== replay-dual test 26: dbench and tar with mds failover ========================================================== 16:25:27 (1710102327) Starting client oleg401-client.virtnet: -o user_xattr,flock oleg401-server@tcp:/lustre /mnt/lustre Started clients oleg401-client.virtnet: 192.168.204.101@tcp:/lustre on /mnt/lustre type lustre (rw,checksum,flock,user_xattr,lruresize,lazystatfs,nouser_fid2path,verbose,noencrypt,statfs_project) Started tar loop with pid 6810 Started dbench loop with 6812 striped dir -i0 -c2 -H fnv_1a_64 /mnt/lustre/d26.replay-dual/run_tar striped dir -i0 -c2 -H crush2 /mnt/lustre2/d26.replay-dual/run_dbench looking for dbench program /usr/bin/dbench found dbench client file /usr/share/dbench/client.txt '/usr/share/dbench/client.txt' -> 'client.txt' running 'dbench 1 -D /mnt/lustre2/d26.replay-dual/run_dbench -t 100' on /mnt/lustre at Sun Mar 10 16:25:30 EDT 2024 waiting for dbench pid 6861 dbench version 4.00 - Copyright Andrew Tridgell 1999-2004 Running for 100 seconds with load 'client.txt' and minimum warmup 20 secs failed to create barrier semaphore 0 of 1 processes prepared for launch 0 sec UUID 1K-blocks Used Available Use% Mounted on lustre-MDT0000_UUID 1414116 2072 1285616 1% /mnt/lustre[MDT:0] lustre-MDT0001_UUID 1414116 1880 1285808 1% /mnt/lustre[MDT:1] 1 of 1 processes prepared for launch 0 sec releasing clients lustre-OST0000_UUID 3833116 1816 3605204 1% /mnt/lustre[OST:0] lustre-OST0001_UUID 3833116 9988 3572172 1% /mnt/lustre[OST:1] filesystem_summary: 7666232 11804 7177376 1% /mnt/lustre 1 91 4.19 MB/sec warmup 1 sec latency 101.606 ms 1 154 3.27 MB/sec warmup 2 sec latency 103.656 ms 1 231 2.96 MB/sec warmup 3 sec latency 88.645 ms 1 312 2.87 MB/sec warmup 4 sec latency 78.505 ms 1 397 2.81 MB/sec warmup 5 sec latency 86.282 ms test_26 fail mds1 1 times 1 485 2.82 MB/sec warmup 6 sec latency 56.014 ms Failing mds1 on oleg401-server 1 583 2.84 MB/sec warmup 7 sec latency 88.232 ms Stopping /mnt/lustre-mds1 (opts:) on oleg401-server 1 671 2.64 MB/sec warmup 8 sec latency 100.723 ms 1 674 2.35 MB/sec warmup 9 sec latency 949.897 ms 16:25:40 (1710102340) shut down 1 674 2.11 MB/sec warmup 10 sec latency 1951.631 ms 1 674 1.92 MB/sec warmup 11 sec latency 2951.800 ms 1 674 1.76 MB/sec warmup 12 sec latency 3952.403 ms 1 674 1.63 MB/sec warmup 13 sec latency 4953.123 ms 1 674 1.51 MB/sec warmup 14 sec latency 5953.308 ms 1 674 1.41 MB/sec warmup 15 sec latency 6955.296 ms 1 674 1.32 MB/sec warmup 16 sec latency 7955.483 ms 1 674 1.24 MB/sec warmup 17 sec latency 8955.664 ms 1 674 1.18 MB/sec warmup 18 sec latency 9955.918 ms 1 674 1.11 MB/sec warmup 19 sec latency 10956.976 ms Failover mds1 to oleg401-server mount facets: mds1 1 674 0.00 MB/sec execute 1 sec latency 12959.089 ms 1 674 0.00 MB/sec execute 2 sec latency 13962.931 ms 1 674 0.00 MB/sec execute 3 sec latency 14966.125 ms 1 674 0.00 MB/sec execute 4 sec latency 15966.586 ms Starting mds1: -o localrecov /dev/mapper/mds1_flakey /mnt/lustre-mds1 1 674 0.00 MB/sec execute 5 sec latency 16966.847 ms 1 674 0.00 MB/sec execute 6 sec latency 17970.542 ms oleg401-server: oleg401-server.virtnet: executing set_default_debug -1 all 1 674 0.00 MB/sec execute 7 sec latency 18973.163 ms pdsh@oleg401-client: oleg401-server: ssh exited with exit code 1 Started lustre-MDT0000 16:25:58 (1710102358) targets are mounted 16:25:58 (1710102358) facet_failover done 1 674 0.00 MB/sec execute 8 sec latency 19989.834 ms 1 674 0.00 MB/sec execute 9 sec latency 20989.985 ms 1 674 0.00 MB/sec execute 10 sec latency 21990.430 ms 1 678 0.00 MB/sec execute 11 sec latency 22933.003 ms oleg401-client.virtnet: executing wait_import_state_mount (FULL|IDLE) mdc.lustre-MDT0000-mdc-*.mds_server_uuid 1 747 0.01 MB/sec execute 12 sec latency 90.249 ms mdc.lustre-MDT0000-mdc-*.mds_server_uuid in FULL state after 0 sec 1 815 0.02 MB/sec execute 13 sec latency 91.331 ms 1 902 0.04 MB/sec execute 14 sec latency 78.528 ms 1 967 0.04 MB/sec execute 15 sec latency 66.508 ms 1 1063 0.11 MB/sec execute 16 sec latency 80.449 ms 1 1167 0.11 MB/sec execute 17 sec latency 40.427 ms UUID 1K-blocks Used Available Use% Mounted on lustre-MDT0000_UUID 1414116 2524 1285164 1% /mnt/lustre[MDT:0] lustre-MDT0001_UUID 1414116 2348 1285340 1% /mnt/lustre[MDT:1] lustre-OST0000_UUID 3833116 7356 3593520 1% /mnt/lustre[OST:0] lustre-OST0001_UUID 3833116 36748 3564384 2% /mnt/lustre[OST:1] filesystem_summary: 7666232 44104 7157904 1% /mnt/lustre 1 1251 0.11 MB/sec execute 18 sec latency 57.122 ms 1 1418 0.25 MB/sec execute 19 sec latency 51.025 ms 1 1499 0.24 MB/sec execute 20 sec latency 56.376 ms 1 1566 0.23 MB/sec execute 21 sec latency 87.770 ms 1 1641 0.23 MB/sec execute 22 sec latency 66.379 ms 1 1721 0.22 MB/sec execute 23 sec latency 60.598 ms test_26 fail mds2 2 times 1 1821 0.21 MB/sec execute 24 sec latency 68.827 ms Failing mds2 on oleg401-server Stopping /mnt/lustre-mds2 (opts:) on oleg401-server 1 1888 0.21 MB/sec execute 25 sec latency 64.090 ms 1 1975 0.20 MB/sec execute 26 sec latency 340.634 ms 16:26:17 (1710102377) shut down 1 1975 0.20 MB/sec execute 27 sec latency 1340.802 ms 1 1975 0.19 MB/sec execute 28 sec latency 2343.709 ms 1 1975 0.18 MB/sec execute 29 sec latency 3343.926 ms 1 1975 0.18 MB/sec execute 30 sec latency 4344.372 ms 1 1975 0.17 MB/sec execute 31 sec latency 5344.647 ms 1 1975 0.16 MB/sec execute 32 sec latency 6345.256 ms 1 1975 0.16 MB/sec execute 33 sec latency 7345.537 ms 1 1975 0.15 MB/sec execute 34 sec latency 8345.804 ms 1 1975 0.15 MB/sec execute 35 sec latency 9346.575 ms 1 1975 0.15 MB/sec execute 36 sec latency 10346.766 ms Failover mds2 to oleg401-server mount facets: mds2 1 1975 0.14 MB/sec execute 37 sec latency 11346.969 ms 1 1975 0.14 MB/sec execute 38 sec latency 12347.144 ms 1 1975 0.14 MB/sec execute 39 sec latency 13348.444 ms 1 1975 0.13 MB/sec execute 40 sec latency 14348.618 ms 1 1975 0.13 MB/sec execute 41 sec latency 15357.606 ms Starting mds2: -o localrecov /dev/mapper/mds2_flakey /mnt/lustre-mds2 1 1975 0.13 MB/sec execute 42 sec latency 16357.828 ms 1 1975 0.12 MB/sec execute 43 sec latency 17359.016 ms 1 1975 0.12 MB/sec execute 44 sec latency 18359.219 ms oleg401-server: oleg401-server.virtnet: executing set_default_debug -1 all 1 1975 0.12 MB/sec execute 45 sec latency 19359.450 ms pdsh@oleg401-client: oleg401-server: ssh exited with exit code 1 Started lustre-MDT0001 16:26:36 (1710102396) targets are mounted 16:26:36 (1710102396) facet_failover done 1 1975 0.11 MB/sec execute 46 sec latency 20359.632 ms 1 1975 0.11 MB/sec execute 47 sec latency 21361.876 ms 1 1975 0.11 MB/sec execute 48 sec latency 22362.268 ms 1 2006 0.11 MB/sec execute 49 sec latency 22951.609 ms oleg401-client.virtnet: executing wait_import_state_mount (FULL|IDLE) mdc.lustre-MDT0001-mdc-*.mds_server_uuid 1 2094 0.11 MB/sec execute 50 sec latency 52.190 ms mdc.lustre-MDT0001-mdc-*.mds_server_uuid in FULL state after 0 sec 1 2150 0.11 MB/sec execute 51 sec latency 130.343 ms 1 2245 0.11 MB/sec execute 52 sec latency 56.034 ms 1 2359 0.13 MB/sec execute 53 sec latency 63.505 ms 1 2435 0.13 MB/sec execute 54 sec latency 62.357 ms 1 2554 0.14 MB/sec execute 55 sec latency 80.778 ms UUID 1K-blocks Used Available Use% Mounted on lustre-MDT0000_UUID 1414116 2644 1285044 1% /mnt/lustre[MDT:0] lustre-MDT0001_UUID 1414116 2496 1285192 1% /mnt/lustre[MDT:1] lustre-OST0000_UUID 3833116 9668 3595584 1% /mnt/lustre[OST:0] lustre-OST0001_UUID 3833116 41176 3563768 2% /mnt/lustre[OST:1] filesystem_summary: 7666232 50844 7159352 1% /mnt/lustre 1 2612 0.14 MB/sec execute 56 sec latency 71.715 ms 1 2896 0.20 MB/sec execute 57 sec latency 70.093 ms 1 2983 0.21 MB/sec execute 58 sec latency 69.750 ms 1 3172 0.22 MB/sec execute 59 sec latency 146.521 ms 1 3313 0.24 MB/sec execute 60 sec latency 57.793 ms 1 3472 0.27 MB/sec execute 61 sec latency 72.189 ms test_26 fail mds1 3 times 1 3577 0.28 MB/sec execute 62 sec latency 51.348 ms 1 3653 0.28 MB/sec execute 63 sec latency 95.749 ms Failing mds1 on oleg401-server Stopping /mnt/lustre-mds1 (opts:) on oleg401-server 1 3747 0.28 MB/sec execute 64 sec latency 71.693 ms 1 3820 0.29 MB/sec execute 65 sec latency 70.017 ms 1 3885 0.28 MB/sec execute 66 sec latency 44.057 ms 1 3963 0.28 MB/sec execute 67 sec latency 51.369 ms 1 4026 0.28 MB/sec execute 68 sec latency 84.151 ms 1 4083 0.27 MB/sec execute 69 sec latency 280.522 ms 1 4083 0.27 MB/sec execute 70 sec latency 1280.812 ms 16:27:01 (1710102421) shut down 1 4083 0.27 MB/sec execute 71 sec latency 2281.037 ms 1 4083 0.26 MB/sec execute 72 sec latency 3281.255 ms 1 4083 0.26 MB/sec execute 73 sec latency 4281.487 ms 1 4083 0.26 MB/sec execute 74 sec latency 5282.830 ms 1 4083 0.25 MB/sec execute 75 sec latency 6283.676 ms 1 4083 0.25 MB/sec execute 76 sec latency 7283.878 ms 1 4083 0.25 MB/sec execute 77 sec latency 8284.120 ms 1 4083 0.24 MB/sec execute 78 sec latency 9284.403 ms 1 4083 0.24 MB/sec execute 79 sec latency 10286.545 ms 1 4083 0.24 MB/sec execute 80 sec latency 11287.127 ms 1 4083 0.23 MB/sec execute 81 sec latency 12291.644 ms Failover mds1 to oleg401-server mount facets: mds1 1 4083 0.23 MB/sec execute 82 sec latency 13291.877 ms 1 4083 0.23 MB/sec execute 83 sec latency 14292.055 ms 1 4083 0.22 MB/sec execute 84 sec latency 15293.275 ms 1 4083 0.22 MB/sec execute 85 sec latency 16304.251 ms Starting mds1: -o localrecov /dev/mapper/mds1_flakey /mnt/lustre-mds1 1 4083 0.22 MB/sec execute 86 sec latency 17304.433 ms 1 4083 0.22 MB/sec execute 87 sec latency 18304.742 ms 1 4083 0.21 MB/sec execute 88 sec latency 19305.045 ms oleg401-server: oleg401-server.virtnet: executing set_default_debug -1 all pdsh@oleg401-client: oleg401-server: ssh exited with exit code 1 1 4083 0.21 MB/sec execute 89 sec latency 20307.275 ms Started lustre-MDT0000 16:27:20 (1710102440) targets are mounted 16:27:20 (1710102440) facet_failover done 1 4083 0.21 MB/sec execute 90 sec latency 21310.177 ms 1 4083 0.21 MB/sec execute 91 sec latency 22310.372 ms 1 4083 0.21 MB/sec execute 92 sec latency 23310.613 ms 1 4083 0.20 MB/sec execute 93 sec latency 24310.982 ms 1 4141 0.20 MB/sec execute 94 sec latency 24749.861 ms oleg401-client.virtnet: executing wait_import_state_mount (FULL|IDLE) mdc.lustre-MDT0000-mdc-*.mds_server_uuid 1 4238 0.20 MB/sec execute 95 sec latency 95.697 ms mdc.lustre-MDT0000-mdc-*.mds_server_uuid in FULL state after 0 sec 1 4302 0.20 MB/sec execute 96 sec latency 83.203 ms 1 4415 0.20 MB/sec execute 97 sec latency 42.452 ms 1 4490 0.20 MB/sec execute 98 sec latency 48.448 ms 1 4597 0.21 MB/sec execute 99 sec latency 47.931 ms UUID 1K-blocks Used Available Use% Mounted on lustre-MDT0000_UUID 1414116 2928 1284760 1% /mnt/lustre[MDT:0] lustre-MDT0001_UUID 1414116 2592 1285096 1% /mnt/lustre[MDT:1] lustre-OST0000_UUID 3833116 12088 3590388 1% /mnt/lustre[OST:0] lustre-OST0001_UUID 3833116 41452 3564540 2% /mnt/lustre[OST:1] filesystem_summary: 7666232 53540 7154928 1% /mnt/lustre 1 cleanup 100 sec 1 cleanup 101 sec 0 cleanup 101 sec Operation Count AvgLat MaxLat ---------------------------------------- NTCreateX 692 29.292 130.315 Close 524 7.750 50.104 Rename 33 43.121 83.180 Unlink 135 12.372 42.584 Qpathinfo 633 47.772 24749.829 Qfileinfo 107 0.866 9.819 Qfsinfo 122 2.878 21.980 Sfileinfo 71 345.203 22932.985 Find 241 101.823 22951.587 WriteX 364 6.096 63.482 ReadX 1053 0.644 46.402 LockX 2 7.062 8.368 UnlockX 2 5.038 6.706 Flush 62 31.570 95.679 Throughput 0.210089 MB/sec 1 clients 1 procs max_latency=24749.861 ms stopping dbench on /mnt/lustre at Sun Mar 10 16:27:32 EDT 2024 with return code 0 clean dbench files on /mnt/lustre /mnt/lustre /mnt/lustre removed 'client.txt' /mnt/lustre dbench successfully finished striped dir -i0 -c2 -H all_char /mnt/lustre2/d26.replay-dual/run_dbench looking for dbench program /usr/bin/dbench found dbench client file /usr/share/dbench/client.txt '/usr/share/dbench/client.txt' -> 'client.txt' running 'dbench 1 -D /mnt/lustre2/d26.replay-dual/run_dbench -t 100' on /mnt/lustre at Sun Mar 10 16:27:35 EDT 2024 waiting for dbench pid 10313 dbench version 4.00 - Copyright Andrew Tridgell 1999-2004 Running for 100 seconds with load 'client.txt' and minimum warmup 20 secs 0 of 1 processes prepared for launch 0 sec 1 of 1 processes prepared for launch 0 sec releasing clients test_26 fail mds2 4 times 1 91 4.19 MB/sec warmup 1 sec latency 52.192 ms Failing mds2 on oleg401-server 1 157 3.27 MB/sec warmup 2 sec latency 101.999 ms Stopping /mnt/lustre-mds2 (opts:) on oleg401-server 1 246 3.13 MB/sec warmup 3 sec latency 81.184 ms 1 326 2.93 MB/sec warmup 4 sec latency 66.209 ms 1 339 2.45 MB/sec warmup 5 sec latency 925.241 ms 1 339 2.04 MB/sec warmup 6 sec latency 1925.377 ms 1 339 1.75 MB/sec warmup 7 sec latency 2925.987 ms 1 339 1.53 MB/sec warmup 8 sec latency 3926.215 ms 1 339 1.36 MB/sec warmup 9 sec latency 4926.826 ms 16:27:45 (1710102465) shut down 1 339 1.23 MB/sec warmup 10 sec latency 5927.506 ms 1 339 1.11 MB/sec warmup 11 sec latency 6941.849 ms 1 339 1.02 MB/sec warmup 12 sec latency 7942.065 ms 1 339 0.94 MB/sec warmup 13 sec latency 8942.468 ms 1 339 0.87 MB/sec warmup 14 sec latency 9943.308 ms 1 339 0.82 MB/sec warmup 15 sec latency 10943.474 ms 1 339 0.77 MB/sec warmup 16 sec latency 11946.950 ms 1 339 0.72 MB/sec warmup 17 sec latency 12947.372 ms 1 339 0.68 MB/sec warmup 18 sec latency 13948.082 ms 1 339 0.64 MB/sec warmup 19 sec latency 14948.677 ms Failover mds2 to oleg401-server mount facets: mds2 1 339 0.00 MB/sec execute 1 sec latency 16948.996 ms 1 339 0.00 MB/sec execute 2 sec latency 17964.867 ms 1 339 0.00 MB/sec execute 3 sec latency 18965.060 ms 1 339 0.00 MB/sec execute 4 sec latency 19965.391 ms Starting mds2: -o localrecov /dev/mapper/mds2_flakey /mnt/lustre-mds2 1 339 0.00 MB/sec execute 5 sec latency 20966.373 ms 1 339 0.00 MB/sec execute 6 sec latency 21966.722 ms oleg401-server: oleg401-server.virtnet: executing set_default_debug -1 all pdsh@oleg401-client: oleg401-server: ssh exited with exit code 1 1 339 0.00 MB/sec execute 7 sec latency 22967.615 ms Started lustre-MDT0001 16:28:03 (1710102483) targets are mounted 16:28:03 (1710102483) facet_failover done 1 339 0.00 MB/sec execute 8 sec latency 23967.812 ms 1 339 0.00 MB/sec execute 9 sec latency 24968.052 ms 1 339 0.00 MB/sec execute 10 sec latency 25968.386 ms oleg401-client.virtnet: executing wait_import_state_mount (FULL|IDLE) mdc.lustre-MDT0001-mdc-*.mds_server_uuid 1 381 0.12 MB/sec execute 11 sec latency 26540.968 ms mdc.lustre-MDT0001-mdc-*.mds_server_uuid in FULL state after 0 sec 1 454 0.30 MB/sec execute 12 sec latency 86.191 ms 1 537 0.50 MB/sec execute 13 sec latency 94.372 ms 1 664 0.63 MB/sec execute 14 sec latency 84.466 ms 1 744 0.60 MB/sec execute 15 sec latency 61.825 ms 1 825 0.58 MB/sec execute 16 sec latency 51.739 ms UUID 1K-blocks Used Available Use% Mounted on lustre-MDT0000_UUID 1414116 3080 1284608 1% /mnt/lustre[MDT:0] lustre-MDT0001_UUID 1414116 2448 1285240 1% /mnt/lustre[MDT:1] lustre-OST0000_UUID 3833116 36088 3554896 2% /mnt/lustre[OST:0] lustre-OST0001_UUID 3833116 3780 3594680 1% /mnt/lustre[OST:1] filesystem_summary: 7666232 39868 7149576 1% /mnt/lustre 1 910 0.56 MB/sec execute 17 sec latency 57.338 ms 1 955 0.53 MB/sec execute 18 sec latency 99.803 ms 1 1045 0.56 MB/sec execute 19 sec latency 78.311 ms 1 1128 0.54 MB/sec execute 20 sec latency 56.119 ms 1 1190 0.51 MB/sec execute 21 sec latency 96.050 ms 1 1273 0.49 MB/sec execute 22 sec latency 105.270 ms test_26 fail mds1 5 times 1 1455 0.60 MB/sec execute 23 sec latency 82.711 ms Failing mds1 on oleg401-server 1 1513 0.57 MB/sec execute 24 sec latency 87.550 ms Stopping /mnt/lustre-mds1 (opts:) on oleg401-server 1 1610 0.55 MB/sec execute 25 sec latency 66.796 ms 1 1610 0.53 MB/sec execute 26 sec latency 1067.141 ms 16:28:22 (1710102502) shut down 1 1610 0.51 MB/sec execute 27 sec latency 2067.367 ms 1 1610 0.50 MB/sec execute 28 sec latency 3067.573 ms 1 1610 0.48 MB/sec execute 29 sec latency 4067.825 ms 1 1610 0.46 MB/sec execute 30 sec latency 5074.267 ms 1 1610 0.45 MB/sec execute 31 sec latency 6075.046 ms 1 1610 0.43 MB/sec execute 32 sec latency 7075.794 ms 1 1610 0.42 MB/sec execute 33 sec latency 8077.375 ms 1 1610 0.41 MB/sec execute 34 sec latency 9077.534 ms 1 1610 0.40 MB/sec execute 35 sec latency 10077.798 ms 1 1610 0.39 MB/sec execute 36 sec latency 11078.019 ms Failover mds1 to oleg401-server mount facets: mds1 1 1610 0.38 MB/sec execute 37 sec latency 12078.399 ms 1 1610 0.37 MB/sec execute 38 sec latency 13078.650 ms 1 1610 0.36 MB/sec execute 39 sec latency 14079.850 ms 1 1610 0.35 MB/sec execute 40 sec latency 15081.348 ms Starting mds1: -o localrecov /dev/mapper/mds1_flakey /mnt/lustre-mds1 1 1610 0.34 MB/sec execute 41 sec latency 16081.555 ms 1 1610 0.33 MB/sec execute 42 sec latency 17081.694 ms 1 1610 0.32 MB/sec execute 43 sec latency 18083.743 ms oleg401-server: oleg401-server.virtnet: executing set_default_debug -1 all 1 1610 0.32 MB/sec execute 44 sec latency 19085.880 ms pdsh@oleg401-client: oleg401-server: ssh exited with exit code 1 Started lustre-MDT0000 16:28:40 (1710102520) targets are mounted 16:28:40 (1710102520) facet_failover done 1 1610 0.31 MB/sec execute 45 sec latency 20086.593 ms 1 1610 0.30 MB/sec execute 46 sec latency 21086.728 ms 1 1616 0.30 MB/sec execute 47 sec latency 21975.770 ms oleg401-client.virtnet: executing wait_import_state_mount (FULL|IDLE) mdc.lustre-MDT0000-mdc-*.mds_server_uuid 1 1686 0.29 MB/sec execute 48 sec latency 116.189 ms mdc.lustre-MDT0000-mdc-*.mds_server_uuid in FULL state after 0 sec 1 1771 0.29 MB/sec execute 49 sec latency 140.828 ms 1 1861 0.28 MB/sec execute 50 sec latency 68.495 ms 1 1980 0.28 MB/sec execute 51 sec latency 81.853 ms 1 2088 0.28 MB/sec execute 52 sec latency 50.793 ms 1 2173 0.27 MB/sec execute 53 sec latency 196.399 ms UUID 1K-blocks Used Available Use% Mounted on lustre-MDT0000_UUID 1414116 2940 1284748 1% /mnt/lustre[MDT:0] lustre-MDT0001_UUID 1414116 2404 1285284 1% /mnt/lustre[MDT:1] lustre-OST0000_UUID 3833116 48256 3558536 2% /mnt/lustre[OST:0] lustre-OST0001_UUID 3833116 10844 3595748 1% /mnt/lustre[OST:1] filesystem_summary: 7666232 59100 7154284 1% /mnt/lustre 1 2300 0.29 MB/sec execute 54 sec latency 53.804 ms 1 2376 0.29 MB/sec execute 55 sec latency 80.357 ms 1 2429 0.28 MB/sec execute 56 sec latency 81.376 ms 1 2490 0.29 MB/sec execute 57 sec latency 107.786 ms 1 2604 0.29 MB/sec execute 58 sec latency 63.978 ms 1 2896 0.35 MB/sec execute 59 sec latency 51.415 ms test_26 fail mds2 6 times 1 2983 0.35 MB/sec execute 60 sec latency 118.637 ms Failing mds2 on oleg401-server 1 3025 0.35 MB/sec execute 61 sec latency 140.522 ms Stopping /mnt/lustre-mds2 (opts:) on oleg401-server 1 3291 0.37 MB/sec execute 62 sec latency 66.495 ms 1 3491 0.41 MB/sec execute 63 sec latency 50.744 ms 16:28:59 (1710102539) shut down 1 3590 0.41 MB/sec execute 64 sec latency 104.979 ms 1 3696 0.41 MB/sec execute 65 sec latency 60.292 ms 1 3793 0.42 MB/sec execute 66 sec latency 57.573 ms 1 3854 0.41 MB/sec execute 67 sec latency 71.285 ms 1 3927 0.41 MB/sec execute 68 sec latency 61.319 ms 1 3991 0.40 MB/sec execute 69 sec latency 71.395 ms 1 4059 0.40 MB/sec execute 70 sec latency 54.113 ms 1 4083 0.39 MB/sec execute 71 sec latency 708.642 ms 1 4083 0.39 MB/sec execute 72 sec latency 1710.346 ms 1 4083 0.38 MB/sec execute 73 sec latency 2711.097 ms Failover mds2 to oleg401-server mount facets: mds2 1 4083 0.38 MB/sec execute 74 sec latency 3711.776 ms 1 4083 0.37 MB/sec execute 75 sec latency 4711.930 ms 1 4083 0.37 MB/sec execute 76 sec latency 5712.438 ms 1 4083 0.36 MB/sec execute 77 sec latency 6712.893 ms 1 4083 0.36 MB/sec execute 78 sec latency 7713.778 ms Starting mds2: -o localrecov /dev/mapper/mds2_flakey /mnt/lustre-mds2 1 4083 0.35 MB/sec execute 79 sec latency 8713.958 ms 1 4083 0.35 MB/sec execute 80 sec latency 9714.237 ms oleg401-server: oleg401-server.virtnet: executing set_default_debug -1 all 1 4083 0.34 MB/sec execute 81 sec latency 10715.241 ms pdsh@oleg401-client: oleg401-server: ssh exited with exit code 1 Started lustre-MDT0001 16:29:17 (1710102557) targets are mounted 16:29:17 (1710102557) facet_failover done 1 4083 0.34 MB/sec execute 82 sec latency 11715.459 ms 1 4083 0.33 MB/sec execute 83 sec latency 12716.189 ms 1 4083 0.33 MB/sec execute 84 sec latency 13716.336 ms oleg401-client.virtnet: executing wait_import_state_mount (FULL|IDLE) mdc.lustre-MDT0001-mdc-*.mds_server_uuid 1 4141 0.33 MB/sec execute 85 sec latency 14121.911 ms mdc.lustre-MDT0001-mdc-*.mds_server_uuid in FULL state after 0 sec 1 4234 0.33 MB/sec execute 86 sec latency 153.108 ms 1 4297 0.32 MB/sec execute 87 sec latency 96.208 ms 1 4377 0.32 MB/sec execute 88 sec latency 80.739 ms 1 4474 0.32 MB/sec execute 89 sec latency 102.384 ms 1 4559 0.32 MB/sec execute 90 sec latency 40.637 ms 1 4699 0.33 MB/sec execute 91 sec latency 38.122 ms 1 4795 0.32 MB/sec execute 92 sec latency 46.990 ms 1 4969 0.35 MB/sec execute 93 sec latency 55.403 ms 1 5066 0.35 MB/sec execute 94 sec latency 49.228 ms 1 5156 0.35 MB/sec execute 95 sec latency 40.385 ms 1 5261 0.34 MB/sec execute 96 sec latency 48.983 ms 1 5358 0.34 MB/sec execute 97 sec latency 44.659 ms 1 5474 0.34 MB/sec execute 98 sec latency 34.009 ms 1 5595 0.34 MB/sec execute 99 sec latency 45.419 ms 1 cleanup 100 sec 1 cleanup 101 sec 0 cleanup 101 sec Operation Count AvgLat MaxLat ---------------------------------------- NTCreateX 937 57.950 26540.953 Close 725 6.922 71.365 Rename 37 42.128 129.364 Unlink 152 12.535 40.096 Qpathinfo 839 24.619 14121.883 Qfileinfo 180 0.774 6.778 Qfsinfo 135 2.065 14.359 Sfileinfo 84 23.635 55.408 Find 303 78.144 21975.760 WriteX 618 6.062 118.619 ReadX 1303 0.318 38.471 LockX 2 6.851 9.404 UnlockX 2 5.160 5.762 Flush 73 30.620 106.879 Throughput 0.337323 MB/sec 1 clients 1 procs max_latency=26540.968 ms stopping dbench on /mnt/lustre at Sun Mar 10 16:29:37 EDT 2024 with return code 0 clean dbench files on /mnt/lustre /mnt/lustre /mnt/lustre removed 'client.txt' /mnt/lustre dbench successfully finished striped dir -i0 -c2 -H all_char /mnt/lustre2/d26.replay-dual/run_dbench looking for dbench program /usr/bin/dbench found dbench client file /usr/share/dbench/client.txt '/usr/share/dbench/client.txt' -> 'client.txt' running 'dbench 1 -D /mnt/lustre2/d26.replay-dual/run_dbench -t 100' on /mnt/lustre at Sun Mar 10 16:29:40 EDT 2024 waiting for dbench pid 13592 dbench version 4.00 - Copyright Andrew Tridgell 1999-2004 Running for 100 seconds with load 'client.txt' and minimum warmup 20 secs 0 of 1 processes prepared for launch 0 sec 1 of 1 processes prepared for launch 0 sec releasing clients 1 103 4.78 MB/sec warmup 1 sec latency 40.732 ms 1 189 3.79 MB/sec warmup 2 sec latency 70.082 ms 1 287 3.57 MB/sec warmup 3 sec latency 49.451 ms 1 374 3.32 MB/sec warmup 4 sec latency 51.964 ms 1 484 3.39 MB/sec warmup 5 sec latency 43.452 ms 1 612 3.49 MB/sec warmup 6 sec latency 90.378 ms 1 711 3.03 MB/sec warmup 7 sec latency 55.779 ms 1 794 2.68 MB/sec warmup 8 sec latency 64.474 ms 1 888 2.41 MB/sec warmup 9 sec latency 66.891 ms 1 959 2.18 MB/sec warmup 10 sec latency 58.374 ms 1 1050 2.08 MB/sec warmup 11 sec latency 89.949 ms 1 1147 1.91 MB/sec warmup 12 sec latency 99.254 ms 1 1235 1.77 MB/sec warmup 13 sec latency 99.339 ms 1 1410 1.86 MB/sec warmup 14 sec latency 47.761 ms 1 1504 1.74 MB/sec warmup 15 sec latency 43.743 ms 1 1614 1.63 MB/sec warmup 16 sec latency 49.908 ms tar: Unexpected EOF in archive tar: Unexpected EOF in archive 1 1692 1.54 MB/sec warmup 17 sec latency 74.393 ms 1 1825 1.46 MB/sec warmup 18 sec latency 40.737 ms 1 1957 1.39 MB/sec warmup 19 sec latency 50.493 ms 1 2162 0.09 MB/sec execute 1 sec latency 54.963 ms tar: Error is not recoverable: exiting now dbench killed by signal 15 stopping dbench on /mnt/lustre at Sun Mar 10 16:30:02 EDT 2024 with return code 0 1 2282 0.31 MB/sec execute 2 sec latency 48.354 ms 13592 pts/0 S+ 0:00 dbench -c client.txt 1 -D /mnt/lustre2/d26.replay-dual/run_dbench -t 100 killed dbench main pid 13592 clean dbench files on /mnt/lustre /mnt/lustre /mnt/lustre removed 'client.txt' /mnt/lustre dbench successfully finished