== replay-dual test 26: dbench and tar with mds failover ========================================================== 22:31:12 (1788489072) Starting client oleg250-client.virtnet: -o user_xattr,flock 192.168.202.150@tcp:/lustre /mnt/lustre Started clients oleg250-client.virtnet: 192.168.202.150@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 65694 Started dbench loop with 65696 striped dir -i0 -c2 -H all_char /mnt/lustre2/d26.replay-dual/run_dbench striped dir -i0 -c2 -H crush2 /mnt/lustre/d26.replay-dual/run_tar 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 1756 1269548 1% /mnt/lustre[MDT:1] lustre-OST0000_UUID 3833116 1564 3605456 1% /mnt/lustre[OST:0] lustre-OST0001_UUID 3833116 17928 3584896 1% /mnt/lustre[OST:1] filesystem_summary: 7666232 19492 7190352 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 3 22:31:19 EDT 2026 waiting for dbench pid 65773 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 0 8 0.00 MB/sec warmup 1 sec latency 23.903 ms 1 45 0.88 MB/sec warmup 2 sec latency 197.900 ms 1 91 1.40 MB/sec warmup 3 sec latency 159.517 ms 1 113 1.31 MB/sec warmup 4 sec latency 287.739 ms 1 137 1.20 MB/sec warmup 5 sec latency 251.403 ms 1 157 1.09 MB/sec warmup 6 sec latency 297.217 ms 1 184 1.05 MB/sec warmup 7 sec latency 176.093 ms 1 197 0.98 MB/sec warmup 8 sec latency 658.250 ms 1 218 0.93 MB/sec warmup 9 sec latency 528.109 ms 1 267 1.00 MB/sec warmup 10 sec latency 125.323 ms 1 313 1.04 MB/sec warmup 11 sec latency 106.815 ms test_26 fail mds1 1 times 1 341 1.02 MB/sec warmup 12 sec latency 194.382 ms 1 372 1.02 MB/sec warmup 13 sec latency 157.874 ms Failing mds1 on oleg250-server 1 397 1.00 MB/sec warmup 14 sec latency 159.047 ms 1 432 1.01 MB/sec warmup 15 sec latency 129.193 ms Stopping /mnt/lustre-mds1 (opts:) on oleg250-server 1 476 1.04 MB/sec warmup 16 sec latency 154.565 ms 1 501 1.03 MB/sec warmup 17 sec latency 535.536 ms 1 501 0.97 MB/sec warmup 18 sec latency 1535.785 ms 22:31:38 (1788489098) shut down 1 501 0.92 MB/sec warmup 19 sec latency 2537.818 ms facet: mds1 facet_host: oleg250-server facet_failover_host: oleg250-server 1 501 0.00 MB/sec execute 1 sec latency 4538.460 ms 1 501 0.00 MB/sec execute 2 sec latency 5540.056 ms 1 501 0.00 MB/sec execute 3 sec latency 6540.944 ms 1 501 0.00 MB/sec execute 4 sec latency 7541.401 ms 1 501 0.00 MB/sec execute 5 sec latency 8541.676 ms 1 501 0.00 MB/sec execute 6 sec latency 9542.125 ms 1 501 0.00 MB/sec execute 7 sec latency 10542.901 ms 1 501 0.00 MB/sec execute 8 sec latency 11543.171 ms 1 501 0.00 MB/sec execute 9 sec latency 12543.906 ms Failover mds1 to oleg250-server mount facets: mds1 1 501 0.00 MB/sec execute 10 sec latency 13548.854 ms 1 501 0.00 MB/sec execute 11 sec latency 14549.190 ms 1 501 0.00 MB/sec execute 12 sec latency 15549.525 ms 1 501 0.00 MB/sec execute 13 sec latency 16550.312 ms 1 501 0.00 MB/sec execute 14 sec latency 17553.171 ms 1 501 0.00 MB/sec execute 15 sec latency 18554.131 ms 1 501 0.00 MB/sec execute 16 sec latency 19554.433 ms 1 501 0.00 MB/sec execute 17 sec latency 20555.025 ms 1 501 0.00 MB/sec execute 18 sec latency 21555.200 ms 1 501 0.00 MB/sec execute 19 sec latency 22555.332 ms Start mds1: mount -t lustre -o localrecov /dev/mapper/mds1_flakey /mnt/lustre-mds1 1 501 0.00 MB/sec execute 20 sec latency 23556.143 ms 1 501 0.00 MB/sec execute 21 sec latency 24557.664 ms 1 501 0.00 MB/sec execute 22 sec latency 25558.063 ms 1 501 0.00 MB/sec execute 23 sec latency 26558.343 ms 1 501 0.00 MB/sec execute 24 sec latency 27560.130 ms 1 501 0.00 MB/sec execute 25 sec latency 28560.790 ms 1 501 0.00 MB/sec execute 26 sec latency 29561.122 ms 1 501 0.00 MB/sec execute 27 sec latency 30561.600 ms 1 501 0.00 MB/sec execute 28 sec latency 31571.927 ms 1 501 0.00 MB/sec execute 29 sec latency 32572.106 ms 1 501 0.00 MB/sec execute 30 sec latency 33572.355 ms 1 501 0.00 MB/sec execute 31 sec latency 34572.846 ms 1 501 0.00 MB/sec execute 32 sec latency 35574.123 ms 1 501 0.00 MB/sec execute 33 sec latency 36574.321 ms oleg250-server: oleg250-server.virtnet: executing set_default_debug -1 all 1 501 0.00 MB/sec execute 34 sec latency 37575.726 ms 1 501 0.00 MB/sec execute 35 sec latency 38576.768 ms pdsh@oleg250-client: oleg250-server: ssh exited with exit code 1 1 503 0.00 MB/sec execute 36 sec latency 39411.891 ms Started lustre-MDT0000 22:32:16 (1788489136) targets are mounted 22:32:16 (1788489136) facet_failover done 1 529 0.02 MB/sec execute 37 sec latency 246.760 ms 1 552 0.04 MB/sec execute 38 sec latency 766.405 ms 1 603 0.09 MB/sec execute 39 sec latency 950.556 ms 1 642 0.09 MB/sec execute 40 sec latency 142.215 ms 1 676 0.09 MB/sec execute 41 sec latency 129.790 ms 1 705 0.09 MB/sec execute 42 sec latency 168.538 ms 1 728 0.09 MB/sec execute 43 sec latency 168.044 ms 1 759 0.09 MB/sec execute 44 sec latency 147.412 ms 1 776 0.09 MB/sec execute 45 sec latency 213.568 ms 1 799 0.09 MB/sec execute 46 sec latency 191.640 ms oleg250-client.virtnet: executing wait_import_state_mount (FULL|IDLE) mdc.lustre-MDT0000-mdc-*.mds_server_uuid 1475 0 1 848 0.09 MB/sec execute 47 sec latency 129.263 ms 1 873 0.09 MB/sec execute 48 sec latency 172.449 ms mdc.lustre-MDT0000-mdc-*.mds_server_uuid in FULL state after 0 sec 1 888 0.09 MB/sec execute 49 sec latency 207.446 ms 1 909 0.09 MB/sec execute 50 sec latency 190.371 ms 1 924 0.08 MB/sec execute 51 sec latency 205.167 ms 1 967 0.08 MB/sec execute 52 sec latency 136.274 ms 1 988 0.08 MB/sec execute 53 sec latency 120.702 ms 1 1048 0.10 MB/sec execute 54 sec latency 108.676 ms 1 1113 0.10 MB/sec execute 55 sec latency 91.321 ms 1 1155 0.10 MB/sec execute 56 sec latency 143.998 ms UUID 1K-blocks Used Available Use% Mounted on lustre-MDT0000_UUID 1414116 2188 1269116 1% /mnt/lustre[MDT:0] lustre-MDT0001_UUID 1414116 1944 1269360 1% /mnt/lustre[MDT:1] lustre-OST0000_UUID 3833116 10088 3594860 1% /mnt/lustre[OST:0] lustre-OST0001_UUID 3833116 45724 3554248 2% /mnt/lustre[OST:1] filesystem_summary: 7666232 55812 7149108 1% /mnt/lustre 1 1172 0.10 MB/sec execute 57 sec latency 661.507 ms 1 1190 0.10 MB/sec execute 58 sec latency 947.815 ms 1 1224 0.09 MB/sec execute 59 sec latency 114.711 ms 1 1252 0.09 MB/sec execute 60 sec latency 135.975 ms 1 1274 0.09 MB/sec execute 61 sec latency 152.549 ms 1 1410 0.14 MB/sec execute 62 sec latency 153.417 ms 1 1459 0.14 MB/sec execute 63 sec latency 161.777 ms 1 1481 0.13 MB/sec execute 64 sec latency 759.689 ms 1 1497 0.13 MB/sec execute 65 sec latency 1233.792 ms 1 1532 0.13 MB/sec execute 66 sec latency 136.335 ms 1 1597 0.13 MB/sec execute 67 sec latency 94.382 ms test_26 fail mds2 2 times 1 1625 0.13 MB/sec execute 68 sec latency 190.121 ms 1 1656 0.13 MB/sec execute 69 sec latency 162.014 ms 1 1685 0.12 MB/sec execute 70 sec latency 132.099 ms Failing mds2 on oleg250-server 1 1706 0.12 MB/sec execute 71 sec latency 212.655 ms Stopping /mnt/lustre-mds2 (opts:) on oleg250-server 1 1757 0.12 MB/sec execute 72 sec latency 102.179 ms 1 1791 0.12 MB/sec execute 73 sec latency 145.440 ms 1 1791 0.12 MB/sec execute 74 sec latency 1040.912 ms 1 1791 0.12 MB/sec execute 75 sec latency 2041.215 ms 22:32:55 (1788489175) shut down facet: mds2 facet_host: oleg250-server facet_failover_host: oleg250-server 1 1791 0.12 MB/sec execute 76 sec latency 3043.308 ms 1 1791 0.11 MB/sec execute 77 sec latency 4043.586 ms 1 1791 0.11 MB/sec execute 78 sec latency 5043.893 ms 1 1791 0.11 MB/sec execute 79 sec latency 6044.222 ms 1 1791 0.11 MB/sec execute 80 sec latency 7046.296 ms 1 1791 0.11 MB/sec execute 81 sec latency 8046.724 ms 1 1791 0.11 MB/sec execute 82 sec latency 9047.010 ms 1 1791 0.11 MB/sec execute 83 sec latency 10047.623 ms 1 1791 0.11 MB/sec execute 84 sec latency 11048.050 ms 1 1791 0.10 MB/sec execute 85 sec latency 12049.141 ms 1 1791 0.10 MB/sec execute 86 sec latency 13049.629 ms Failover mds2 to oleg250-server mount facets: mds2 1 1791 0.10 MB/sec execute 87 sec latency 14050.075 ms 1 1791 0.10 MB/sec execute 88 sec latency 15050.316 ms 1 1791 0.10 MB/sec execute 89 sec latency 16050.856 ms 1 1791 0.10 MB/sec execute 90 sec latency 17051.432 ms 1 1791 0.10 MB/sec execute 91 sec latency 18053.310 ms 1 1791 0.10 MB/sec execute 92 sec latency 19053.485 ms 1 1791 0.10 MB/sec execute 93 sec latency 20053.718 ms 1 1791 0.09 MB/sec execute 94 sec latency 21053.924 ms 1 1791 0.09 MB/sec execute 95 sec latency 22054.572 ms 1 1791 0.09 MB/sec execute 96 sec latency 23054.863 ms Start mds2: mount -t lustre -o localrecov /dev/mapper/mds2_flakey /mnt/lustre-mds2 1 1791 0.09 MB/sec execute 97 sec latency 24056.071 ms 1 1791 0.09 MB/sec execute 98 sec latency 25057.308 ms 1 1791 0.09 MB/sec execute 99 sec latency 26061.883 ms 1 cleanup 100 sec 1 cleanup 101 sec 1 cleanup 102 sec 1 cleanup 103 sec oleg250-server: oleg250-server.virtnet: executing set_default_debug -1 all 1 cleanup 104 sec pdsh@oleg250-client: oleg250-server: ssh exited with exit code 1 1 cleanup 105 sec Started lustre-MDT0001 22:33:25 (1788489205) targets are mounted 22:33:25 (1788489205) facet_failover done 1 cleanup 106 sec 1 cleanup 107 sec 1 cleanup 108 sec 1 cleanup 109 sec 1 cleanup 110 sec 0 cleanup 110 sec Operation Count AvgLat MaxLat ---------------------------------------- NTCreateX 222 399.886 39411.870 Close 188 18.400 77.105 Rename 11 121.422 212.622 Unlink 24 35.329 67.144 Qpathinfo 173 19.461 138.576 Qfileinfo 51 1.950 30.200 Qfsinfo 33 4.679 28.117 Sfileinfo 34 80.995 156.378 Find 64 15.390 99.985 WriteX 185 16.740 93.141 ReadX 279 0.155 5.939 Flush 27 111.069 1233.754 Throughput 0.0892516 MB/sec 1 clients 1 procs max_latency=39411.891 ms stopping dbench on /mnt/lustre at Thu Sep 3 22:33:30 EDT 2026 with return code 0 clean dbench files on /mnt/lustre /mnt/lustre /mnt/lustre removed 'client.txt' /mnt/lustre dbench successfully finished oleg250-client.virtnet: executing wait_import_state_mount (FULL|IDLE) mdc.lustre-MDT0001-mdc-*.mds_server_uuid 1475 0 striped dir -i0 -c2 -H crush /mnt/lustre2/d26.replay-dual/run_dbench mdc.lustre-MDT0001-mdc-*.mds_server_uuid in FULL state after 0 sec 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 3 22:33:41 EDT 2026 waiting for dbench pid 67866 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 0 8 0.00 MB/sec warmup 1 sec latency 8.721 ms 1 51 1.05 MB/sec warmup 2 sec latency 67.233 ms 1 111 1.74 MB/sec warmup 3 sec latency 99.246 ms 1 130 1.44 MB/sec warmup 4 sec latency 255.555 ms UUID 1K-blocks Used Available Use% Mounted on lustre-MDT0000_UUID 1414116 2376 1268928 1% /mnt/lustre[MDT:0] lustre-MDT0001_UUID 1414116 2136 1269168 1% /mnt/lustre[MDT:1] lustre-OST0000_UUID 3833116 2600 3604336 1% /mnt/lustre[OST:0] lustre-OST0001_UUID 3833116 44640 3560636 2% /mnt/lustre[OST:1] filesystem_summary: 7666232 47240 7164972 1% /mnt/lustre 1 147 1.25 MB/sec warmup 5 sec latency 295.261 ms 1 173 1.17 MB/sec warmup 6 sec latency 152.246 ms 1 184 1.05 MB/sec warmup 7 sec latency 698.298 ms 1 207 1.01 MB/sec warmup 8 sec latency 753.060 ms 1 231 0.98 MB/sec warmup 9 sec latency 200.977 ms 1 260 0.98 MB/sec warmup 10 sec latency 218.629 ms 1 289 0.97 MB/sec warmup 11 sec latency 171.749 ms 1 331 1.00 MB/sec warmup 12 sec latency 142.739 ms 1 357 0.98 MB/sec warmup 13 sec latency 191.885 ms 1 389 0.98 MB/sec warmup 14 sec latency 160.000 ms 1 427 1.00 MB/sec warmup 15 sec latency 188.660 ms test_26 fail mds1 3 times 1 461 1.01 MB/sec warmup 16 sec latency 132.980 ms 1 493 1.01 MB/sec warmup 17 sec latency 139.353 ms 1 522 1.00 MB/sec warmup 18 sec latency 160.049 ms Failing mds1 on oleg250-server 1 603 1.10 MB/sec warmup 19 sec latency 143.843 ms Stopping /mnt/lustre-mds1 (opts:) on oleg250-server 1 674 0.11 MB/sec execute 1 sec latency 303.977 ms 1 674 0.06 MB/sec execute 2 sec latency 1304.182 ms 1 674 0.04 MB/sec execute 3 sec latency 2305.834 ms 22:34:04 (1788489244) shut down facet: mds1 facet_host: oleg250-server facet_failover_host: oleg250-server 1 674 0.03 MB/sec execute 4 sec latency 3306.227 ms 1 674 0.02 MB/sec execute 5 sec latency 4309.224 ms 1 674 0.02 MB/sec execute 6 sec latency 5310.420 ms 1 674 0.02 MB/sec execute 7 sec latency 6313.168 ms 1 674 0.01 MB/sec execute 8 sec latency 7314.186 ms 1 674 0.01 MB/sec execute 9 sec latency 8314.680 ms 1 674 0.01 MB/sec execute 10 sec latency 9315.042 ms 1 674 0.01 MB/sec execute 11 sec latency 10315.259 ms 1 674 0.01 MB/sec execute 12 sec latency 11315.858 ms 1 674 0.01 MB/sec execute 13 sec latency 12316.113 ms Failover mds1 to oleg250-server 1 674 0.01 MB/sec execute 14 sec latency 13316.373 ms mount facets: mds1 1 674 0.01 MB/sec execute 15 sec latency 14316.795 ms 1 674 0.01 MB/sec execute 16 sec latency 15318.107 ms 1 674 0.01 MB/sec execute 17 sec latency 16318.953 ms 1 674 0.01 MB/sec execute 18 sec latency 17319.339 ms 1 674 0.01 MB/sec execute 19 sec latency 18320.001 ms 1 674 0.01 MB/sec execute 20 sec latency 19320.201 ms 1 674 0.01 MB/sec execute 21 sec latency 20320.352 ms 1 674 0.00 MB/sec execute 22 sec latency 21322.252 ms 1 674 0.00 MB/sec execute 23 sec latency 22323.170 ms 1 674 0.00 MB/sec execute 24 sec latency 23323.391 ms 1 674 0.00 MB/sec execute 25 sec latency 24323.601 ms Start mds1: mount -t lustre -o localrecov /dev/mapper/mds1_flakey /mnt/lustre-mds1 1 674 0.00 MB/sec execute 26 sec latency 25325.036 ms 1 674 0.00 MB/sec execute 27 sec latency 26325.637 ms 1 674 0.00 MB/sec execute 28 sec latency 27326.625 ms 1 674 0.00 MB/sec execute 29 sec latency 28332.294 ms 1 674 0.00 MB/sec execute 30 sec latency 29335.055 ms 1 674 0.00 MB/sec execute 31 sec latency 30336.113 ms 1 674 0.00 MB/sec execute 32 sec latency 31339.343 ms 1 674 0.00 MB/sec execute 33 sec latency 32344.834 ms 1 674 0.00 MB/sec execute 34 sec latency 33345.132 ms oleg250-server: oleg250-server.virtnet: executing set_default_debug -1 all 1 674 0.00 MB/sec execute 35 sec latency 34346.294 ms 1 674 0.00 MB/sec execute 36 sec latency 35347.345 ms pdsh@oleg250-client: oleg250-server: ssh exited with exit code 1 1 674 0.00 MB/sec execute 37 sec latency 36348.360 ms Started lustre-MDT0000 22:34:39 (1788489279) targets are mounted 22:34:39 (1788489279) facet_failover done 1 705 0.00 MB/sec execute 38 sec latency 36389.067 ms 1 736 0.01 MB/sec execute 39 sec latency 112.704 ms 1 778 0.01 MB/sec execute 40 sec latency 142.892 ms 1 806 0.01 MB/sec execute 41 sec latency 133.316 ms 1 851 0.01 MB/sec execute 42 sec latency 149.667 ms 1 884 0.01 MB/sec execute 43 sec latency 159.077 ms 1 884 0.01 MB/sec execute 44 sec latency 1025.998 ms 1 888 0.01 MB/sec execute 45 sec latency 1747.038 ms oleg250-client.virtnet: executing wait_import_state_mount (FULL|IDLE) mdc.lustre-MDT0000-mdc-*.mds_server_uuid 1475 0 1 916 0.02 MB/sec execute 46 sec latency 148.028 ms 1 952 0.02 MB/sec execute 47 sec latency 142.999 ms mdc.lustre-MDT0000-mdc-*.mds_server_uuid in FULL state after 0 sec 1 972 0.02 MB/sec execute 48 sec latency 190.008 ms 1 993 0.02 MB/sec execute 49 sec latency 128.692 ms 1 1050 0.04 MB/sec execute 50 sec latency 81.897 ms 1 1095 0.04 MB/sec execute 51 sec latency 114.185 ms 1 1147 0.04 MB/sec execute 52 sec latency 84.703 ms 1 1193 0.04 MB/sec execute 53 sec latency 114.705 ms 1 1231 0.04 MB/sec execute 54 sec latency 107.298 ms 1 1277 0.04 MB/sec execute 55 sec latency 111.876 ms 1 1405 0.09 MB/sec execute 56 sec latency 80.596 ms 1 1451 0.09 MB/sec execute 57 sec latency 80.331 ms tar: Unexpected EOF in archive tar: Unexpected EOF in archive 1 1474 0.09 MB/sec execute 58 sec latency 415.745 ms 1 1512 0.08 MB/sec execute 59 sec latency 88.437 ms 1 1561 0.08 MB/sec execute 60 sec latency 137.585 ms 1 1591 0.08 MB/sec execute 61 sec latency 95.059 ms 1 1643 0.08 MB/sec execute 62 sec latency 84.042 ms 1 1680 0.08 MB/sec execute 63 sec latency 112.043 ms 1 1707 0.08 MB/sec execute 64 sec latency 144.245 ms 1 1762 0.08 MB/sec execute 65 sec latency 113.883 ms 1 1807 0.08 MB/sec execute 66 sec latency 142.344 ms 1 1857 0.08 MB/sec execute 67 sec latency 95.934 ms 1 1911 0.08 MB/sec execute 68 sec latency 90.904 ms tar: Error is not recoverable: exiting now 1 1973 0.08 MB/sec execute 69 sec latency 94.386 ms dbench killed by signal 15 stopping dbench on /mnt/lustre at Thu Sep 3 22:35:10 EDT 2026 with return code 0 67866 pts/0 S+ 0:00 dbench -c client.txt 1 -D /mnt/lustre2/d26.replay-dual/run_dbench -t 100 killed dbench main pid 67866 clean dbench files on /mnt/lustre /mnt/lustre /mnt/lustre removed 'client.txt' /mnt/lustre dbench successfully finished