== replay-dual test 26: dbench and tar with mds failover ========================================================== 15:28:27 (1789586907) Starting client oleg309-client.virtnet: -o user_xattr,flock 192.168.203.109@tcp:/lustre /mnt/lustre Started clients oleg309-client.virtnet: 192.168.203.109@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 64027 Started dbench loop with 64029 striped dir -i0 -c2 -H all_char /mnt/lustre/d26.replay-dual/run_tar striped dir -i0 -c2 -H crush2 /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 2210560 2432 2206080 1% /mnt/lustre[MDT:0] lustre-MDT0001_UUID 2210688 2048 2206592 1% /mnt/lustre[MDT:1] lustre-OST0000_UUID 3771392 2048 3765248 1% /mnt/lustre[OST:0] lustre-OST0001_UUID 3771392 18432 3741696 1% /mnt/lustre[OST:1] filesystem_summary: 7542784 20480 7506944 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 Wed Sep 16 15:28:33 EDT 2026 waiting for dbench pid 64106 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 3 0.00 MB/sec warmup 1 sec latency 773.848 ms 1 17 0.10 MB/sec warmup 2 sec latency 118.302 ms 1 71 1.05 MB/sec warmup 3 sec latency 141.475 ms 1 114 1.31 MB/sec warmup 4 sec latency 164.847 ms test_26 fail mds1 1 times 1 155 1.31 MB/sec warmup 5 sec latency 134.509 ms 1 181 1.22 MB/sec warmup 6 sec latency 188.137 ms 1 199 1.12 MB/sec warmup 7 sec latency 193.846 ms Failing mds1 on oleg309-server 1 231 1.11 MB/sec warmup 8 sec latency 227.019 ms Stopping /mnt/lustre-mds1 (opts:) on oleg309-server 1 263 1.10 MB/sec warmup 9 sec latency 136.617 ms 1 290 1.07 MB/sec warmup 10 sec latency 233.061 ms 1 339 1.11 MB/sec warmup 11 sec latency 131.505 ms 15:28:45 (1789586925) shut down facet: mds1 facet_host: oleg309-server facet_failover_host: oleg309-server 1 339 1.02 MB/sec warmup 12 sec latency 1131.627 ms 1 339 0.94 MB/sec warmup 13 sec latency 2132.439 ms 1 339 0.87 MB/sec warmup 14 sec latency 3135.944 ms 1 339 0.82 MB/sec warmup 15 sec latency 4137.812 ms 1 339 0.77 MB/sec warmup 16 sec latency 5140.535 ms 1 339 0.72 MB/sec warmup 17 sec latency 6140.709 ms 1 339 0.68 MB/sec warmup 18 sec latency 7140.918 ms 1 339 0.64 MB/sec warmup 19 sec latency 8141.121 ms 1 339 0.00 MB/sec execute 1 sec latency 10141.577 ms 1 339 0.00 MB/sec execute 2 sec latency 11141.946 ms Failover mds1 to oleg309-server mount facets: mds1 1 339 0.00 MB/sec execute 3 sec latency 12142.617 ms 1 339 0.00 MB/sec execute 4 sec latency 13143.225 ms 1 339 0.00 MB/sec execute 5 sec latency 14143.422 ms Start mds1: mount -t lustre -o localrecov lustre-mdt1/mdt1 /mnt/lustre-mds1 1 339 0.00 MB/sec execute 6 sec latency 15143.643 ms 1 339 0.00 MB/sec execute 7 sec latency 16143.901 ms 1 339 0.00 MB/sec execute 8 sec latency 17144.680 ms 1 339 0.00 MB/sec execute 9 sec latency 18144.908 ms 1 339 0.00 MB/sec execute 10 sec latency 19145.103 ms 1 339 0.00 MB/sec execute 11 sec latency 20146.020 ms 1 339 0.00 MB/sec execute 12 sec latency 21149.588 ms 1 339 0.00 MB/sec execute 13 sec latency 22150.008 ms 1 339 0.00 MB/sec execute 14 sec latency 23150.411 ms 1 339 0.00 MB/sec execute 15 sec latency 24150.697 ms 1 339 0.00 MB/sec execute 16 sec latency 25160.083 ms 1 339 0.00 MB/sec execute 17 sec latency 26160.403 ms 1 339 0.00 MB/sec execute 18 sec latency 27160.973 ms 1 339 0.00 MB/sec execute 19 sec latency 28163.457 ms 1 339 0.00 MB/sec execute 20 sec latency 29166.748 ms 1 339 0.00 MB/sec execute 21 sec latency 30167.147 ms 1 339 0.00 MB/sec execute 22 sec latency 31168.186 ms 1 339 0.00 MB/sec execute 23 sec latency 32168.405 ms oleg309-server: oleg309-server.virtnet: executing set_default_debug -1 all 1 339 0.00 MB/sec execute 24 sec latency 33168.664 ms pdsh@oleg309-client: oleg309-server: ssh exited with exit code 1 1 342 0.00 MB/sec execute 25 sec latency 34021.337 ms Started lustre-MDT0000 15:29:19 (1789586959) targets are mounted 15:29:19 (1789586959) facet_failover done 1 373 0.04 MB/sec execute 26 sec latency 144.651 ms 1 421 0.10 MB/sec execute 27 sec latency 106.071 ms 1 459 0.14 MB/sec execute 28 sec latency 193.782 ms 1 497 0.17 MB/sec execute 29 sec latency 128.736 ms 1 522 0.19 MB/sec execute 30 sec latency 176.185 ms 1 600 0.28 MB/sec execute 31 sec latency 241.697 ms 1 648 0.27 MB/sec execute 32 sec latency 104.208 ms oleg309-client.virtnet: executing wait_import_state_mount (FULL|IDLE) mdc.lustre-MDT0000-mdc-*.mds_server_uuid 1475 0 1 674 0.27 MB/sec execute 33 sec latency 234.761 ms 1 700 0.26 MB/sec execute 34 sec latency 171.879 ms mdc.lustre-MDT0000-mdc-*.mds_server_uuid in FULL state after 0 sec 1 733 0.26 MB/sec execute 35 sec latency 138.635 ms 1 765 0.25 MB/sec execute 36 sec latency 124.158 ms 1 789 0.25 MB/sec execute 37 sec latency 272.664 ms 1 806 0.24 MB/sec execute 38 sec latency 372.536 ms 1 851 0.24 MB/sec execute 39 sec latency 174.463 ms 1 885 0.23 MB/sec execute 40 sec latency 143.840 ms 1 913 0.23 MB/sec execute 41 sec latency 157.934 ms UUID 1K-blocks Used Available Use% Mounted on lustre-MDT0000_UUID 2210560 2560 2205952 1% /mnt/lustre[MDT:0] lustre-MDT0001_UUID 2210688 2432 2206208 1% /mnt/lustre[MDT:1] lustre-OST0000_UUID 3771392 9216 3697664 1% /mnt/lustre[OST:0] lustre-OST0001_UUID 3771392 36864 3710976 1% /mnt/lustre[OST:1] filesystem_summary: 7542784 46080 7408640 1% /mnt/lustre 1 929 0.23 MB/sec execute 42 sec latency 236.136 ms 1 970 0.22 MB/sec execute 43 sec latency 89.094 ms 1 1009 0.23 MB/sec execute 44 sec latency 126.634 ms 1 1082 0.24 MB/sec execute 45 sec latency 76.866 ms 1 1136 0.23 MB/sec execute 46 sec latency 74.809 ms test_26 fail mds2 2 times 1 1185 0.23 MB/sec execute 47 sec latency 74.555 ms 1 1225 0.22 MB/sec execute 48 sec latency 80.579 ms Failing mds2 on oleg309-server 1 1262 0.22 MB/sec execute 49 sec latency 97.145 ms Stopping /mnt/lustre-mds2 (opts:) on oleg309-server 1 1403 0.27 MB/sec execute 50 sec latency 80.895 ms 1 1433 0.27 MB/sec execute 51 sec latency 185.196 ms 1 1433 0.26 MB/sec execute 52 sec latency 1185.445 ms 15:29:46 (1789586986) shut down facet: mds2 facet_host: oleg309-server facet_failover_host: oleg309-server 1 1433 0.26 MB/sec execute 53 sec latency 2185.640 ms 1 1433 0.25 MB/sec execute 54 sec latency 3186.350 ms 1 1433 0.25 MB/sec execute 55 sec latency 4186.799 ms 1 1433 0.25 MB/sec execute 56 sec latency 5187.425 ms 1 1433 0.24 MB/sec execute 57 sec latency 6187.663 ms 1 1433 0.24 MB/sec execute 58 sec latency 7187.845 ms 1 1433 0.23 MB/sec execute 59 sec latency 8193.251 ms 1 1433 0.23 MB/sec execute 60 sec latency 9195.748 ms 1 1433 0.23 MB/sec execute 61 sec latency 10195.933 ms 1 1433 0.22 MB/sec execute 62 sec latency 11196.396 ms Failover mds2 to oleg309-server 1 1433 0.22 MB/sec execute 63 sec latency 12197.257 ms mount facets: mds2 1 1433 0.21 MB/sec execute 64 sec latency 13197.497 ms 1 1433 0.21 MB/sec execute 65 sec latency 14197.955 ms 1 1433 0.21 MB/sec execute 66 sec latency 15207.412 ms 1 1433 0.21 MB/sec execute 67 sec latency 16207.651 ms Start mds2: mount -t lustre -o localrecov lustre-mdt2/mdt2 /mnt/lustre-mds2 1 1433 0.20 MB/sec execute 68 sec latency 17208.381 ms 1 1433 0.20 MB/sec execute 69 sec latency 18211.030 ms 1 1433 0.20 MB/sec execute 70 sec latency 19219.990 ms 1 1433 0.19 MB/sec execute 71 sec latency 20220.569 ms 1 1433 0.19 MB/sec execute 72 sec latency 21224.431 ms 1 1433 0.19 MB/sec execute 73 sec latency 22224.687 ms oleg309-server: oleg309-server.virtnet: executing set_default_debug -1 all 1 1433 0.19 MB/sec execute 74 sec latency 23225.639 ms pdsh@oleg309-client: oleg309-server: ssh exited with exit code 1 1 1433 0.18 MB/sec execute 75 sec latency 24225.810 ms Started lustre-MDT0001 15:30:09 (1789587009) targets are mounted 15:30:09 (1789587009) facet_failover done 1 1465 0.18 MB/sec execute 76 sec latency 24544.273 ms 1 1502 0.18 MB/sec execute 77 sec latency 214.323 ms 1 1558 0.18 MB/sec execute 78 sec latency 92.710 ms 1 1613 0.18 MB/sec execute 79 sec latency 112.500 ms 1 1646 0.17 MB/sec execute 80 sec latency 254.623 ms 1 1670 0.17 MB/sec execute 81 sec latency 222.949 ms 1 1696 0.17 MB/sec execute 82 sec latency 134.405 ms 1 1748 0.17 MB/sec execute 83 sec latency 121.365 ms oleg309-client.virtnet: executing wait_import_state_mount (FULL|IDLE) mdc.lustre-MDT0001-mdc-*.mds_server_uuid 1475 0 1 1785 0.17 MB/sec execute 84 sec latency 138.722 ms 1 1820 0.17 MB/sec execute 85 sec latency 192.992 ms 1 1848 0.16 MB/sec execute 86 sec latency 166.599 ms mdc.lustre-MDT0001-mdc-*.mds_server_uuid in FULL state after 0 sec 1 1884 0.16 MB/sec execute 87 sec latency 132.971 ms 1 1924 0.16 MB/sec execute 88 sec latency 261.786 ms 1 1978 0.16 MB/sec execute 89 sec latency 143.978 ms 1 2018 0.16 MB/sec execute 90 sec latency 127.899 ms 1 2080 0.16 MB/sec execute 91 sec latency 114.883 ms 1 2115 0.16 MB/sec execute 92 sec latency 156.987 ms 1 2156 0.16 MB/sec execute 93 sec latency 102.414 ms UUID 1K-blocks Used Available Use% Mounted on lustre-MDT0000_UUID 2210432 2816 2205568 1% /mnt/lustre[MDT:0] lustre-MDT0001_UUID 2210688 2688 2205952 1% /mnt/lustre[MDT:1] lustre-OST0000_UUID 3771392 24576 3735552 1% /mnt/lustre[OST:0] lustre-OST0001_UUID 3771392 39936 3727360 2% /mnt/lustre[OST:1] filesystem_summary: 7542784 64512 7462912 1% /mnt/lustre 1 2185 0.15 MB/sec execute 94 sec latency 133.898 ms 1 2228 0.15 MB/sec execute 95 sec latency 137.347 ms 1 2288 0.16 MB/sec execute 96 sec latency 104.340 ms 1 2345 0.16 MB/sec execute 97 sec latency 87.940 ms 1 2373 0.16 MB/sec execute 98 sec latency 213.312 ms 1 2407 0.16 MB/sec execute 99 sec latency 141.983 ms test_26 fail mds1 3 times 1 cleanup 100 sec 1 cleanup 101 sec 1 cleanup 102 sec Failing mds1 on oleg309-server 1 cleanup 103 sec Stopping /mnt/lustre-mds1 (opts:) on oleg309-server 1 cleanup 104 sec 1 cleanup 105 sec 1 cleanup 106 sec 1 cleanup 107 sec 1 cleanup 108 sec 15:30:42 (1789587042) shut down 1 cleanup 109 sec facet: mds1 facet_host: oleg309-server facet_failover_host: oleg309-server 1 cleanup 110 sec 1 cleanup 111 sec 1 cleanup 112 sec 1 cleanup 113 sec 1 cleanup 114 sec 1 cleanup 115 sec 1 cleanup 116 sec 1 cleanup 117 sec 1 cleanup 118 sec 1 cleanup 119 sec Failover mds1 to oleg309-server mount facets: mds1 1 cleanup 120 sec 1 cleanup 121 sec 1 cleanup 122 sec 1 cleanup 123 sec Start mds1: mount -t lustre -o localrecov lustre-mdt1/mdt1 /mnt/lustre-mds1 1 cleanup 124 sec 1 cleanup 125 sec 1 cleanup 126 sec 1 cleanup 127 sec 1 cleanup 128 sec 1 cleanup 129 sec oleg309-server: oleg309-server.virtnet: executing set_default_debug -1 all 1 cleanup 130 sec pdsh@oleg309-client: oleg309-server: ssh exited with exit code 1 1 cleanup 131 sec 1 cleanup 132 sec Started lustre-MDT0000 15:31:06 (1789587066) targets are mounted 15:31:06 (1789587066) facet_failover done 1 cleanup 133 sec 0 cleanup 133 sec Operation Count AvgLat MaxLat ---------------------------------------- NTCreateX 361 158.342 34021.321 Close 314 94.363 24544.247 Rename 14 82.314 157.912 Unlink 35 30.622 58.432 Qpathinfo 322 18.535 166.570 Qfileinfo 96 2.386 22.546 Qfsinfo 40 5.867 47.341 Sfileinfo 44 60.088 161.864 Find 109 14.579 104.195 WriteX 351 12.539 74.165 ReadX 382 1.489 74.795 Flush 33 135.677 372.518 Throughput 0.158934 MB/sec 1 clients 1 procs max_latency=34021.337 ms stopping dbench on /mnt/lustre at Wed Sep 16 15:31:07 EDT 2026 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 oleg309-client.virtnet: executing wait_import_state_mount (FULL|IDLE) mdc.lustre-MDT0000-mdc-*.mds_server_uuid 1475 0 mdc.lustre-MDT0000-mdc-*.mds_server_uuid in FULL state after 0 sec '/usr/share/dbench/client.txt' -> 'client.txt' running 'dbench 1 -D /mnt/lustre2/d26.replay-dual/run_dbench -t 100' on /mnt/lustre at Wed Sep 16 15:31:16 EDT 2026 waiting for dbench pid 67003 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 3 0.00 MB/sec warmup 1 sec latency 740.573 ms 0 10 0.00 MB/sec warmup 2 sec latency 49.088 ms 1 51 0.70 MB/sec warmup 3 sec latency 152.684 ms 1 111 1.30 MB/sec warmup 4 sec latency 169.125 ms 1 147 1.25 MB/sec warmup 5 sec latency 158.308 ms 1 181 1.22 MB/sec warmup 6 sec latency 132.361 ms UUID 1K-blocks Used Available Use% Mounted on lustre-MDT0000_UUID 2210432 3072 2205312 1% /mnt/lustre[MDT:0] lustre-MDT0001_UUID 2210560 2816 2205696 1% /mnt/lustre[MDT:1] lustre-OST0000_UUID 3771392 15360 3749888 1% /mnt/lustre[OST:0] lustre-OST0001_UUID 3771392 40960 3688448 2% /mnt/lustre[OST:1] filesystem_summary: 7542784 56320 7438336 1% /mnt/lustre 1 200 1.12 MB/sec warmup 7 sec latency 209.392 ms 1 224 1.07 MB/sec warmup 8 sec latency 192.783 ms 1 271 1.13 MB/sec warmup 9 sec latency 158.667 ms 1 315 1.15 MB/sec warmup 10 sec latency 128.820 ms 1 341 1.11 MB/sec warmup 11 sec latency 181.501 ms 1 374 1.10 MB/sec warmup 12 sec latency 205.680 ms 1 413 1.12 MB/sec warmup 13 sec latency 138.191 ms test_26 fail mds2 4 times 1 445 1.11 MB/sec warmup 14 sec latency 184.657 ms 1 478 1.11 MB/sec warmup 15 sec latency 184.885 ms Failing mds2 on oleg309-server 1 516 1.10 MB/sec warmup 16 sec latency 130.579 ms Stopping /mnt/lustre-mds2 (opts:) on oleg309-server 1 611 1.23 MB/sec warmup 17 sec latency 207.355 ms 1 669 1.17 MB/sec warmup 18 sec latency 144.694 ms 1 706 1.11 MB/sec warmup 19 sec latency 155.968 ms 1 791 0.14 MB/sec execute 1 sec latency 145.296 ms 1 851 0.13 MB/sec execute 2 sec latency 80.425 ms 1 897 0.12 MB/sec execute 3 sec latency 156.909 ms 1 926 0.11 MB/sec execute 4 sec latency 177.766 ms 1 977 0.11 MB/sec execute 5 sec latency 74.611 ms 15:31:42 (1789587102) shut down facet: mds2 facet_host: oleg309-server facet_failover_host: oleg309-server 1 995 0.10 MB/sec execute 6 sec latency 384.669 ms 1 995 0.08 MB/sec execute 7 sec latency 1385.693 ms 1 995 0.07 MB/sec execute 8 sec latency 2386.009 ms 1 995 0.07 MB/sec execute 9 sec latency 3386.934 ms 1 995 0.06 MB/sec execute 10 sec latency 4390.138 ms 1 995 0.05 MB/sec execute 11 sec latency 5391.662 ms 1 995 0.05 MB/sec execute 12 sec latency 6392.225 ms 1 995 0.05 MB/sec execute 13 sec latency 7392.361 ms 1 995 0.04 MB/sec execute 14 sec latency 8392.546 ms 1 995 0.04 MB/sec execute 15 sec latency 9392.752 ms Failover mds2 to oleg309-server mount facets: mds2 1 995 0.04 MB/sec execute 16 sec latency 10395.117 ms 1 995 0.03 MB/sec execute 17 sec latency 11396.227 ms 1 995 0.03 MB/sec execute 18 sec latency 12396.592 ms 1 995 0.03 MB/sec execute 19 sec latency 13396.811 ms Start mds2: mount -t lustre -o localrecov lustre-mdt2/mdt2 /mnt/lustre-mds2 1 995 0.03 MB/sec execute 20 sec latency 14397.496 ms 1 995 0.03 MB/sec execute 21 sec latency 15397.687 ms 1 995 0.03 MB/sec execute 22 sec latency 16398.045 ms 1 995 0.03 MB/sec execute 23 sec latency 17398.287 ms 1 995 0.02 MB/sec execute 24 sec latency 18399.488 ms oleg309-server: oleg309-server.virtnet: executing set_default_debug -1 all 1 995 0.02 MB/sec execute 25 sec latency 19400.299 ms pdsh@oleg309-client: oleg309-server: ssh exited with exit code 1 1 995 0.02 MB/sec execute 26 sec latency 20403.054 ms Started lustre-MDT0001 15:32:03 (1789587123) targets are mounted 15:32:03 (1789587123) facet_failover done 1 995 0.02 MB/sec execute 27 sec latency 21403.735 ms 1 1042 0.06 MB/sec execute 28 sec latency 21590.772 ms 1 1103 0.06 MB/sec execute 29 sec latency 78.429 ms 1 1156 0.06 MB/sec execute 30 sec latency 102.365 ms 1 1202 0.06 MB/sec execute 31 sec latency 75.967 ms 1 1244 0.06 MB/sec execute 32 sec latency 104.260 ms oleg309-client.virtnet: executing wait_import_state_mount (FULL|IDLE) mdc.lustre-MDT0001-mdc-*.mds_server_uuid 1475 0 1 1284 0.06 MB/sec execute 33 sec latency 105.879 ms 1 1406 0.14 MB/sec execute 34 sec latency 90.262 ms mdc.lustre-MDT0001-mdc-*.mds_server_uuid in FULL state after 0 sec 1 1447 0.13 MB/sec execute 35 sec latency 123.248 ms 1 1492 0.13 MB/sec execute 36 sec latency 117.236 ms 1 1550 0.13 MB/sec execute 37 sec latency 100.594 ms 1 1603 0.13 MB/sec execute 38 sec latency 83.307 ms 1 1646 0.12 MB/sec execute 39 sec latency 201.344 ms 1 1696 0.12 MB/sec execute 40 sec latency 87.476 ms 1 1770 0.12 MB/sec execute 41 sec latency 73.152 ms tar: Unexpected EOF in archive tar: Unexpected EOF in archive 1 1837 0.12 MB/sec execute 42 sec latency 62.839 ms 1 1886 0.12 MB/sec execute 43 sec latency 199.200 ms 1 1965 0.12 MB/sec execute 44 sec latency 78.350 ms 1 2012 0.12 MB/sec execute 45 sec latency 106.638 ms 1 2083 0.12 MB/sec execute 46 sec latency 122.149 ms 1 2150 0.11 MB/sec execute 47 sec latency 101.879 ms 1 2185 0.11 MB/sec execute 48 sec latency 179.725 ms 1 2284 0.12 MB/sec execute 49 sec latency 85.673 ms 1 2347 0.13 MB/sec execute 50 sec latency 206.113 ms 1 2389 0.13 MB/sec execute 51 sec latency 85.735 ms tar: Error is not recoverable: exiting now 1 2428 0.13 MB/sec execute 52 sec latency 163.470 ms dbench killed by signal 15 stopping dbench on /mnt/lustre at Wed Sep 16 15:32:29 EDT 2026 with return code 0 67003 pts/0 S+ 0:00 dbench -c client.txt 1 -D /mnt/lustre2/d26.replay-dual/run_dbench -t 100 killed dbench main pid 67003 clean dbench files on /mnt/lustre /mnt/lustre /mnt/lustre removed 'client.txt' /mnt/lustre dbench successfully finished