== replay-dual test 26: dbench and tar with mds failover ========================================================== 16:42:28 (1748983348) Starting client oleg150-client.virtnet: -o user_xattr,flock oleg150-server@tcp:/lustre /mnt/lustre Started clients oleg150-client.virtnet: 192.168.201.150@tcp:/lustre on /mnt/lustre type lustre (rw,checksum,flock,user_xattr,lruresize,lazystatfs,nouser_fid2path,verbose,noencrypt) Started tar loop with pid 17523 Started dbench loop with 17524 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 Tue Jun 3 16:42:29 EDT 2025 waiting for dbench pid 17549 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 1 174 7.06 MB/sec warmup 1 sec latency 34.723 ms UUID 1K-blocks Used Available Use% Mounted on lustre-MDT0000_UUID 1414116 1872 1285816 1% /mnt/lustre[MDT:0] lustre-OST0000_UUID 3833116 25844 3556200 1% /mnt/lustre[OST:0] lustre-OST0001_UUID 3833116 1252 3605768 1% /mnt/lustre[OST:1] filesystem_summary: 7666232 27096 7161968 1% /mnt/lustre 1 342 6.13 MB/sec warmup 2 sec latency 31.045 ms 1 494 5.73 MB/sec warmup 3 sec latency 29.599 ms 1 717 5.32 MB/sec warmup 4 sec latency 37.878 ms 1 900 4.34 MB/sec warmup 5 sec latency 26.549 ms test_26 fail mds1 1 times 1 1106 3.83 MB/sec warmup 6 sec latency 23.964 ms Failing mds1 on oleg150-server Stopping /mnt/lustre-mds1 (opts:) on oleg150-server 1 1376 3.71 MB/sec warmup 7 sec latency 29.071 ms reboot facets: mds1 1 1376 3.25 MB/sec warmup 8 sec latency 1029.260 ms 1 1376 2.89 MB/sec warmup 9 sec latency 2029.483 ms 1 1376 2.60 MB/sec warmup 10 sec latency 3029.703 ms 1 1376 2.36 MB/sec warmup 11 sec latency 4029.894 ms 1 1376 2.17 MB/sec warmup 12 sec latency 5030.095 ms 1 1376 2.00 MB/sec warmup 13 sec latency 6030.309 ms 1 1376 1.86 MB/sec warmup 14 sec latency 7030.519 ms 1 1376 1.73 MB/sec warmup 15 sec latency 8030.765 ms 1 1376 1.62 MB/sec warmup 16 sec latency 9030.965 ms 1 1376 1.53 MB/sec warmup 17 sec latency 10031.175 ms Failover mds1 to oleg150-server 1 1376 1.44 MB/sec warmup 18 sec latency 11031.383 ms mount facets: mds1 1 1376 1.37 MB/sec warmup 19 sec latency 12031.563 ms 1 1376 0.00 MB/sec execute 1 sec latency 14031.866 ms Starting mds1: -o localrecov /dev/mapper/mds1_flakey /mnt/lustre-mds1 1 1376 0.00 MB/sec execute 2 sec latency 15032.049 ms 1 1376 0.00 MB/sec execute 3 sec latency 16032.245 ms oleg150-server: oleg150-server.virtnet: executing set_default_debug -1 all 8 pdsh@oleg150-client: oleg150-server: ssh exited with exit code 1 1 1376 0.00 MB/sec execute 4 sec latency 17032.396 ms Started lustre-MDT0000 1 1376 0.00 MB/sec execute 5 sec latency 18032.592 ms 1 1555 0.02 MB/sec execute 6 sec latency 18097.319 ms 1 1762 0.04 MB/sec execute 7 sec latency 30.666 ms oleg150-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 2006 0.06 MB/sec execute 8 sec latency 24.606 ms 1 2345 0.21 MB/sec execute 9 sec latency 25.623 ms 1 2716 0.59 MB/sec execute 10 sec latency 17.526 ms 1 3013 0.69 MB/sec execute 11 sec latency 24.355 ms UUID 1K-blocks Used Available Use% Mounted on lustre-MDT0000_UUID 1414116 2280 1285408 1% /mnt/lustre[MDT:0] lustre-OST0000_UUID 3833116 39092 3565696 2% /mnt/lustre[OST:0] lustre-OST0001_UUID 3833116 10892 3593752 1% /mnt/lustre[OST:1] filesystem_summary: 7666232 49984 7159448 1% /mnt/lustre 1 3395 0.79 MB/sec execute 12 sec latency 26.654 ms 1 3702 0.98 MB/sec execute 13 sec latency 17.375 ms 1 3860 0.98 MB/sec execute 14 sec latency 27.533 ms 1 4028 0.94 MB/sec execute 15 sec latency 18.451 ms 1 4236 0.89 MB/sec execute 16 sec latency 21.802 ms test_26 fail mds1 2 times Failing mds1 on oleg150-server 1 4421 0.86 MB/sec execute 17 sec latency 41.653 ms Stopping /mnt/lustre-mds1 (opts:) on oleg150-server 1 4538 0.83 MB/sec execute 18 sec latency 327.911 ms reboot facets: mds1 1 4538 0.79 MB/sec execute 19 sec latency 1328.140 ms 1 4538 0.75 MB/sec execute 20 sec latency 2328.378 ms 1 4538 0.71 MB/sec execute 21 sec latency 3328.636 ms 1 4538 0.68 MB/sec execute 22 sec latency 4328.856 ms 1 4538 0.65 MB/sec execute 23 sec latency 5329.059 ms 1 4538 0.62 MB/sec execute 24 sec latency 6329.279 ms 1 4538 0.60 MB/sec execute 25 sec latency 7329.543 ms 1 4538 0.57 MB/sec execute 26 sec latency 8329.755 ms 1 4538 0.55 MB/sec execute 27 sec latency 9329.926 ms 1 4538 0.53 MB/sec execute 28 sec latency 10330.138 ms Failover mds1 to oleg150-server 1 4538 0.52 MB/sec execute 29 sec latency 11330.349 ms mount facets: mds1 1 4538 0.50 MB/sec execute 30 sec latency 12330.551 ms 1 4538 0.48 MB/sec execute 31 sec latency 13330.766 ms 1 4538 0.47 MB/sec execute 32 sec latency 14331.033 ms Starting mds1: -o localrecov /dev/mapper/mds1_flakey /mnt/lustre-mds1 1 4538 0.45 MB/sec execute 33 sec latency 15331.306 ms 1 4538 0.44 MB/sec execute 34 sec latency 16331.522 ms 1 4538 0.43 MB/sec execute 35 sec latency 17331.767 ms 1 4538 0.41 MB/sec execute 36 sec latency 18331.998 ms oleg150-server: oleg150-server.virtnet: executing set_default_debug -1 all 8 1 4538 0.40 MB/sec execute 37 sec latency 19332.240 ms pdsh@oleg150-client: oleg150-server: ssh exited with exit code 1 Started lustre-MDT0000 1 4538 0.39 MB/sec execute 38 sec latency 20332.440 ms 1 4538 0.38 MB/sec execute 39 sec latency 21332.646 ms 1 4538 0.37 MB/sec execute 40 sec latency 22332.863 ms 1 4538 0.36 MB/sec execute 41 sec latency 23333.111 ms 1 4538 0.36 MB/sec execute 42 sec latency 24333.335 ms 1 4538 0.35 MB/sec execute 43 sec latency 25333.540 ms 1 4538 0.34 MB/sec execute 44 sec latency 26333.798 ms 1 4538 0.33 MB/sec execute 45 sec latency 27334.093 ms 1 4538 0.32 MB/sec execute 46 sec latency 28334.307 ms 1 4538 0.32 MB/sec execute 47 sec latency 29334.503 ms 1 4538 0.31 MB/sec execute 48 sec latency 30334.729 ms 1 4538 0.30 MB/sec execute 49 sec latency 31334.937 ms 1 4538 0.30 MB/sec execute 50 sec latency 32335.131 ms 1 4651 0.31 MB/sec execute 51 sec latency 32860.971 ms oleg150-client.virtnet: executing wait_import_state_mount (FULL|IDLE) mdc.lustre-MDT0000-mdc-*.mds_server_uuid 1 4972 0.37 MB/sec execute 52 sec latency 22.456 ms mdc.lustre-MDT0000-mdc-*.mds_server_uuid in FULL state after 0 sec 1 5208 0.36 MB/sec execute 53 sec latency 25.125 ms 1 5393 0.36 MB/sec execute 54 sec latency 27.887 ms 1 5625 0.36 MB/sec execute 55 sec latency 26.360 ms 1 5905 0.38 MB/sec execute 56 sec latency 17.507 ms UUID 1K-blocks Used Available Use% Mounted on lustre-MDT0000_UUID 1414116 2808 1284880 1% /mnt/lustre[MDT:0] lustre-OST0000_UUID 3833116 42272 3563708 2% /mnt/lustre[OST:0] lustre-OST0001_UUID 3833116 16008 3589412 1% /mnt/lustre[OST:1] filesystem_summary: 7666232 58280 7153120 1% /mnt/lustre 1 6155 0.39 MB/sec execute 57 sec latency 24.161 ms 1 6537 0.46 MB/sec execute 58 sec latency 444.988 ms 1 6905 0.48 MB/sec execute 59 sec latency 563.257 ms 1 7190 0.53 MB/sec execute 60 sec latency 20.841 ms test_26 fail mds1 3 times 1 7360 0.54 MB/sec execute 61 sec latency 27.425 ms Failing mds1 on oleg150-server Stopping /mnt/lustre-mds1 (opts:) on oleg150-server 1 7476 0.54 MB/sec execute 62 sec latency 189.613 ms reboot facets: mds1 1 7476 0.53 MB/sec execute 63 sec latency 1189.857 ms 1 7476 0.52 MB/sec execute 64 sec latency 2190.082 ms 1 7476 0.51 MB/sec execute 65 sec latency 3190.390 ms 1 7476 0.50 MB/sec execute 66 sec latency 4190.639 ms 1 7476 0.50 MB/sec execute 67 sec latency 5190.843 ms 1 7476 0.49 MB/sec execute 68 sec latency 6191.071 ms 1 7476 0.48 MB/sec execute 69 sec latency 7191.403 ms 1 7476 0.47 MB/sec execute 70 sec latency 8191.629 ms 1 7476 0.47 MB/sec execute 71 sec latency 9191.854 ms 1 7476 0.46 MB/sec execute 72 sec latency 10192.193 ms Failover mds1 to oleg150-server 1 7476 0.45 MB/sec execute 73 sec latency 11192.405 ms mount facets: mds1 1 7476 0.45 MB/sec execute 74 sec latency 12192.609 ms 1 7476 0.44 MB/sec execute 75 sec latency 13192.810 ms 1 7476 0.44 MB/sec execute 76 sec latency 14193.024 ms Starting mds1: -o localrecov /dev/mapper/mds1_flakey /mnt/lustre-mds1 1 7476 0.43 MB/sec execute 77 sec latency 15193.223 ms 1 7476 0.43 MB/sec execute 78 sec latency 16193.450 ms 1 7476 0.42 MB/sec execute 79 sec latency 17193.662 ms 1 7476 0.41 MB/sec execute 80 sec latency 18193.954 ms oleg150-server: oleg150-server.virtnet: executing set_default_debug -1 all 8 pdsh@oleg150-client: oleg150-server: ssh exited with exit code 1 Started lustre-MDT0000 1 7476 0.41 MB/sec execute 81 sec latency 19194.141 ms 1 7476 0.40 MB/sec execute 82 sec latency 20194.433 ms 1 7476 0.40 MB/sec execute 83 sec latency 21194.612 ms 1 7476 0.40 MB/sec execute 84 sec latency 22194.821 ms 1 7476 0.39 MB/sec execute 85 sec latency 23195.092 ms 1 7476 0.39 MB/sec execute 86 sec latency 24195.374 ms 1 7476 0.38 MB/sec execute 87 sec latency 25195.615 ms 1 7476 0.38 MB/sec execute 88 sec latency 26195.878 ms 1 7476 0.37 MB/sec execute 89 sec latency 27196.100 ms 1 7476 0.37 MB/sec execute 90 sec latency 28196.310 ms 1 7476 0.36 MB/sec execute 91 sec latency 29196.527 ms 1 7536 0.36 MB/sec execute 92 sec latency 29788.220 ms 1 7700 0.36 MB/sec execute 93 sec latency 24.126 ms oleg150-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 7899 0.36 MB/sec execute 94 sec latency 36.619 ms 1 8088 0.36 MB/sec execute 95 sec latency 24.822 ms 1 8306 0.37 MB/sec execute 96 sec latency 25.449 ms 1 8574 0.39 MB/sec execute 97 sec latency 25.291 ms UUID 1K-blocks Used Available Use% Mounted on lustre-MDT0000_UUID 1414116 3300 1284388 1% /mnt/lustre[MDT:0] lustre-OST0000_UUID 3833116 43544 3562820 2% /mnt/lustre[OST:0] lustre-OST0001_UUID 3833116 16988 3589760 1% /mnt/lustre[OST:1] filesystem_summary: 7666232 60532 7152580 1% /mnt/lustre 1 8755 0.39 MB/sec execute 98 sec latency 41.300 ms 1 8942 0.39 MB/sec execute 99 sec latency 26.223 ms 1 cleanup 100 sec 0 cleanup 101 sec Operation Count AvgLat MaxLat ---------------------------------------- NTCreateX 1398 33.671 29788.204 Close 1022 2.447 7.621 Rename 53 636.550 32860.951 Unlink 258 5.680 17.868 Qpathinfo 1293 3.143 31.383 Qfileinfo 223 0.496 3.504 Qfsinfo 204 0.569 2.801 Sfileinfo 100 8.094 13.901 Find 472 39.887 18097.291 WriteX 667 3.054 11.508 ReadX 2031 0.211 25.612 LockX 4 2.320 2.941 UnlockX 4 2.295 2.758 Flush 92 19.384 563.241 Throughput 0.389787 MB/sec 1 clients 1 procs max_latency=32860.971 ms stopping dbench on /mnt/lustre at Tue Jun 3 16:44:30 EDT 2025 with return code 0 clean dbench files on /mnt/lustre /mnt/lustre /mnt/lustre removed 'client.txt' /mnt/lustre dbench successfully finished 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 Tue Jun 3 16:44:31 EDT 2025 waiting for dbench pid 21297 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 mds1 4 times Failing mds1 on oleg150-server Stopping /mnt/lustre-mds1 (opts:) on oleg150-server 1 197 7.84 MB/sec warmup 1 sec latency 28.692 ms 1 248 4.70 MB/sec warmup 2 sec latency 647.640 ms reboot facets: mds1 1 248 3.13 MB/sec warmup 3 sec latency 1647.893 ms 1 248 2.35 MB/sec warmup 4 sec latency 2648.192 ms 1 248 1.88 MB/sec warmup 5 sec latency 3648.449 ms 1 248 1.57 MB/sec warmup 6 sec latency 4648.732 ms 1 248 1.34 MB/sec warmup 7 sec latency 5649.032 ms 1 248 1.18 MB/sec warmup 8 sec latency 6649.235 ms 1 248 1.04 MB/sec warmup 9 sec latency 7649.444 ms 1 248 0.94 MB/sec warmup 10 sec latency 8649.665 ms 1 248 0.85 MB/sec warmup 11 sec latency 9649.836 ms 1 248 0.78 MB/sec warmup 12 sec latency 10650.147 ms Failover mds1 to oleg150-server mount facets: mds1 1 248 0.72 MB/sec warmup 13 sec latency 11650.362 ms 1 248 0.67 MB/sec warmup 14 sec latency 12650.543 ms 1 248 0.63 MB/sec warmup 15 sec latency 13650.673 ms Starting mds1: -o localrecov /dev/mapper/mds1_flakey /mnt/lustre-mds1 1 248 0.59 MB/sec warmup 16 sec latency 14650.872 ms 1 248 0.55 MB/sec warmup 17 sec latency 15651.089 ms 1 248 0.52 MB/sec warmup 18 sec latency 16651.236 ms 1 248 0.49 MB/sec warmup 19 sec latency 17651.435 ms 1 248 0.00 MB/sec execute 1 sec latency 19651.841 ms oleg150-server: oleg150-server.virtnet: executing set_default_debug -1 all 8 pdsh@oleg150-client: oleg150-server: ssh exited with exit code 1 Started lustre-MDT0000 1 248 0.00 MB/sec execute 2 sec latency 20652.030 ms 1 248 0.00 MB/sec execute 3 sec latency 21652.194 ms 1 248 0.00 MB/sec execute 4 sec latency 22652.393 ms 1 248 0.00 MB/sec execute 5 sec latency 23652.582 ms 1 248 0.00 MB/sec execute 6 sec latency 24652.780 ms 1 248 0.00 MB/sec execute 7 sec latency 25652.936 ms 1 248 0.00 MB/sec execute 8 sec latency 26653.091 ms 1 248 0.00 MB/sec execute 9 sec latency 27653.290 ms 1 248 0.00 MB/sec execute 10 sec latency 28653.445 ms 1 248 0.00 MB/sec execute 11 sec latency 29653.637 ms 1 248 0.00 MB/sec execute 12 sec latency 30653.836 ms 1 413 0.40 MB/sec execute 13 sec latency 30807.044 ms oleg150-client.virtnet: executing wait_import_state_mount (FULL|IDLE) mdc.lustre-MDT0000-mdc-*.mds_server_uuid 1 642 0.83 MB/sec execute 14 sec latency 28.867 ms mdc.lustre-MDT0000-mdc-*.mds_server_uuid in FULL state after 0 sec 1 813 0.81 MB/sec execute 15 sec latency 26.156 ms 1 987 0.78 MB/sec execute 16 sec latency 24.004 ms 1 1202 0.80 MB/sec execute 17 sec latency 25.391 ms 1 1481 0.92 MB/sec execute 18 sec latency 27.209 ms UUID 1K-blocks Used Available Use% Mounted on lustre-MDT0000_UUID 1414116 3840 1283848 1% /mnt/lustre[MDT:0] lustre-OST0000_UUID 3833116 37416 3563220 2% /mnt/lustre[OST:0] lustre-OST0001_UUID 3833116 14284 3585124 1% /mnt/lustre[OST:1] filesystem_summary: 7666232 51700 7148344 1% /mnt/lustre 1 1691 0.89 MB/sec execute 19 sec latency 29.962 ms 1 1884 0.85 MB/sec execute 20 sec latency 206.433 ms 1 2118 0.82 MB/sec execute 21 sec latency 541.235 ms 1 2367 0.84 MB/sec execute 22 sec latency 28.454 ms test_26 fail mds1 5 times 1 2582 0.85 MB/sec execute 23 sec latency 23.072 ms Failing mds1 on oleg150-server Stopping /mnt/lustre-mds1 (opts:) on oleg150-server 1 2973 1.00 MB/sec execute 24 sec latency 27.502 ms reboot facets: mds1 1 2973 0.96 MB/sec execute 25 sec latency 1027.690 ms 1 2973 0.93 MB/sec execute 26 sec latency 2027.908 ms 1 2973 0.89 MB/sec execute 27 sec latency 3028.149 ms 1 2973 0.86 MB/sec execute 28 sec latency 4028.360 ms 1 2973 0.83 MB/sec execute 29 sec latency 5028.565 ms 1 2973 0.80 MB/sec execute 30 sec latency 6028.772 ms 1 2973 0.78 MB/sec execute 31 sec latency 7028.960 ms 1 2973 0.75 MB/sec execute 32 sec latency 8029.144 ms 1 2973 0.73 MB/sec execute 33 sec latency 9029.355 ms 1 2973 0.71 MB/sec execute 34 sec latency 10029.552 ms Failover mds1 to oleg150-server 1 2973 0.69 MB/sec execute 35 sec latency 11029.742 ms mount facets: mds1 1 2973 0.67 MB/sec execute 36 sec latency 12029.926 ms 1 2973 0.65 MB/sec execute 37 sec latency 13030.151 ms Starting mds1: -o localrecov /dev/mapper/mds1_flakey /mnt/lustre-mds1 1 2973 0.63 MB/sec execute 38 sec latency 14030.302 ms 1 2973 0.62 MB/sec execute 39 sec latency 15030.518 ms 1 2973 0.60 MB/sec execute 40 sec latency 16030.667 ms 1 2973 0.59 MB/sec execute 41 sec latency 17030.824 ms 1 2973 0.57 MB/sec execute 42 sec latency 18030.992 ms 1 2973 0.56 MB/sec execute 43 sec latency 19031.202 ms 1 2973 0.55 MB/sec execute 44 sec latency 20031.412 ms oleg150-server: oleg150-server.virtnet: executing set_default_debug -1 all 8 pdsh@oleg150-client: oleg150-server: ssh exited with exit code 1 1 2973 0.54 MB/sec execute 45 sec latency 21031.631 ms Started lustre-MDT0000 1 2973 0.52 MB/sec execute 46 sec latency 22031.892 ms 1 2973 0.51 MB/sec execute 47 sec latency 23032.100 ms 1 2973 0.50 MB/sec execute 48 sec latency 24032.310 ms 1 2973 0.49 MB/sec execute 49 sec latency 25032.520 ms 1 2973 0.48 MB/sec execute 50 sec latency 26032.655 ms 1 2973 0.47 MB/sec execute 51 sec latency 27032.858 ms 1 2973 0.46 MB/sec execute 52 sec latency 28033.066 ms 1 2973 0.45 MB/sec execute 53 sec latency 29033.258 ms 1 2974 0.45 MB/sec execute 54 sec latency 30002.091 ms 1 3351 0.47 MB/sec execute 55 sec latency 37.355 ms oleg150-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 3683 0.52 MB/sec execute 56 sec latency 19.335 ms 1 3871 0.53 MB/sec execute 57 sec latency 22.828 ms 1 4007 0.53 MB/sec execute 58 sec latency 55.331 ms 1 4184 0.52 MB/sec execute 59 sec latency 26.260 ms tar: Unexpected EOF in archive tar: Unexpected EOF in archive 1 4372 0.52 MB/sec execute 60 sec latency 37.585 ms 1 4549 0.52 MB/sec execute 61 sec latency 25.942 ms 1 4767 0.53 MB/sec execute 62 sec latency 25.306 ms tar: Error is not recoverable: exiting now dbench killed by signal 15 stopping dbench on /mnt/lustre at Tue Jun 3 16:45:54 EDT 2025 with return code 0 21297 pts/0 S+ 0:00 dbench -c client.txt 1 -D /mnt/lustre2/d26.replay-dual/run_dbench -t 100 killed dbench main pid 21297 clean dbench files on /mnt/lustre /mnt/lustre /mnt/lustre removed 'client.txt' /mnt/lustre dbench successfully finished