcc6ed10e160714e77","unresolved":false,"context_lines":[{"line_number":93,"context_line":"    for proc in ceph-mon ceph-osd; do"},{"line_number":94,"context_line":"        sudo pkill -f $proc"},{"line_number":95,"context_line":"        if ! timeout 15 sh -c \"while sudo pkill --signal 0 $proc; do sleep 1; done\"; then"},{"line_number":96,"context_line":"            err $LINENO \"$proc has not shutdown in the expected timeframe.\""},{"line_number":97,"context_line":"        fi"},{"line_number":98,"context_line":"    done"},{"line_number":99,"context_line":""}],"source_content_type":"application/octet-stream","patch_set":1,"id":"fa98f980_e0706a42","line":96,"in_reply_to":"fa98f980_021f41b8","updated":"2014-09-19 12:48:56.000000000","message":"I considered this and was kind of torn on this question... it seems like it may be better to continue and clean up as much as possible rather than abort the cleanup process early?  (No worse than today, at least.)","commit_id":"81e8ba21b5ba5d3a502a8e249c2efea0b426949a"},{"author":{"_account_id":6984,"name":"Sébastien Han","email":"seb@redhat.com","username":"leseb"},"change_message_id":"bf35a95efb2382609202bcfa7160342404a121ab","unresolved":false,"context_lines":[{"line_number":93,"context_line":"    for proc in ceph-mon ceph-osd; do"},{"line_number":94,"context_line":"        sudo pkill -f $proc"},{"line_number":95,"context_line":"        if ! timeout 15 sh -c \"while sudo pkill --signal 0 $proc; do sleep 1; done\"; then"},{"line_number":96,"context_line":"            err $LINENO \"$proc has not shutdown in the expected timeframe.\""},{"line_number":97,"context_line":"        fi"},{"line_number":98,"context_line":"    done"},{"line_number":99,"context_line":""}],"source_content_type":"application/octet-stream","patch_set":1,"id":"da9df570_458d1c7b","line":96,"in_reply_to":"fa98f980_9aa4634c","updated":"2014-09-21 23:18:48.000000000","message":"why don\u0027t we just \"pkill -9\" and avoid the loop? Since it\u0027s a cleanup I\u0027m not sure if we should gracefully wait for the process to die...","commit_id":"81e8ba21b5ba5d3a502a8e249c2efea0b426949a"},{"author":{"_account_id":970,"name":"Dean Troyer","email":"dtroyer@gmail.com","username":"dtroyer"},"change_message_id":"d9dcccafa10907aef07f1daa1e92941c96914190","unresolved":false,"context_lines":[{"line_number":143,"context_line":""},{"line_number":144,"context_line":"function cleanup_ceph_embedded {"},{"line_number":145,"context_line":"    sudo killall -w -9 ceph-mon"},{"line_number":146,"context_line":"    sudo killall -w -9 ceph-osd"},{"line_number":147,"context_line":"    sudo rm -rf ${CEPH_DATA_DIR}/*/*"},{"line_number":148,"context_line":"    if egrep -q ${CEPH_DATA_DIR} /proc/mounts; then"},{"line_number":149,"context_line":"        sudo umount ${CEPH_DATA_DIR}"}],"source_content_type":"application/x-shellscript","patch_set":2,"id":"fa81d914_30e8fee1","line":146,"updated":"2015-02-03 09:20:28.000000000","message":"Is the SIGKILL necessary?  The combination of -w and -9 seems odd.","commit_id":"a27b74cdf74a83ef2d197d026526211d12990228"}]}