== replay-dual test 26: dbench and tar with mds failover ========================================================== 11:45:54 (1746114354) Starting client oleg408-client.virtnet: -o user_xattr,flock 192.168.204.108@tcp:/lustre /mnt/lustre Started clients oleg408-client.virtnet: 192.168.204.108@tcp:/lustre on /mnt/lustre type lustre (rw,checksum,noencrypt,flock,lazystatfs,lruresize,nolock,statfs_project,nouser_fid2path,user_xattr,verbose) Started tar loop with pid 4167 Started dbench loop with 4168 striped dir -i0 -c2 -H crush /mnt/lustre2/d26.replay-dual/run_dbench striped dir -i0 -c2 -H crush /mnt/lustre/d26.replay-dual/run_tar 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 May 1 11:45:56 EDT 2025 waiting for dbench pid 4209 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 1 of 1 processes prepared for launch 0 sec releasing clients 1 194 7.71 MB/sec warmup 1 sec latency 32.176 ms UUID 1K-blocks Used Available Use% Mounted on lustre-MDT0000_UUID 1414116 2268 1285420 1% /mnt/lustre[MDT:0] lustre-MDT0001_UUID 1414116 2040 1285648 1% /mnt/lustre[MDT:1] lustre-OST0000_UUID 3833116 26120 3555956 1% /mnt/lustre[OST:0] lustre-OST0001_UUID 3833116 1536 3605484 1% /mnt/lustre[OST:1] filesystem_summary: 7666232 27656 7161440 1% /mnt/lustre 1 378 6.72 MB/sec warmup 2 sec latency 23.766 ms 1 472 5.47 MB/sec warmup 3 sec latency 445.493 ms 1 718 5.32 MB/sec warmup 4 sec latency 22.012 ms test_26 fail mds1 1 times 1 921 4.35 MB/sec warmup 5 sec latency 23.050 ms Failing mds1 on oleg408-server Stopping /mnt/lustre-mds1 (opts:) on oleg408-server 1 1147 3.83 MB/sec warmup 6 sec latency 104.004 ms 11:46:02 (1746114362) shut down facet: mds1 facet_host: oleg408-server facet_failover_host: oleg408-server 1 1147 3.28 MB/sec warmup 7 sec latency 1104.183 ms 1 1147 2.87 MB/sec warmup 8 sec latency 2104.384 ms 1 1147 2.55 MB/sec warmup 9 sec latency 3104.606 ms 1 1147 2.30 MB/sec warmup 10 sec latency 4104.792 ms 1 1147 2.09 MB/sec warmup 11 sec latency 5105.013 ms 1 1147 1.92 MB/sec warmup 12 sec latency 6105.229 ms 1 1147 1.77 MB/sec warmup 13 sec latency 7105.428 ms 1 1147 1.64 MB/sec warmup 14 sec latency 8105.647 ms 1 1147 1.53 MB/sec warmup 15 sec latency 9105.973 ms 1 1147 1.44 MB/sec warmup 16 sec latency 10106.209 ms Failover mds1 to oleg408-server mount facets: mds1 1 1147 1.35 MB/sec warmup 17 sec latency 11106.406 ms 1 1147 1.28 MB/sec warmup 18 sec latency 12106.598 ms Starting mds1: -o localrecov /dev/mapper/mds1_flakey /mnt/lustre-mds1 1 1147 1.21 MB/sec warmup 19 sec latency 13106.760 ms oleg408-server: oleg408-server.virtnet: executing set_default_debug -1 all pdsh@oleg408-client: oleg408-server: ssh exited with exit code 1 Started lustre-MDT0000 11:46:16 (1746114376) targets are mounted 11:46:16 (1746114376) facet_failover done 1 1147 0.00 MB/sec execute 1 sec latency 15107.127 ms 1 1147 0.00 MB/sec execute 2 sec latency 16107.312 ms 1 1147 0.00 MB/sec execute 3 sec latency 17107.476 ms 1 1147 0.00 MB/sec execute 4 sec latency 18107.682 ms 1 1147 0.00 MB/sec execute 5 sec latency 19107.811 ms oleg408-client.virtnet: executing wait_import_state_mount (FULL|IDLE) mdc.lustre-MDT0000-mdc-*.mds_server_uuid 1 1318 0.19 MB/sec execute 6 sec latency 19343.346 ms mdc.lustre-MDT0000-mdc-*.mds_server_uuid in FULL state after 0 sec 1 1563 0.44 MB/sec execute 7 sec latency 25.928 ms 1 1717 0.41 MB/sec execute 8 sec latency 26.392 ms 1 1967 0.38 MB/sec execute 9 sec latency 23.463 ms 1 2164 0.37 MB/sec execute 10 sec latency 23.145 ms UUID 1K-blocks Used Available Use% Mounted on lustre-MDT0000_UUID 1414116 2932 1284756 1% /mnt/lustre[MDT:0] lustre-MDT0001_UUID 1414116 2604 1285084 1% /mnt/lustre[MDT:1] lustre-OST0000_UUID 3833116 31864 3564820 1% /mnt/lustre[OST:0] lustre-OST0001_UUID 3833116 6500 3592444 1% /mnt/lustre[OST:1] filesystem_summary: 7666232 38364 7157264 1% /mnt/lustre 1 2418 0.46 MB/sec execute 11 sec latency 25.047 ms 1 2768 0.75 MB/sec execute 12 sec latency 101.421 ms 1 3024 0.81 MB/sec execute 13 sec latency 238.465 ms 1 3494 1.10 MB/sec execute 14 sec latency 27.048 ms test_26 fail mds2 2 times Failing mds2 on oleg408-server 1 3775 1.12 MB/sec execute 15 sec latency 19.422 ms Stopping /mnt/lustre-mds2 (opts:) on oleg408-server 1 3839 1.05 MB/sec execute 16 sec latency 543.308 ms 11:46:32 (1746114392) shut down facet: mds2 facet_host: oleg408-server facet_failover_host: oleg408-server 1 3839 0.99 MB/sec execute 17 sec latency 1543.451 ms 1 3839 0.93 MB/sec execute 18 sec latency 2543.731 ms 1 3839 0.88 MB/sec execute 19 sec latency 3543.904 ms 1 3839 0.84 MB/sec execute 20 sec latency 4544.059 ms 1 3839 0.80 MB/sec execute 21 sec latency 5544.209 ms 1 3839 0.76 MB/sec execute 22 sec latency 6544.351 ms 1 3839 0.73 MB/sec execute 23 sec latency 7544.488 ms 1 3839 0.70 MB/sec execute 24 sec latency 8544.790 ms 1 3839 0.67 MB/sec execute 25 sec latency 9544.956 ms 1 3839 0.64 MB/sec execute 26 sec latency 10545.138 ms Failover mds2 to oleg408-server mount facets: mds2 1 3839 0.62 MB/sec execute 27 sec latency 11545.286 ms 1 3839 0.60 MB/sec execute 28 sec latency 12545.427 ms Starting mds2: -o localrecov /dev/mapper/mds2_flakey /mnt/lustre-mds2 1 3839 0.58 MB/sec execute 29 sec latency 13545.582 ms oleg408-server: oleg408-server.virtnet: executing set_default_debug -1 all 1 3839 0.56 MB/sec execute 30 sec latency 14545.835 ms pdsh@oleg408-client: oleg408-server: ssh exited with exit code 1 Started lustre-MDT0001 11:46:46 (1746114406) targets are mounted 11:46:46 (1746114406) facet_failover done 1 3839 0.54 MB/sec execute 31 sec latency 15546.081 ms 1 3839 0.52 MB/sec execute 32 sec latency 16546.319 ms 1 3839 0.51 MB/sec execute 33 sec latency 17546.559 ms 1 3839 0.49 MB/sec execute 34 sec latency 18546.834 ms oleg408-client.virtnet: executing wait_import_state_mount (FULL|IDLE) mdc.lustre-MDT0001-mdc-*.mds_server_uuid 1 3961 0.49 MB/sec execute 35 sec latency 19011.468 ms mdc.lustre-MDT0001-mdc-*.mds_server_uuid in FULL state after 0 sec 1 4139 0.47 MB/sec execute 36 sec latency 27.515 ms 1 4346 0.47 MB/sec execute 37 sec latency 23.162 ms 1 4521 0.47 MB/sec execute 38 sec latency 27.355 ms 1 4729 0.49 MB/sec execute 39 sec latency 21.895 ms UUID 1K-blocks Used Available Use% Mounted on lustre-MDT0000_UUID 1414116 3972 1283716 1% /mnt/lustre[MDT:0] lustre-MDT0001_UUID 1414116 2440 1285248 1% /mnt/lustre[MDT:1] lustre-OST0000_UUID 3833116 40496 3560172 2% /mnt/lustre[OST:0] lustre-OST0001_UUID 3833116 11356 3583920 1% /mnt/lustre[OST:1] filesystem_summary: 7666232 51852 7144092 1% /mnt/lustre 1 5017 0.55 MB/sec execute 40 sec latency 25.295 ms 1 5227 0.54 MB/sec execute 41 sec latency 207.228 ms 1 5408 0.54 MB/sec execute 42 sec latency 21.592 ms 1 5645 0.53 MB/sec execute 43 sec latency 20.585 ms test_26 fail mds1 3 times Failing mds1 on oleg408-server Stopping /mnt/lustre-mds1 (opts:) on oleg408-server 1 5918 0.55 MB/sec execute 44 sec latency 17.953 ms 11:47:00 (1746114420) shut down facet: mds1 facet_host: oleg408-server facet_failover_host: oleg408-server 1 6145 0.56 MB/sec execute 45 sec latency 21.714 ms 1 6548 0.64 MB/sec execute 46 sec latency 26.370 ms 1 6927 0.67 MB/sec execute 47 sec latency 23.884 ms 1 7224 0.73 MB/sec execute 48 sec latency 20.356 ms 1 7396 0.73 MB/sec execute 49 sec latency 23.817 ms 1 7566 0.72 MB/sec execute 50 sec latency 23.410 ms 1 7627 0.71 MB/sec execute 51 sec latency 587.410 ms 1 7627 0.70 MB/sec execute 52 sec latency 1587.671 ms 1 7627 0.68 MB/sec execute 53 sec latency 2587.885 ms 1 7627 0.67 MB/sec execute 54 sec latency 3588.070 ms Failover mds1 to oleg408-server mount facets: mds1 1 7627 0.66 MB/sec execute 55 sec latency 4588.255 ms 1 7627 0.65 MB/sec execute 56 sec latency 5588.586 ms 1 7627 0.64 MB/sec execute 57 sec latency 6588.751 ms Starting mds1: -o localrecov /dev/mapper/mds1_flakey /mnt/lustre-mds1 1 7627 0.62 MB/sec execute 58 sec latency 7588.916 ms oleg408-server: oleg408-server.virtnet: executing set_default_debug -1 all pdsh@oleg408-client: oleg408-server: ssh exited with exit code 1 1 7627 0.61 MB/sec execute 59 sec latency 8589.094 ms Started lustre-MDT0000 11:47:15 (1746114435) targets are mounted 11:47:15 (1746114435) facet_failover done 1 7627 0.60 MB/sec execute 60 sec latency 9589.363 ms 1 7627 0.59 MB/sec execute 61 sec latency 10589.614 ms 1 7627 0.58 MB/sec execute 62 sec latency 11589.842 ms 1 7627 0.57 MB/sec execute 63 sec latency 12590.119 ms 1 7627 0.57 MB/sec execute 64 sec latency 13590.323 ms 1 7753 0.56 MB/sec execute 65 sec latency 14196.599 ms oleg408-client.virtnet: executing wait_import_state_mount (FULL|IDLE) mdc.lustre-MDT0000-mdc-*.mds_server_uuid mdc.lustre-MDT0000-mdc-*.mds_server_uuid in FULL state after 0 sec 1 7960 0.56 MB/sec execute 66 sec latency 29.233 ms 1 8202 0.57 MB/sec execute 67 sec latency 25.925 ms 1 8581 0.61 MB/sec execute 68 sec latency 16.020 ms 1 8814 0.60 MB/sec execute 69 sec latency 23.086 ms UUID 1K-blocks Used Available Use% Mounted on lustre-MDT0000_UUID 1414116 4380 1283308 1% /mnt/lustre[MDT:0] lustre-MDT0001_UUID 1414116 2380 1285308 1% /mnt/lustre[MDT:1] lustre-OST0000_UUID 3833116 38888 3553332 2% /mnt/lustre[OST:0] lustre-OST0001_UUID 3833116 11180 3575328 1% /mnt/lustre[OST:1] filesystem_summary: 7666232 50068 7128660 1% /mnt/lustre 1 9089 0.60 MB/sec execute 70 sec latency 19.606 ms 1 9315 0.59 MB/sec execute 71 sec latency 20.169 ms 1 9533 0.60 MB/sec execute 72 sec latency 63.883 ms 1 9859 0.65 MB/sec execute 73 sec latency 25.611 ms test_26 fail mds2 4 times Failing mds2 on oleg408-server 1 10392 0.68 MB/sec execute 74 sec latency 30.939 ms Stopping /mnt/lustre-mds2 (opts:) on oleg408-server 11:47:31 (1746114451) shut down facet: mds2 facet_host: oleg408-server facet_failover_host: oleg408-server 1 10589 0.71 MB/sec execute 75 sec latency 640.353 ms 1 10589 0.70 MB/sec execute 76 sec latency 1640.552 ms 1 10589 0.70 MB/sec execute 77 sec latency 2640.792 ms 1 10589 0.69 MB/sec execute 78 sec latency 3641.034 ms 1 10589 0.68 MB/sec execute 79 sec latency 4641.223 ms 1 10589 0.67 MB/sec execute 80 sec latency 5641.419 ms 1 10589 0.66 MB/sec execute 81 sec latency 6641.649 ms 1 10589 0.65 MB/sec execute 82 sec latency 7641.867 ms 1 10589 0.65 MB/sec execute 83 sec latency 8642.092 ms 1 10589 0.64 MB/sec execute 84 sec latency 9642.261 ms 1 10589 0.63 MB/sec execute 85 sec latency 10642.459 ms Failover mds2 to oleg408-server mount facets: mds2 1 10589 0.62 MB/sec execute 86 sec latency 11642.668 ms 1 10589 0.62 MB/sec execute 87 sec latency 12642.860 ms Starting mds2: -o localrecov /dev/mapper/mds2_flakey /mnt/lustre-mds2 1 10589 0.61 MB/sec execute 88 sec latency 13643.051 ms oleg408-server: oleg408-server.virtnet: executing set_default_debug -1 all pdsh@oleg408-client: oleg408-server: ssh exited with exit code 1 1 10589 0.60 MB/sec execute 89 sec latency 14643.235 ms Started lustre-MDT0001 11:47:45 (1746114465) targets are mounted 11:47:45 (1746114465) facet_failover done 1 10589 0.60 MB/sec execute 90 sec latency 15643.413 ms 1 10589 0.59 MB/sec execute 91 sec latency 16643.620 ms 1 10589 0.58 MB/sec execute 92 sec latency 17643.854 ms 1 10589 0.58 MB/sec execute 93 sec latency 18644.074 ms 1 10675 0.57 MB/sec execute 94 sec latency 19212.263 ms oleg408-client.virtnet: executing wait_import_state_mount (FULL|IDLE) mdc.lustre-MDT0001-mdc-*.mds_server_uuid mdc.lustre-MDT0001-mdc-*.mds_server_uuid in FULL state after 0 sec 1 10912 0.58 MB/sec execute 95 sec latency 18.927 ms 1 11073 0.58 MB/sec execute 96 sec latency 23.405 ms 1 11226 0.57 MB/sec execute 97 sec latency 26.116 ms 1 11424 0.57 MB/sec execute 98 sec latency 25.995 ms UUID 1K-blocks Used Available Use% Mounted on lustre-MDT0000_UUID 1414116 5012 1282676 1% /mnt/lustre[MDT:0] lustre-MDT0001_UUID 1414116 3076 1284612 1% /mnt/lustre[MDT:1] lustre-OST0000_UUID 3833116 38992 3549784 2% /mnt/lustre[OST:0] lustre-OST0001_UUID 3833116 20836 3558124 1% /mnt/lustre[OST:1] filesystem_summary: 7666232 59828 7107908 1% /mnt/lustre 1 11629 0.57 MB/sec execute 99 sec latency 25.919 ms 1 cleanup 100 sec 0 cleanup 101 sec Operation Count AvgLat MaxLat ---------------------------------------- NTCreateX 1842 21.877 19011.447 Close 1350 2.142 15.744 Rename 79 262.970 19343.325 Unlink 372 5.288 17.320 Qpathinfo 1693 14.809 19212.253 Qfileinfo 291 0.590 3.056 Qfsinfo 307 1.225 10.265 Sfileinfo 152 11.103 25.910 Find 646 25.267 14196.579 WriteX 909 2.648 12.995 ReadX 2891 0.150 17.415 LockX 6 2.420 2.799 UnlockX 6 2.483 2.934 Flush 129 12.694 238.449 Throughput 0.567363 MB/sec 1 clients 1 procs max_latency=19343.346 ms stopping dbench on /mnt/lustre at Thu May 1 11:47:56 EDT 2025 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 crush /mnt/lustre2/d26.replay-dual/run_dbench looking for dbench program /usr/bin/dbench found dbench client file /usr/share/dbench/client.txt test_26 fail mds1 5 times '/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 May 1 11:47:58 EDT 2025 waiting for dbench pid 8862 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 Failing mds1 on oleg408-server Stopping /mnt/lustre-mds1 (opts:) on oleg408-server 1 137 6.02 MB/sec warmup 1 sec latency 398.957 ms 11:47:59 (1746114479) shut down facet: mds1 facet_host: oleg408-server facet_failover_host: oleg408-server 1 137 3.01 MB/sec warmup 2 sec latency 1399.153 ms 1 137 2.01 MB/sec warmup 3 sec latency 2399.397 ms 1 137 1.51 MB/sec warmup 4 sec latency 3399.626 ms 1 137 1.20 MB/sec warmup 5 sec latency 4399.829 ms 1 137 1.00 MB/sec warmup 6 sec latency 5400.069 ms 1 137 0.86 MB/sec warmup 7 sec latency 6400.305 ms 1 137 0.75 MB/sec warmup 8 sec latency 7400.500 ms 1 137 0.67 MB/sec warmup 9 sec latency 8400.750 ms 1 137 0.60 MB/sec warmup 10 sec latency 9400.994 ms 1 137 0.55 MB/sec warmup 11 sec latency 10401.236 ms Failover mds1 to oleg408-server mount facets: mds1 1 137 0.50 MB/sec warmup 12 sec latency 11401.427 ms 1 137 0.46 MB/sec warmup 13 sec latency 12401.628 ms Starting mds1: -o localrecov /dev/mapper/mds1_flakey /mnt/lustre-mds1 1 137 0.43 MB/sec warmup 14 sec latency 13401.836 ms oleg408-server: oleg408-server.virtnet: executing set_default_debug -1 all 1 137 0.40 MB/sec warmup 15 sec latency 14402.020 ms pdsh@oleg408-client: oleg408-server: ssh exited with exit code 1 Started lustre-MDT0000 11:48:14 (1746114494) targets are mounted 11:48:14 (1746114494) facet_failover done 1 137 0.38 MB/sec warmup 16 sec latency 15402.225 ms 1 137 0.35 MB/sec warmup 17 sec latency 16402.499 ms 1 137 0.33 MB/sec warmup 18 sec latency 17402.736 ms 1 137 0.32 MB/sec warmup 19 sec latency 18402.949 ms oleg408-client.virtnet: executing wait_import_state_mount (FULL|IDLE) mdc.lustre-MDT0000-mdc-*.mds_server_uuid mdc.lustre-MDT0000-mdc-*.mds_server_uuid in FULL state after 0 sec 1 329 5.85 MB/sec execute 1 sec latency 19366.523 ms 1 533 6.24 MB/sec execute 2 sec latency 21.429 ms 1 803 5.13 MB/sec execute 3 sec latency 21.758 ms 1 990 3.95 MB/sec execute 4 sec latency 23.102 ms UUID 1K-blocks Used Available Use% Mounted on lustre-MDT0000_UUID 1414116 5452 1282236 1% /mnt/lustre[MDT:0] lustre-MDT0001_UUID 1414116 3488 1284200 1% /mnt/lustre[MDT:1] lustre-OST0000_UUID 3833116 1764 3573156 1% /mnt/lustre[OST:0] lustre-OST0001_UUID 3833116 35204 3529300 1% /mnt/lustre[OST:1] filesystem_summary: 7666232 36968 7102456 1% /mnt/lustre 1 1204 3.39 MB/sec execute 5 sec latency 24.362 ms 1 1487 3.33 MB/sec execute 6 sec latency 22.737 ms 1 1735 2.88 MB/sec execute 7 sec latency 29.065 ms 1 2026 2.56 MB/sec execute 8 sec latency 27.313 ms test_26 fail mds2 6 times 1 2340 2.42 MB/sec execute 9 sec latency 18.330 ms Failing mds2 on oleg408-server Stopping /mnt/lustre-mds2 (opts:) on oleg408-server 1 2555 2.28 MB/sec execute 10 sec latency 146.735 ms 11:48:29 (1746114509) shut down facet: mds2 facet_host: oleg408-server facet_failover_host: oleg408-server 1 2555 2.07 MB/sec execute 11 sec latency 1146.887 ms 1 2555 1.90 MB/sec execute 12 sec latency 2147.305 ms 1 2555 1.76 MB/sec execute 13 sec latency 3147.471 ms 1 2555 1.63 MB/sec execute 14 sec latency 4147.671 ms 1 2555 1.52 MB/sec execute 15 sec latency 5147.864 ms 1 2555 1.43 MB/sec execute 16 sec latency 6147.997 ms 1 2555 1.34 MB/sec execute 17 sec latency 7148.131 ms 1 2555 1.27 MB/sec execute 18 sec latency 8148.242 ms 1 2555 1.20 MB/sec execute 19 sec latency 9148.393 ms 1 2555 1.14 MB/sec execute 20 sec latency 10148.575 ms Failover mds2 to oleg408-server mount facets: mds2 1 2555 1.09 MB/sec execute 21 sec latency 11148.732 ms 1 2555 1.04 MB/sec execute 22 sec latency 12148.966 ms Starting mds2: -o localrecov /dev/mapper/mds2_flakey /mnt/lustre-mds2 1 2555 0.99 MB/sec execute 23 sec latency 13149.149 ms oleg408-server: oleg408-server.virtnet: executing set_default_debug -1 all 1 2555 0.95 MB/sec execute 24 sec latency 14149.359 ms pdsh@oleg408-client: oleg408-server: ssh exited with exit code 1 Started lustre-MDT0001 11:48:43 (1746114523) targets are mounted 11:48:43 (1746114523) facet_failover done 1 2555 0.91 MB/sec execute 25 sec latency 15149.543 ms 1 2555 0.88 MB/sec execute 26 sec latency 16149.744 ms 1 2555 0.85 MB/sec execute 27 sec latency 17150.016 ms 1 2555 0.81 MB/sec execute 28 sec latency 18150.235 ms 1 2574 0.79 MB/sec execute 29 sec latency 19025.580 ms oleg408-client.virtnet: executing wait_import_state_mount (FULL|IDLE) mdc.lustre-MDT0001-mdc-*.mds_server_uuid mdc.lustre-MDT0001-mdc-*.mds_server_uuid in FULL state after 0 sec 1 2990 0.91 MB/sec execute 30 sec latency 20.372 ms 1 3374 0.95 MB/sec execute 31 sec latency 24.229 ms 1 3695 1.02 MB/sec execute 32 sec latency 17.814 ms 1 3910 1.03 MB/sec execute 33 sec latency 17.347 ms UUID 1K-blocks Used Available Use% Mounted on lustre-MDT0000_UUID 1414116 4448 1283240 1% /mnt/lustre[MDT:0] lustre-MDT0001_UUID 1414116 2700 1284988 1% /mnt/lustre[MDT:1] 1 4068 1.00 MB/sec execute 34 sec latency 26.689 ms lustre-OST0000_UUID 3833116 12056 3585036 1% /mnt/lustre[OST:0] lustre-OST0001_UUID 3833116 53544 3543392 2% /mnt/lustre[OST:1] filesystem_summary: 7666232 65600 7128428 1% /mnt/lustre 1 4270 0.98 MB/sec execute 35 sec latency 38.138 ms 1 4381 0.96 MB/sec execute 36 sec latency 468.661 ms 1 4564 0.94 MB/sec execute 37 sec latency 33.612 ms test_26 fail mds1 7 times 1 4821 0.95 MB/sec execute 38 sec latency 23.558 ms Failing mds1 on oleg408-server Stopping /mnt/lustre-mds1 (opts:) on oleg408-server 1 5057 1.00 MB/sec execute 39 sec latency 121.899 ms 11:48:58 (1746114538) shut down facet: mds1 facet_host: oleg408-server facet_failover_host: oleg408-server 1 5057 0.98 MB/sec execute 40 sec latency 1122.105 ms 1 5057 0.95 MB/sec execute 41 sec latency 2122.367 ms 1 5057 0.93 MB/sec execute 42 sec latency 3122.566 ms 1 5057 0.91 MB/sec execute 43 sec latency 4122.769 ms 1 5057 0.89 MB/sec execute 44 sec latency 5122.969 ms 1 5057 0.87 MB/sec execute 45 sec latency 6123.195 ms 1 5057 0.85 MB/sec execute 46 sec latency 7123.471 ms 1 5057 0.83 MB/sec execute 47 sec latency 8123.728 ms 1 5057 0.81 MB/sec execute 48 sec latency 9123.926 ms 1 5057 0.80 MB/sec execute 49 sec latency 10124.186 ms Failover mds1 to oleg408-server mount facets: mds1 1 5057 0.78 MB/sec execute 50 sec latency 11124.382 ms 1 5057 0.77 MB/sec execute 51 sec latency 12124.555 ms Starting mds1: -o localrecov /dev/mapper/mds1_flakey /mnt/lustre-mds1 1 5057 0.75 MB/sec execute 52 sec latency 13124.752 ms 1 5057 0.74 MB/sec execute 53 sec latency 14124.928 ms oleg408-server: oleg408-server.virtnet: executing set_default_debug -1 all pdsh@oleg408-client: oleg408-server: ssh exited with exit code 1 Started lustre-MDT0000 11:49:12 (1746114552) targets are mounted 11:49:12 (1746114552) facet_failover done 1 5057 0.72 MB/sec execute 54 sec latency 15125.158 ms 1 5057 0.71 MB/sec execute 55 sec latency 16125.357 ms 1 5057 0.70 MB/sec execute 56 sec latency 17125.548 ms 1 5057 0.69 MB/sec execute 57 sec latency 18125.741 ms 1 5057 0.67 MB/sec execute 58 sec latency 19126.040 ms 1 5143 0.66 MB/sec execute 59 sec latency 19616.400 ms oleg408-client.virtnet: executing wait_import_state_mount (FULL|IDLE) mdc.lustre-MDT0000-mdc-*.mds_server_uuid mdc.lustre-MDT0000-mdc-*.mds_server_uuid in FULL state after 0 sec 1 5344 0.66 MB/sec execute 60 sec latency 27.577 ms 1 5545 0.65 MB/sec execute 61 sec latency 28.769 ms 1 5738 0.64 MB/sec execute 62 sec latency 21.548 ms 1 5992 0.65 MB/sec execute 63 sec latency 26.392 ms 1 6320 0.70 MB/sec execute 64 sec latency 21.881 ms 1 6741 0.73 MB/sec execute 65 sec latency 24.650 ms 1 7073 0.78 MB/sec execute 66 sec latency 34.537 ms 1 7336 0.79 MB/sec execute 67 sec latency 23.085 ms 1 7531 0.78 MB/sec execute 68 sec latency 23.812 ms 1 7766 0.77 MB/sec execute 69 sec latency 21.798 ms 1 7990 0.77 MB/sec execute 70 sec latency 44.925 ms 1 8227 0.78 MB/sec execute 71 sec latency 44.070 ms 1 8498 0.81 MB/sec execute 72 sec latency 23.998 ms 1 8699 0.80 MB/sec execute 73 sec latency 34.954 ms 1 8910 0.79 MB/sec execute 74 sec latency 19.959 ms 1 9140 0.78 MB/sec execute 75 sec latency 21.259 ms 1 9405 0.79 MB/sec execute 76 sec latency 22.536 ms dbench killed by signal 15 stopping dbench on /mnt/lustre at Thu May 1 11:49:35 EDT 2025 with return code 0 8862 pts/0 S+ 0:00 dbench -c client.txt 1 -D /mnt/lustre2/d26.replay-dual/run_dbench -t 100 killed dbench main pid 8862 clean dbench files on /mnt/lustre /mnt/lustre /mnt/lustre removed 'client.txt' /mnt/lustre dbench successfully finished