== replay-dual test 26: dbench and tar with mds failover ========================================================== 06:56:27 (1787396187) Starting client oleg633-client.virtnet: -o user_xattr,flock 192.168.206.133@tcp:/lustre /mnt/lustre Started clients oleg633-client.virtnet: 192.168.206.133@tcp:/lustre on /mnt/lustre type lustre (rw,flock,checksum,encrypt,lazystatfs,lruresize,nolock,statfs_project,nouser_fid2path,user_xattr,verbose) Started tar loop with pid 44902 Started dbench loop with 44903 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 2210560 2048 2206464 1% /mnt/lustre[MDT:0] lustre-OST0000_UUID 3771392 2048 3760128 1% /mnt/lustre[OST:0] lustre-OST0001_UUID 3771392 18432 3726336 1% /mnt/lustre[OST:1] filesystem_summary: 7542784 20480 7486464 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 Sat Aug 22 06:56:33 EDT 2026 waiting for dbench pid 44970 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 14 0.00 MB/sec warmup 1 sec latency 46.613 ms 1 66 1.44 MB/sec warmup 2 sec latency 113.307 ms 1 111 1.75 MB/sec warmup 3 sec latency 186.743 ms 1 147 1.57 MB/sec warmup 4 sec latency 204.792 ms 1 176 1.41 MB/sec warmup 5 sec latency 167.830 ms test_26 fail mds1 1 times 1 200 1.30 MB/sec warmup 6 sec latency 204.837 ms 1 231 1.27 MB/sec warmup 7 sec latency 204.183 ms Failing mds1 on oleg633-server 1 263 1.24 MB/sec warmup 8 sec latency 158.193 ms Stopping /mnt/lustre-mds1 (opts:) on oleg633-server 1 289 1.19 MB/sec warmup 9 sec latency 161.116 ms 1 322 1.17 MB/sec warmup 10 sec latency 137.965 ms 1 322 1.07 MB/sec warmup 11 sec latency 1040.172 ms 1 322 0.98 MB/sec warmup 12 sec latency 2040.385 ms 06:56:45 (1787396205) shut down facet: mds1 facet_host: oleg633-server facet_failover_host: oleg633-server 1 322 0.90 MB/sec warmup 13 sec latency 3040.644 ms 1 322 0.84 MB/sec warmup 14 sec latency 4041.012 ms 1 322 0.78 MB/sec warmup 15 sec latency 5043.113 ms 1 322 0.73 MB/sec warmup 16 sec latency 6044.055 ms 1 322 0.69 MB/sec warmup 17 sec latency 7045.254 ms 1 322 0.65 MB/sec warmup 18 sec latency 8046.701 ms 1 322 0.62 MB/sec warmup 19 sec latency 9048.163 ms 1 322 0.00 MB/sec execute 1 sec latency 11048.646 ms 1 322 0.00 MB/sec execute 2 sec latency 12048.928 ms 1 322 0.00 MB/sec execute 3 sec latency 13049.076 ms Failover mds1 to oleg633-server mount facets: mds1 1 322 0.00 MB/sec execute 4 sec latency 14049.283 ms 1 322 0.00 MB/sec execute 5 sec latency 15050.511 ms 1 322 0.00 MB/sec execute 6 sec latency 16050.778 ms Start mds1: mount -t lustre -o localrecov lustre-mdt1/mdt1 /mnt/lustre-mds1 1 322 0.00 MB/sec execute 7 sec latency 17050.916 ms 1 322 0.00 MB/sec execute 8 sec latency 18051.167 ms 1 322 0.00 MB/sec execute 9 sec latency 19056.795 ms 1 322 0.00 MB/sec execute 10 sec latency 20057.021 ms 1 322 0.00 MB/sec execute 11 sec latency 21057.294 ms 1 322 0.00 MB/sec execute 12 sec latency 22057.717 ms oleg633-server: oleg633-server.virtnet: executing set_default_debug -1 all 1 322 0.00 MB/sec execute 13 sec latency 23064.820 ms pdsh@oleg633-client: oleg633-server: ssh exited with exit code 1 1 339 0.04 MB/sec execute 14 sec latency 23346.905 ms Started lustre-MDT0000 06:57:08 (1787396228) targets are mounted 06:57:08 (1787396228) facet_failover done 1 373 0.10 MB/sec execute 15 sec latency 146.474 ms 1 414 0.18 MB/sec execute 16 sec latency 130.287 ms 1 448 0.23 MB/sec execute 17 sec latency 186.488 ms 1 477 0.27 MB/sec execute 18 sec latency 171.634 ms 1 503 0.30 MB/sec execute 19 sec latency 224.062 ms 1 536 0.35 MB/sec execute 20 sec latency 184.834 ms 1 606 0.44 MB/sec execute 21 sec latency 149.072 ms 1 651 0.42 MB/sec execute 22 sec latency 115.857 ms oleg633-client.virtnet: executing wait_import_state_mount (FULL|IDLE) mdc.lustre-MDT0000-mdc-*.mds_server_uuid 1 674 0.41 MB/sec execute 23 sec latency 292.184 ms 1 704 0.39 MB/sec execute 24 sec latency 140.083 ms mdc.lustre-MDT0000-mdc-*.mds_server_uuid in FULL state after 0 sec 1 726 0.38 MB/sec execute 25 sec latency 230.025 ms 1 758 0.37 MB/sec execute 26 sec latency 114.348 ms 1 786 0.36 MB/sec execute 27 sec latency 155.214 ms 1 806 0.35 MB/sec execute 28 sec latency 250.215 ms 1 862 0.34 MB/sec execute 29 sec latency 99.811 ms 1 887 0.33 MB/sec execute 30 sec latency 211.818 ms 1 916 0.32 MB/sec execute 31 sec latency 221.924 ms UUID 1K-blocks Used Available Use% Mounted on lustre-MDT0000_UUID 2210560 2304 2206208 1% /mnt/lustre[MDT:0] lustre-OST0000_UUID 3771392 19456 3723264 1% /mnt/lustre[OST:0] lustre-OST0001_UUID 3771392 36864 3700736 1% /mnt/lustre[OST:1] filesystem_summary: 7542784 56320 7424000 1% /mnt/lustre 1 948 0.31 MB/sec execute 32 sec latency 119.536 ms 1 970 0.30 MB/sec execute 33 sec latency 148.830 ms 1 990 0.30 MB/sec execute 34 sec latency 262.512 ms 1 1051 0.32 MB/sec execute 35 sec latency 100.579 ms 1 1086 0.31 MB/sec execute 36 sec latency 102.597 ms 1 1138 0.30 MB/sec execute 37 sec latency 76.448 ms test_26 fail mds1 2 times 1 1181 0.30 MB/sec execute 38 sec latency 105.844 ms 1 1219 0.29 MB/sec execute 39 sec latency 87.498 ms Failing mds1 on oleg633-server 1 1244 0.28 MB/sec execute 40 sec latency 132.545 ms 1 1274 0.28 MB/sec execute 41 sec latency 185.400 ms Stopping /mnt/lustre-mds1 (opts:) on oleg633-server 1 1405 0.34 MB/sec execute 42 sec latency 118.185 ms 1 1409 0.33 MB/sec execute 43 sec latency 877.741 ms 1 1409 0.32 MB/sec execute 44 sec latency 1878.447 ms 06:57:38 (1787396258) shut down facet: mds1 facet_host: oleg633-server facet_failover_host: oleg633-server 1 1409 0.32 MB/sec execute 45 sec latency 2880.940 ms 1 1409 0.31 MB/sec execute 46 sec latency 3881.180 ms 1 1409 0.30 MB/sec execute 47 sec latency 4881.449 ms 1 1409 0.30 MB/sec execute 48 sec latency 5883.798 ms 1 1409 0.29 MB/sec execute 49 sec latency 6884.436 ms 1 1409 0.28 MB/sec execute 50 sec latency 7885.157 ms 1 1409 0.28 MB/sec execute 51 sec latency 8885.428 ms 1 1409 0.27 MB/sec execute 52 sec latency 9885.776 ms 1 1409 0.27 MB/sec execute 53 sec latency 10886.294 ms 1 1409 0.26 MB/sec execute 54 sec latency 11886.706 ms 1 1409 0.26 MB/sec execute 55 sec latency 12887.027 ms Failover mds1 to oleg633-server mount facets: mds1 1 1409 0.25 MB/sec execute 56 sec latency 13888.151 ms 1 1409 0.25 MB/sec execute 57 sec latency 14888.408 ms 1 1409 0.25 MB/sec execute 58 sec latency 15889.638 ms 1 1409 0.24 MB/sec execute 59 sec latency 16891.017 ms Start mds1: mount -t lustre -o localrecov lustre-mdt1/mdt1 /mnt/lustre-mds1 1 1409 0.24 MB/sec execute 60 sec latency 17894.065 ms 1 1409 0.23 MB/sec execute 61 sec latency 18894.451 ms 1 1409 0.23 MB/sec execute 62 sec latency 19896.765 ms 1 1409 0.23 MB/sec execute 63 sec latency 20897.164 ms 1 1409 0.22 MB/sec execute 64 sec latency 21897.412 ms 1 1409 0.22 MB/sec execute 65 sec latency 22897.756 ms oleg633-server: oleg633-server.virtnet: executing set_default_debug -1 all 1 1447 0.22 MB/sec execute 66 sec latency 22960.797 ms 1 1481 0.21 MB/sec execute 67 sec latency 232.750 ms pdsh@oleg633-client: oleg633-server: ssh exited with exit code 1 1 1518 0.21 MB/sec execute 68 sec latency 287.491 ms Started lustre-MDT0000 06:58:01 (1787396281) targets are mounted 06:58:01 (1787396281) facet_failover done 1 1568 0.21 MB/sec execute 69 sec latency 135.231 ms 1 1616 0.21 MB/sec execute 70 sec latency 100.868 ms 1 1646 0.20 MB/sec execute 71 sec latency 294.562 ms 1 1674 0.20 MB/sec execute 72 sec latency 162.472 ms 1 1702 0.20 MB/sec execute 73 sec latency 141.495 ms 1 1757 0.20 MB/sec execute 74 sec latency 127.935 ms 1 1797 0.19 MB/sec execute 75 sec latency 112.607 ms 1 1830 0.19 MB/sec execute 76 sec latency 135.467 ms 1 1861 0.19 MB/sec execute 77 sec latency 163.352 ms oleg633-client.virtnet: executing wait_import_state_mount (FULL|IDLE) mdc.lustre-MDT0000-mdc-*.mds_server_uuid 1 1888 0.19 MB/sec execute 78 sec latency 253.850 ms 1 1942 0.19 MB/sec execute 79 sec latency 153.839 ms mdc.lustre-MDT0000-mdc-*.mds_server_uuid in FULL state after 0 sec 1 1984 0.18 MB/sec execute 80 sec latency 90.002 ms 1 2016 0.18 MB/sec execute 81 sec latency 117.725 ms 1 2076 0.18 MB/sec execute 82 sec latency 105.120 ms 1 2133 0.18 MB/sec execute 83 sec latency 111.445 ms 1 2160 0.18 MB/sec execute 84 sec latency 129.378 ms 1 2204 0.18 MB/sec execute 85 sec latency 125.167 ms 1 2287 0.18 MB/sec execute 86 sec latency 76.731 ms UUID 1K-blocks Used Available Use% Mounted on lustre-MDT0000_UUID 2210560 2688 2205824 1% /mnt/lustre[MDT:0] lustre-OST0000_UUID 3771392 32768 3724288 1% /mnt/lustre[OST:0] lustre-OST0001_UUID 3771392 40960 3716096 2% /mnt/lustre[OST:1] filesystem_summary: 7542784 73728 7440384 1% /mnt/lustre 1 2345 0.19 MB/sec execute 87 sec latency 90.110 ms 1 2367 0.18 MB/sec execute 88 sec latency 336.309 ms 1 2394 0.18 MB/sec execute 89 sec latency 287.926 ms 1 2426 0.18 MB/sec execute 90 sec latency 150.081 ms 1 2466 0.19 MB/sec execute 91 sec latency 112.130 ms 1 2497 0.18 MB/sec execute 92 sec latency 270.435 ms test_26 fail mds1 3 times 1 2566 0.18 MB/sec execute 93 sec latency 344.153 ms 1 2567 0.18 MB/sec execute 94 sec latency 1325.251 ms Failing mds1 on oleg633-server 1 2608 0.18 MB/sec execute 95 sec latency 90.939 ms Stopping /mnt/lustre-mds1 (opts:) on oleg633-server 1 2759 0.21 MB/sec execute 96 sec latency 94.718 ms 1 2776 0.21 MB/sec execute 97 sec latency 431.051 ms 1 2776 0.21 MB/sec execute 98 sec latency 1431.525 ms 1 2776 0.20 MB/sec execute 99 sec latency 2432.031 ms 06:58:32 (1787396312) shut down facet: mds1 facet_host: oleg633-server facet_failover_host: oleg633-server 1 cleanup 100 sec 1 cleanup 101 sec 1 cleanup 102 sec 1 cleanup 103 sec 1 cleanup 104 sec 1 cleanup 105 sec 1 cleanup 106 sec 1 cleanup 107 sec 1 cleanup 108 sec 1 cleanup 109 sec Failover mds1 to oleg633-server mount facets: mds1 1 cleanup 110 sec 1 cleanup 111 sec 1 cleanup 112 sec 1 cleanup 113 sec Start mds1: mount -t lustre -o localrecov lustre-mdt1/mdt1 /mnt/lustre-mds1 1 cleanup 114 sec 1 cleanup 115 sec 1 cleanup 116 sec 1 cleanup 117 sec 1 cleanup 118 sec 1 cleanup 119 sec oleg633-server: oleg633-server.virtnet: executing set_default_debug -1 all 1 cleanup 120 sec 1 cleanup 121 sec pdsh@oleg633-client: oleg633-server: ssh exited with exit code 1 1 cleanup 122 sec Started lustre-MDT0000 06:58:55 (1787396335) targets are mounted 06:58:55 (1787396335) facet_failover done 1 cleanup 123 sec 1 cleanup 124 sec 1 cleanup 125 sec 1 cleanup 126 sec 1 cleanup 127 sec 1 cleanup 128 sec 1 cleanup 129 sec 1 cleanup 130 sec 1 cleanup 131 sec 1 cleanup 132 sec 1 cleanup 133 sec 1 cleanup 134 sec 1 cleanup 135 sec 1 cleanup 136 sec 1 cleanup 137 sec 1 cleanup 138 sec 0 cleanup 139 sec Operation Count AvgLat MaxLat ---------------------------------------- NTCreateX 396 164.751 38733.038 Close 341 150.231 23346.879 Rename 18 96.225 185.380 Unlink 41 35.167 78.464 Qpathinfo 360 19.881 110.442 Qfileinfo 109 3.206 14.441 Qfsinfo 52 4.039 47.666 Sfileinfo 50 63.579 102.577 Find 126 7.500 95.849 WriteX 387 15.296 91.573 ReadX 540 1.823 85.722 Flush 35 164.973 336.289 Throughput 0.2039 MB/sec 1 clients 1 procs max_latency=23346.905 ms stopping dbench on /mnt/lustre at Sat Aug 22 06:59:12 EDT 2026 with return code 0 clean dbench files on /mnt/lustre /mnt/lustre /mnt/lustre removed 'client.txt' /mnt/lustre dbench successfully finished oleg633-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 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 Sat Aug 22 06:59:19 EDT 2026 waiting for dbench pid 47893 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 31 1.05 MB/sec warmup 1 sec latency 130.520 ms 1 95 2.13 MB/sec warmup 2 sec latency 139.528 ms 1 137 2.01 MB/sec warmup 3 sec latency 158.516 ms UUID 1K-blocks Used Available Use% Mounted on lustre-MDT0000_UUID 2210560 3072 2205440 1% /mnt/lustre[MDT:0] lustre-OST0000_UUID 3771392 23552 3736576 1% /mnt/lustre[OST:0] lustre-OST0001_UUID 3771392 28672 3706880 1% /mnt/lustre[OST:1] filesystem_summary: 7542784 52224 7443456 1% /mnt/lustre 1 166 1.70 MB/sec warmup 4 sec latency 156.912 ms 1 200 1.57 MB/sec warmup 5 sec latency 144.385 ms 1 255 1.61 MB/sec warmup 6 sec latency 87.950 ms 1 290 1.53 MB/sec warmup 7 sec latency 143.075 ms 1 339 1.53 MB/sec warmup 8 sec latency 107.271 ms 1 389 1.53 MB/sec warmup 9 sec latency 115.676 ms test_26 fail mds1 4 times 1 424 1.48 MB/sec warmup 10 sec latency 138.953 ms 1 461 1.47 MB/sec warmup 11 sec latency 142.462 ms Failing mds1 on oleg633-server 1 499 1.45 MB/sec warmup 12 sec latency 131.926 ms Stopping /mnt/lustre-mds1 (opts:) on oleg633-server 1 538 1.45 MB/sec warmup 13 sec latency 124.794 ms 1 541 1.36 MB/sec warmup 14 sec latency 972.042 ms 1 541 1.27 MB/sec warmup 15 sec latency 1972.438 ms 06:59:35 (1787396375) shut down facet: mds1 facet_host: oleg633-server facet_failover_host: oleg633-server 1 541 1.19 MB/sec warmup 16 sec latency 2973.175 ms 1 541 1.12 MB/sec warmup 17 sec latency 3974.533 ms 1 541 1.06 MB/sec warmup 18 sec latency 4975.517 ms 1 541 1.00 MB/sec warmup 19 sec latency 5976.422 ms 1 541 0.00 MB/sec execute 1 sec latency 7976.955 ms 1 541 0.00 MB/sec execute 2 sec latency 8977.974 ms 1 541 0.00 MB/sec execute 3 sec latency 9978.735 ms 1 541 0.00 MB/sec execute 4 sec latency 10979.000 ms 1 541 0.00 MB/sec execute 5 sec latency 11979.175 ms Failover mds1 to oleg633-server 1 541 0.00 MB/sec execute 6 sec latency 12979.470 ms mount facets: mds1 1 541 0.00 MB/sec execute 7 sec latency 13985.903 ms 1 541 0.00 MB/sec execute 8 sec latency 14986.191 ms 1 541 0.00 MB/sec execute 9 sec latency 15986.516 ms 1 541 0.00 MB/sec execute 10 sec latency 16986.676 ms Start mds1: mount -t lustre -o localrecov lustre-mdt1/mdt1 /mnt/lustre-mds1 1 541 0.00 MB/sec execute 11 sec latency 17986.886 ms 1 541 0.00 MB/sec execute 12 sec latency 18987.081 ms 1 541 0.00 MB/sec execute 13 sec latency 19988.539 ms 1 541 0.00 MB/sec execute 14 sec latency 20988.996 ms 1 541 0.00 MB/sec execute 15 sec latency 21989.625 ms 1 541 0.00 MB/sec execute 16 sec latency 22993.020 ms 1 541 0.00 MB/sec execute 17 sec latency 23993.502 ms 1 541 0.00 MB/sec execute 18 sec latency 24993.723 ms 1 541 0.00 MB/sec execute 19 sec latency 25994.073 ms 1 541 0.00 MB/sec execute 20 sec latency 26994.272 ms 1 541 0.00 MB/sec execute 21 sec latency 27994.996 ms 1 541 0.00 MB/sec execute 22 sec latency 28997.845 ms 1 541 0.00 MB/sec execute 23 sec latency 29998.005 ms 1 541 0.00 MB/sec execute 24 sec latency 30998.924 ms 1 541 0.00 MB/sec execute 25 sec latency 32006.368 ms oleg633-server: oleg633-server.virtnet: executing set_default_debug -1 all 1 541 0.00 MB/sec execute 26 sec latency 33006.587 ms pdsh@oleg633-client: oleg633-server: ssh exited with exit code 1 1 541 0.00 MB/sec execute 27 sec latency 34006.841 ms Started lustre-MDT0000 07:00:07 (1787396407) targets are mounted 07:00:07 (1787396407) facet_failover done 1 541 0.00 MB/sec execute 28 sec latency 35008.800 ms 1 541 0.00 MB/sec execute 29 sec latency 36009.034 ms 1 541 0.00 MB/sec execute 30 sec latency 37009.349 ms 1 541 0.00 MB/sec execute 31 sec latency 38009.563 ms 1 541 0.00 MB/sec execute 32 sec latency 39010.752 ms 1 541 0.00 MB/sec execute 33 sec latency 40016.899 ms 1 551 0.00 MB/sec execute 34 sec latency 40750.219 ms 1 636 0.06 MB/sec execute 35 sec latency 135.566 ms 1 664 0.06 MB/sec execute 36 sec latency 156.606 ms 1 678 0.06 MB/sec execute 37 sec latency 468.404 ms 1 710 0.06 MB/sec execute 38 sec latency 154.416 ms 1 731 0.06 MB/sec execute 39 sec latency 235.186 ms oleg633-client.virtnet: executing wait_import_state_mount (FULL|IDLE) mdc.lustre-MDT0000-mdc-*.mds_server_uuid 1 765 0.06 MB/sec execute 40 sec latency 125.106 ms 1 789 0.06 MB/sec execute 41 sec latency 133.904 ms mdc.lustre-MDT0000-mdc-*.mds_server_uuid in FULL state after 0 sec 1 803 0.06 MB/sec execute 42 sec latency 317.971 ms 1 851 0.06 MB/sec execute 43 sec latency 106.174 ms 1 882 0.06 MB/sec execute 44 sec latency 211.467 ms 1 913 0.06 MB/sec execute 45 sec latency 131.892 ms 1 939 0.06 MB/sec execute 46 sec latency 212.578 ms 1 977 0.06 MB/sec execute 47 sec latency 113.690 ms 1 1002 0.06 MB/sec execute 48 sec latency 139.435 ms 1 1077 0.08 MB/sec execute 49 sec latency 131.700 ms 1 1136 0.08 MB/sec execute 50 sec latency 90.165 ms 1 1190 0.08 MB/sec execute 51 sec latency 87.943 ms 1 1235 0.08 MB/sec execute 52 sec latency 69.390 ms 1 1268 0.08 MB/sec execute 53 sec latency 137.014 ms 1 1403 0.13 MB/sec execute 54 sec latency 138.630 ms 1 1461 0.13 MB/sec execute 55 sec latency 90.997 ms 1 1499 0.12 MB/sec execute 56 sec latency 333.370 ms 1 1547 0.12 MB/sec execute 57 sec latency 115.794 ms 1 1579 0.12 MB/sec execute 58 sec latency 93.522 ms dbench killed by signal 15 stopping dbench on /mnt/lustre at Sat Aug 22 07:00:38 EDT 2026 with return code 0 1 1632 0.12 MB/sec execute 59 sec latency 81.630 ms 47893 pts/0 S+ 0:00 dbench -c client.txt 1 -D /mnt/lustre2/d26.replay-dual/run_dbench -t 100 killed dbench main pid 47893 clean dbench files on /mnt/lustre /mnt/lustre /mnt/lustre removed 'client.txt' /mnt/lustre dbench successfully finished