== replay-dual test 26: dbench and tar with mds failover ========================================================== 16:17:09 (1789676229) Starting client oleg455-client.virtnet: -o user_xattr,flock 192.168.204.155@tcp:/lustre /mnt/lustre Started clients oleg455-client.virtnet: 192.168.204.155@tcp:/lustre on /mnt/lustre type lustre (rw,flock,checksum,encrypt,lazystatfs,lruresize,nolock,statfs_project,nosync_on_close,nouser_fid2path,user_xattr,verbose) Started tar loop with pid 65652 Started dbench loop with 65653 striped dir -i0 -c2 -H all_char /mnt/lustre/d26.replay-dual/run_tar striped dir -i0 -c2 -H crush /mnt/lustre2/d26.replay-dual/run_dbench looking for dbench program /usr/bin/dbench found dbench client file /usr/share/dbench/client.txt UUID 1K-blocks Used Available Use% Mounted on lustre-MDT0000_UUID 1414116 1968 1269336 1% /mnt/lustre[MDT:0] lustre-MDT0001_UUID 1414116 1768 1269536 1% /mnt/lustre[MDT:1] lustre-OST0000_UUID 3833116 1564 3605456 1% /mnt/lustre[OST:0] lustre-OST0001_UUID 3833116 17928 3589036 1% /mnt/lustre[OST:1] filesystem_summary: 7666232 19492 7194492 1% /mnt/lustre '/usr/share/dbench/client.txt' -> 'client.txt' running 'dbench 1 -D /mnt/lustre2/d26.replay-dual/run_dbench -t 100' on /mnt/lustre at Thu Sep 17 16:17:15 EDT 2026 waiting for dbench pid 65731 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 1 of 1 processes prepared for launch 0 sec releasing clients 1 18 0.26 MB/sec warmup 1 sec latency 19.607 ms 1 72 1.57 MB/sec warmup 2 sec latency 148.494 ms 1 113 1.74 MB/sec warmup 3 sec latency 187.312 ms 1 147 1.57 MB/sec warmup 4 sec latency 172.784 ms 1 173 1.41 MB/sec warmup 5 sec latency 172.733 ms 1 187 1.25 MB/sec warmup 6 sec latency 534.029 ms 1 223 1.23 MB/sec warmup 7 sec latency 143.595 ms 1 255 1.21 MB/sec warmup 8 sec latency 154.025 ms 1 297 1.22 MB/sec warmup 9 sec latency 161.911 ms test_26 fail mds1 1 times 1 323 1.17 MB/sec warmup 10 sec latency 174.433 ms 1 357 1.16 MB/sec warmup 11 sec latency 173.620 ms Failing mds1 on oleg455-server 1 402 1.18 MB/sec warmup 12 sec latency 163.867 ms Stopping /mnt/lustre-mds1 (opts:) on oleg455-server 1 437 1.18 MB/sec warmup 13 sec latency 174.751 ms 1 461 1.15 MB/sec warmup 14 sec latency 334.067 ms 1 461 1.08 MB/sec warmup 15 sec latency 1334.493 ms 1 461 1.01 MB/sec warmup 16 sec latency 2334.877 ms 16:17:31 (1789676251) shut down facet: mds1 facet_host: oleg455-server facet_failover_host: oleg455-server 1 461 0.95 MB/sec warmup 17 sec latency 3336.553 ms 1 461 0.90 MB/sec warmup 18 sec latency 4337.785 ms 1 461 0.85 MB/sec warmup 19 sec latency 5337.980 ms 1 461 0.00 MB/sec execute 1 sec latency 7339.704 ms 1 461 0.00 MB/sec execute 2 sec latency 8339.976 ms 1 461 0.00 MB/sec execute 3 sec latency 9340.192 ms 1 461 0.00 MB/sec execute 4 sec latency 10342.822 ms 1 461 0.00 MB/sec execute 5 sec latency 11343.805 ms 1 461 0.00 MB/sec execute 6 sec latency 12344.558 ms Failover mds1 to oleg455-server mount facets: mds1 1 461 0.00 MB/sec execute 7 sec latency 13345.670 ms 1 461 0.00 MB/sec execute 8 sec latency 14346.415 ms 1 461 0.00 MB/sec execute 9 sec latency 15346.668 ms 1 461 0.00 MB/sec execute 10 sec latency 16346.890 ms 1 461 0.00 MB/sec execute 11 sec latency 17352.696 ms 1 461 0.00 MB/sec execute 12 sec latency 18352.978 ms 1 461 0.00 MB/sec execute 13 sec latency 19353.255 ms 1 461 0.00 MB/sec execute 14 sec latency 20354.893 ms 1 461 0.00 MB/sec execute 15 sec latency 21356.733 ms 1 461 0.00 MB/sec execute 16 sec latency 22357.000 ms Start mds1: mount -t lustre -o localrecov /dev/mapper/mds1_flakey /mnt/lustre-mds1 1 461 0.00 MB/sec execute 17 sec latency 23360.163 ms 1 461 0.00 MB/sec execute 18 sec latency 24360.479 ms 1 461 0.00 MB/sec execute 19 sec latency 25362.337 ms 1 461 0.00 MB/sec execute 20 sec latency 26362.596 ms 1 461 0.00 MB/sec execute 21 sec latency 27363.420 ms 1 461 0.00 MB/sec execute 22 sec latency 28363.795 ms 1 461 0.00 MB/sec execute 23 sec latency 29364.050 ms 1 461 0.00 MB/sec execute 24 sec latency 30364.252 ms 1 461 0.00 MB/sec execute 25 sec latency 31364.484 ms oleg455-server: oleg455-server.virtnet: executing set_default_debug -1 all 1 461 0.00 MB/sec execute 26 sec latency 32364.928 ms pdsh@oleg455-client: oleg455-server: ssh exited with exit code 1 1 461 0.00 MB/sec execute 27 sec latency 33365.322 ms Started lustre-MDT0000 16:18:03 (1789676283) targets are mounted 16:18:03 (1789676283) facet_failover done 1 461 0.00 MB/sec execute 28 sec latency 34368.234 ms 1 466 0.00 MB/sec execute 29 sec latency 35319.380 ms 1 503 0.04 MB/sec execute 30 sec latency 152.673 ms 1 542 0.09 MB/sec execute 31 sec latency 130.698 ms 1 585 0.12 MB/sec execute 32 sec latency 608.965 ms 1 642 0.15 MB/sec execute 33 sec latency 125.004 ms 1 674 0.15 MB/sec execute 34 sec latency 143.057 ms oleg455-client.virtnet: executing wait_import_state_mount (FULL|IDLE) mdc.lustre-MDT0000-mdc-*.mds_server_uuid 1475 0 1 705 0.14 MB/sec execute 35 sec latency 124.631 ms 1 732 0.14 MB/sec execute 36 sec latency 129.648 ms mdc.lustre-MDT0000-mdc-*.mds_server_uuid in FULL state after 0 sec 1 765 0.14 MB/sec execute 37 sec latency 142.142 ms 1 798 0.14 MB/sec execute 38 sec latency 124.107 ms 1 845 0.14 MB/sec execute 39 sec latency 113.189 ms 1 885 0.14 MB/sec execute 40 sec latency 110.892 ms 1 918 0.14 MB/sec execute 41 sec latency 121.298 ms 1 952 0.13 MB/sec execute 42 sec latency 121.900 ms 1 984 0.13 MB/sec execute 43 sec latency 123.335 ms UUID 1K-blocks Used Available Use% Mounted on lustre-MDT0000_UUID 1414116 2144 1269160 1% /mnt/lustre[MDT:0] lustre-MDT0001_UUID 1414116 1936 1269368 1% /mnt/lustre[MDT:1] lustre-OST0000_UUID 3833116 21088 3583744 1% /mnt/lustre[OST:0] lustre-OST0001_UUID 3833116 33192 3570580 1% /mnt/lustre[OST:1] filesystem_summary: 7666232 54280 7154324 1% /mnt/lustre 1 1038 0.15 MB/sec execute 44 sec latency 108.580 ms 1 1076 0.15 MB/sec execute 45 sec latency 135.109 ms 1 1127 0.15 MB/sec execute 46 sec latency 83.129 ms 1 1157 0.15 MB/sec execute 47 sec latency 142.759 ms 1 1172 0.14 MB/sec execute 48 sec latency 678.334 ms 1 1179 0.14 MB/sec execute 49 sec latency 1279.550 ms 1 1215 0.14 MB/sec execute 50 sec latency 119.754 ms 1 1255 0.14 MB/sec execute 51 sec latency 103.014 ms 1 1380 0.19 MB/sec execute 52 sec latency 124.247 ms 1 1410 0.19 MB/sec execute 53 sec latency 108.127 ms 1 1473 0.18 MB/sec execute 54 sec latency 89.714 ms test_26 fail mds2 2 times 1 1509 0.18 MB/sec execute 55 sec latency 98.768 ms 1 1552 0.18 MB/sec execute 56 sec latency 129.674 ms Failing mds2 on oleg455-server 1 1578 0.17 MB/sec execute 57 sec latency 88.258 ms Stopping /mnt/lustre-mds2 (opts:) on oleg455-server 1 1619 0.17 MB/sec execute 58 sec latency 156.534 ms 1 1653 0.17 MB/sec execute 59 sec latency 150.301 ms 1 1711 0.17 MB/sec execute 60 sec latency 183.618 ms 1 1821 0.17 MB/sec execute 61 sec latency 46.667 ms 1 1884 0.16 MB/sec execute 62 sec latency 84.761 ms 1 1978 0.16 MB/sec execute 63 sec latency 51.725 ms 1 2068 0.16 MB/sec execute 64 sec latency 72.330 ms 1 2145 0.16 MB/sec execute 65 sec latency 166.778 ms 1 2181 0.16 MB/sec execute 66 sec latency 95.791 ms 1 2244 0.16 MB/sec execute 67 sec latency 93.884 ms 16:18:43 (1789676323) shut down facet: mds2 facet_host: oleg455-server facet_failover_host: oleg455-server 1 2259 0.16 MB/sec execute 68 sec latency 743.421 ms 1 2259 0.15 MB/sec execute 69 sec latency 1744.234 ms 1 2259 0.15 MB/sec execute 70 sec latency 2744.498 ms 1 2259 0.15 MB/sec execute 71 sec latency 3744.901 ms 1 2259 0.15 MB/sec execute 72 sec latency 4745.289 ms 1 2259 0.15 MB/sec execute 73 sec latency 5745.477 ms 1 2259 0.14 MB/sec execute 74 sec latency 6746.209 ms 1 2259 0.14 MB/sec execute 75 sec latency 7746.403 ms 1 2259 0.14 MB/sec execute 76 sec latency 8746.809 ms 1 2259 0.14 MB/sec execute 77 sec latency 9747.041 ms Failover mds2 to oleg455-server 1 2259 0.14 MB/sec execute 78 sec latency 10752.727 ms mount facets: mds2 1 2259 0.13 MB/sec execute 79 sec latency 11752.906 ms 1 2259 0.13 MB/sec execute 80 sec latency 12753.622 ms 1 2259 0.13 MB/sec execute 81 sec latency 13756.867 ms 1 2259 0.13 MB/sec execute 82 sec latency 14757.799 ms 1 2259 0.13 MB/sec execute 83 sec latency 15758.037 ms 1 2259 0.13 MB/sec execute 84 sec latency 16758.320 ms 1 2259 0.12 MB/sec execute 85 sec latency 17758.635 ms 1 2259 0.12 MB/sec execute 86 sec latency 18761.728 ms Start mds2: mount -t lustre -o localrecov /dev/mapper/mds2_flakey /mnt/lustre-mds2 1 2259 0.12 MB/sec execute 87 sec latency 19765.608 ms 1 2259 0.12 MB/sec execute 88 sec latency 20766.912 ms 1 2259 0.12 MB/sec execute 89 sec latency 21767.126 ms 1 2259 0.12 MB/sec execute 90 sec latency 22768.612 ms 1 2259 0.12 MB/sec execute 91 sec latency 23768.987 ms 1 2259 0.12 MB/sec execute 92 sec latency 24769.345 ms 1 2259 0.11 MB/sec execute 93 sec latency 25770.198 ms 1 2259 0.11 MB/sec execute 94 sec latency 26772.464 ms oleg455-server: oleg455-server.virtnet: executing set_default_debug -1 all 1 2259 0.11 MB/sec execute 95 sec latency 27773.565 ms 1 2259 0.11 MB/sec execute 96 sec latency 28773.770 ms pdsh@oleg455-client: oleg455-server: ssh exited with exit code 1 1 2295 0.12 MB/sec execute 97 sec latency 29066.571 ms Started lustre-MDT0001 16:19:13 (1789676353) targets are mounted 16:19:13 (1789676353) facet_failover done 1 2350 0.12 MB/sec execute 98 sec latency 124.210 ms 1 2388 0.12 MB/sec execute 99 sec latency 93.813 ms 1 cleanup 100 sec 1 cleanup 101 sec 1 cleanup 102 sec 1 cleanup 103 sec 1 cleanup 104 sec 0 cleanup 104 sec Operation Count AvgLat MaxLat ---------------------------------------- NTCreateX 341 242.597 35319.365 Close 294 11.419 43.080 Rename 13 94.705 142.735 Unlink 33 27.309 52.165 Qpathinfo 317 15.198 135.088 Qfileinfo 81 2.067 19.371 Qfsinfo 39 4.571 22.687 Sfileinfo 43 63.335 118.079 Find 107 13.817 71.676 WriteX 275 13.531 93.867 ReadX 381 3.199 118.558 Flush 33 63.219 143.043 Throughput 0.119007 MB/sec 1 clients 1 procs max_latency=35319.380 ms stopping dbench on /mnt/lustre at Thu Sep 17 16:19:20 EDT 2026 with return code 0 clean dbench files on /mnt/lustre /mnt/lustre /mnt/lustre oleg455-client.virtnet: executing wait_import_state_mount (FULL|IDLE) mdc.lustre-MDT0001-mdc-*.mds_server_uuid 1475 0 removed 'client.txt' /mnt/lustre dbench successfully finished mdc.lustre-MDT0001-mdc-*.mds_server_uuid in FULL state after 0 sec 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 Thu Sep 17 16:19:30 EDT 2026 waiting for dbench pid 67855 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 63 2.66 MB/sec warmup 1 sec latency 96.873 ms UUID 1K-blocks Used Available Use% Mounted on lustre-MDT0000_UUID 1414116 2356 1268948 1% /mnt/lustre[MDT:0] lustre-MDT0001_UUID 1414116 2244 1269060 1% /mnt/lustre[MDT:1] lustre-OST0000_UUID 3833116 36552 3567944 2% /mnt/lustre[OST:0] lustre-OST0001_UUID 3833116 10080 3595396 1% /mnt/lustre[OST:1] filesystem_summary: 7666232 46632 7163340 1% /mnt/lustre 1 112 2.61 MB/sec warmup 2 sec latency 204.136 ms 1 139 2.00 MB/sec warmup 3 sec latency 207.723 ms 1 178 1.79 MB/sec warmup 4 sec latency 161.860 ms 1 206 1.62 MB/sec warmup 5 sec latency 347.468 ms 1 242 1.52 MB/sec warmup 6 sec latency 146.288 ms 1 280 1.49 MB/sec warmup 7 sec latency 146.790 ms 1 286 1.33 MB/sec warmup 8 sec latency 927.793 ms 1 305 1.24 MB/sec warmup 9 sec latency 1188.578 ms 1 331 1.20 MB/sec warmup 10 sec latency 179.274 ms 1 365 1.18 MB/sec warmup 11 sec latency 195.501 ms test_26 fail mds1 3 times 1 397 1.17 MB/sec warmup 12 sec latency 151.594 ms 1 432 1.16 MB/sec warmup 13 sec latency 150.854 ms Failing mds1 on oleg455-server 1 468 1.17 MB/sec warmup 14 sec latency 112.867 ms Stopping /mnt/lustre-mds1 (opts:) on oleg455-server 1 504 1.16 MB/sec warmup 15 sec latency 140.890 ms 1 569 1.19 MB/sec warmup 16 sec latency 113.044 ms 1 600 1.23 MB/sec warmup 17 sec latency 601.221 ms 1 600 1.16 MB/sec warmup 18 sec latency 1602.313 ms 16:19:48 (1789676388) shut down facet: mds1 facet_host: oleg455-server facet_failover_host: oleg455-server 1 600 1.10 MB/sec warmup 19 sec latency 2602.526 ms 1 600 0.00 MB/sec execute 1 sec latency 4603.246 ms 1 600 0.00 MB/sec execute 2 sec latency 5603.447 ms 1 600 0.00 MB/sec execute 3 sec latency 6607.860 ms 1 600 0.00 MB/sec execute 4 sec latency 7609.750 ms 1 600 0.00 MB/sec execute 5 sec latency 8609.939 ms 1 600 0.00 MB/sec execute 6 sec latency 9610.140 ms 1 600 0.00 MB/sec execute 7 sec latency 10610.334 ms 1 600 0.00 MB/sec execute 8 sec latency 11612.040 ms Failover mds1 to oleg455-server 1 600 0.00 MB/sec execute 9 sec latency 12612.285 ms mount facets: mds1 1 600 0.00 MB/sec execute 10 sec latency 13612.488 ms 1 600 0.00 MB/sec execute 11 sec latency 14613.442 ms 1 600 0.00 MB/sec execute 12 sec latency 15613.704 ms 1 600 0.00 MB/sec execute 13 sec latency 16615.145 ms 1 600 0.00 MB/sec execute 14 sec latency 17615.318 ms 1 600 0.00 MB/sec execute 15 sec latency 18615.592 ms 1 600 0.00 MB/sec execute 16 sec latency 19615.996 ms 1 600 0.00 MB/sec execute 17 sec latency 20616.519 ms 1 600 0.00 MB/sec execute 18 sec latency 21616.724 ms 1 600 0.00 MB/sec execute 19 sec latency 22617.071 ms Start mds1: mount -t lustre -o localrecov /dev/mapper/mds1_flakey /mnt/lustre-mds1 1 600 0.00 MB/sec execute 20 sec latency 23618.183 ms 1 600 0.00 MB/sec execute 21 sec latency 24618.825 ms 1 600 0.00 MB/sec execute 22 sec latency 25619.005 ms 1 600 0.00 MB/sec execute 23 sec latency 26619.399 ms 1 600 0.00 MB/sec execute 24 sec latency 27620.901 ms 1 600 0.00 MB/sec execute 25 sec latency 28621.099 ms 1 600 0.00 MB/sec execute 26 sec latency 29626.074 ms 1 600 0.00 MB/sec execute 27 sec latency 30627.874 ms 1 600 0.00 MB/sec execute 28 sec latency 31628.120 ms 1 600 0.00 MB/sec execute 29 sec latency 32628.718 ms 1 600 0.00 MB/sec execute 30 sec latency 33629.192 ms oleg455-server: oleg455-server.virtnet: executing set_default_debug -1 all 1 600 0.00 MB/sec execute 31 sec latency 34629.414 ms 1 600 0.00 MB/sec execute 32 sec latency 35629.654 ms pdsh@oleg455-client: oleg455-server: ssh exited with exit code 1 1 600 0.00 MB/sec execute 33 sec latency 36630.211 ms 1 639 0.00 MB/sec execute 34 sec latency 36783.063 ms Started lustre-MDT0000 16:20:24 (1789676424) targets are mounted 16:20:24 (1789676424) facet_failover done 1 673 0.01 MB/sec execute 35 sec latency 140.281 ms 1 708 0.01 MB/sec execute 36 sec latency 130.526 ms 1 735 0.01 MB/sec execute 37 sec latency 141.689 ms 1 776 0.01 MB/sec execute 38 sec latency 131.565 ms 1 803 0.01 MB/sec execute 39 sec latency 141.373 ms 1 849 0.02 MB/sec execute 40 sec latency 153.374 ms 1 882 0.02 MB/sec execute 41 sec latency 141.290 ms oleg455-client.virtnet: executing wait_import_state_mount (FULL|IDLE) mdc.lustre-MDT0000-mdc-*.mds_server_uuid 1475 0 1 884 0.02 MB/sec execute 42 sec latency 933.219 ms 1 895 0.02 MB/sec execute 43 sec latency 1382.862 ms mdc.lustre-MDT0000-mdc-*.mds_server_uuid in FULL state after 0 sec 1 919 0.02 MB/sec execute 44 sec latency 115.879 ms 1 955 0.02 MB/sec execute 45 sec latency 126.730 ms 1 981 0.02 MB/sec execute 46 sec latency 215.050 ms 1 1005 0.02 MB/sec execute 47 sec latency 141.061 ms 1 1063 0.04 MB/sec execute 48 sec latency 207.417 ms tar: Unexpected EOF in archive tar: Unexpected EOF in archive 1 1127 0.04 MB/sec execute 49 sec latency 103.402 ms 1 1172 0.04 MB/sec execute 50 sec latency 93.771 ms 1 1223 0.04 MB/sec execute 51 sec latency 77.051 ms 1 1251 0.04 MB/sec execute 52 sec latency 119.805 ms 1 1391 0.09 MB/sec execute 53 sec latency 80.871 ms 1 1429 0.09 MB/sec execute 54 sec latency 90.654 ms 1 1467 0.09 MB/sec execute 55 sec latency 255.201 ms 1 1504 0.09 MB/sec execute 56 sec latency 368.963 ms 1 1557 0.09 MB/sec execute 57 sec latency 220.655 ms 1 1609 0.09 MB/sec execute 58 sec latency 108.090 ms 1 1656 0.09 MB/sec execute 59 sec latency 139.023 ms tar: Error is not recoverable: exiting now dbench killed by signal 15 stopping dbench on /mnt/lustre at Thu Sep 17 16:20:49 EDT 2026 with return code 0 67855 pts/0 S+ 0:00 dbench -c client.txt 1 -D /mnt/lustre2/d26.replay-dual/run_dbench -t 100 killed dbench main pid 67855 clean dbench files on /mnt/lustre /mnt/lustre /mnt/lustre removed 'client.txt' /mnt/lustre dbench successfully finished