== replay-dual test 26: dbench and tar with mds failover ========================================================== 11:11:21 (1780326681) Starting client oleg303-client.virtnet: -o user_xattr,flock 192.168.203.103@tcp:/lustre /mnt/lustre Started clients oleg303-client.virtnet: 192.168.203.103@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 64746 Started dbench loop with 64747 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 '/usr/share/dbench/client.txt' -> 'client.txt' running 'dbench 1 -D /mnt/lustre2/d26.replay-dual/run_dbench -t 100' on /mnt/lustre at Mon Jun 1 11:11:22 EDT 2026 waiting for dbench pid 64789 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 1 147 6.28 MB/sec warmup 1 sec latency 49.427 ms UUID 1K-blocks Used Available Use% Mounted on lustre-MDT0000_UUID 1414116 2000 1269304 1% /mnt/lustre[MDT:0] lustre-MDT0001_UUID 1414116 1792 1269512 1% /mnt/lustre[MDT:1] lustre-OST0000_UUID 3833116 28188 3573840 1% /mnt/lustre[OST:0] lustre-OST0001_UUID 3833116 4616 3600964 1% /mnt/lustre[OST:1] filesystem_summary: 7666232 32804 7174804 1% /mnt/lustre 1 279 5.22 MB/sec warmup 2 sec latency 163.205 ms 1 453 5.30 MB/sec warmup 3 sec latency 26.774 ms 1 653 5.27 MB/sec warmup 4 sec latency 40.731 ms 1 803 4.30 MB/sec warmup 5 sec latency 29.579 ms test_26 fail mds1 1 times Failing mds1 on oleg303-server 1 884 3.61 MB/sec warmup 6 sec latency 506.643 ms Stopping /mnt/lustre-mds1 (opts:) on oleg303-server 1 990 3.13 MB/sec warmup 7 sec latency 859.583 ms 1 995 2.74 MB/sec warmup 8 sec latency 976.021 ms 1 995 2.43 MB/sec warmup 9 sec latency 1976.183 ms 1 995 2.19 MB/sec warmup 10 sec latency 2976.336 ms 1 995 1.99 MB/sec warmup 11 sec latency 3976.524 ms 1 995 1.83 MB/sec warmup 12 sec latency 4976.667 ms 1 995 1.69 MB/sec warmup 13 sec latency 5976.811 ms 11:11:36 (1780326696) shut down facet: mds1 facet_host: oleg303-server facet_failover_host: oleg303-server 1 995 1.56 MB/sec warmup 14 sec latency 6976.988 ms 1 995 1.46 MB/sec warmup 15 sec latency 7977.131 ms 1 995 1.37 MB/sec warmup 16 sec latency 8977.286 ms 1 995 1.29 MB/sec warmup 17 sec latency 9977.448 ms 1 995 1.22 MB/sec warmup 18 sec latency 10977.640 ms 1 995 1.15 MB/sec warmup 19 sec latency 11977.808 ms 1 995 0.00 MB/sec execute 1 sec latency 13978.065 ms 1 995 0.00 MB/sec execute 2 sec latency 14978.243 ms 1 995 0.00 MB/sec execute 3 sec latency 15978.368 ms Failover mds1 to oleg303-server mount facets: mds1 1 995 0.00 MB/sec execute 4 sec latency 16978.559 ms 1 995 0.00 MB/sec execute 5 sec latency 17978.720 ms 1 995 0.00 MB/sec execute 6 sec latency 18978.857 ms Start mds1: mount -t lustre -o localrecov /dev/mapper/mds1_flakey /mnt/lustre-mds1 1 995 0.00 MB/sec execute 7 sec latency 19979.009 ms oleg303-server: oleg303-server.virtnet: executing set_default_debug -1 all pdsh@oleg303-client: oleg303-server: ssh exited with exit code 1 1 995 0.00 MB/sec execute 8 sec latency 20979.116 ms Started lustre-MDT0000 11:11:51 (1780326711) targets are mounted 11:11:51 (1780326711) facet_failover done 1 995 0.00 MB/sec execute 9 sec latency 21979.288 ms 1 995 0.00 MB/sec execute 10 sec latency 22979.450 ms 1 995 0.00 MB/sec execute 11 sec latency 23979.690 ms 1 995 0.00 MB/sec execute 12 sec latency 24979.900 ms 1 1062 0.08 MB/sec execute 13 sec latency 25522.976 ms oleg303-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 1244 0.08 MB/sec execute 14 sec latency 23.197 ms 1 1486 0.28 MB/sec execute 15 sec latency 123.101 ms 1 1686 0.27 MB/sec execute 16 sec latency 35.392 ms 1 1849 0.26 MB/sec execute 17 sec latency 57.966 ms UUID 1K-blocks Used Available Use% Mounted on lustre-MDT0000_UUID 1414116 2420 1268884 1% /mnt/lustre[MDT:0] lustre-MDT0001_UUID 1414116 2124 1269180 1% /mnt/lustre[MDT:1] lustre-OST0000_UUID 3833116 49576 3555788 2% /mnt/lustre[OST:0] lustre-OST0001_UUID 3833116 5848 3591528 1% /mnt/lustre[OST:1] filesystem_summary: 7666232 55424 7147316 1% /mnt/lustre 1 2008 0.26 MB/sec execute 18 sec latency 39.650 ms 1 2181 0.25 MB/sec execute 19 sec latency 22.777 ms 1 2394 0.30 MB/sec execute 20 sec latency 33.497 ms 1 2582 0.34 MB/sec execute 21 sec latency 30.332 ms 1 2969 0.53 MB/sec execute 22 sec latency 28.385 ms test_26 fail mds2 2 times Failing mds2 on oleg303-server 1 3326 0.58 MB/sec execute 23 sec latency 33.871 ms Stopping /mnt/lustre-mds2 (opts:) on oleg303-server 1 3641 0.70 MB/sec execute 24 sec latency 28.123 ms 1 3885 0.71 MB/sec execute 25 sec latency 20.880 ms 1 4083 0.70 MB/sec execute 26 sec latency 90.178 ms 1 4083 0.67 MB/sec execute 27 sec latency 1090.328 ms 1 4083 0.65 MB/sec execute 28 sec latency 2090.452 ms 1 4083 0.63 MB/sec execute 29 sec latency 3090.611 ms 1 4083 0.61 MB/sec execute 30 sec latency 4090.758 ms 11:12:13 (1780326733) shut down facet: mds2 facet_host: oleg303-server facet_failover_host: oleg303-server 1 4083 0.59 MB/sec execute 31 sec latency 5090.909 ms 1 4083 0.57 MB/sec execute 32 sec latency 6091.128 ms 1 4083 0.55 MB/sec execute 33 sec latency 7091.336 ms 1 4083 0.53 MB/sec execute 34 sec latency 8091.508 ms 1 4083 0.52 MB/sec execute 35 sec latency 9091.715 ms 1 4083 0.50 MB/sec execute 36 sec latency 10091.929 ms 1 4083 0.49 MB/sec execute 37 sec latency 11092.108 ms 1 4083 0.48 MB/sec execute 38 sec latency 12092.259 ms 1 4083 0.47 MB/sec execute 39 sec latency 13092.423 ms 1 4083 0.45 MB/sec execute 40 sec latency 14092.615 ms Failover mds2 to oleg303-server mount facets: mds2 1 4083 0.44 MB/sec execute 41 sec latency 15092.757 ms 1 4083 0.43 MB/sec execute 42 sec latency 16092.909 ms Start mds2: mount -t lustre -o localrecov /dev/mapper/mds2_flakey /mnt/lustre-mds2 1 4083 0.42 MB/sec execute 43 sec latency 17093.004 ms 1 4083 0.41 MB/sec execute 44 sec latency 18093.148 ms oleg303-server: oleg303-server.virtnet: executing set_default_debug -1 all pdsh@oleg303-client: oleg303-server: ssh exited with exit code 1 1 4083 0.40 MB/sec execute 45 sec latency 19093.269 ms Started lustre-MDT0001 11:12:28 (1780326748) targets are mounted 11:12:28 (1780326748) facet_failover done 1 4083 0.40 MB/sec execute 46 sec latency 20093.409 ms 1 4083 0.39 MB/sec execute 47 sec latency 21093.563 ms 1 4083 0.38 MB/sec execute 48 sec latency 22093.685 ms 1 4083 0.37 MB/sec execute 49 sec latency 23093.819 ms 1 4221 0.37 MB/sec execute 50 sec latency 23436.727 ms oleg303-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 4347 0.36 MB/sec execute 51 sec latency 63.355 ms 1 4488 0.36 MB/sec execute 52 sec latency 56.440 ms 1 4642 0.38 MB/sec execute 53 sec latency 42.346 ms 1 4799 0.38 MB/sec execute 54 sec latency 27.214 ms 1 5012 0.42 MB/sec execute 55 sec latency 38.588 ms UUID 1K-blocks Used Available Use% Mounted on lustre-MDT0000_UUID 1414116 2992 1268312 1% /mnt/lustre[MDT:0] lustre-MDT0001_UUID 1414116 2768 1268536 1% /mnt/lustre[MDT:1] lustre-OST0000_UUID 3833116 51244 3554888 2% /mnt/lustre[OST:0] lustre-OST0001_UUID 3833116 19108 3587056 1% /mnt/lustre[OST:1] filesystem_summary: 7666232 70352 7141944 1% /mnt/lustre 1 5127 0.42 MB/sec execute 56 sec latency 47.233 ms 1 5218 0.41 MB/sec execute 57 sec latency 45.709 ms 1 5303 0.40 MB/sec execute 58 sec latency 67.757 ms 1 5390 0.40 MB/sec execute 59 sec latency 65.153 ms 1 5489 0.39 MB/sec execute 60 sec latency 54.764 ms test_26 fail mds1 3 times 1 5583 0.39 MB/sec execute 61 sec latency 47.076 ms Failing mds1 on oleg303-server Stopping /mnt/lustre-mds1 (opts:) on oleg303-server 1 5679 0.39 MB/sec execute 62 sec latency 87.159 ms 1 5728 0.38 MB/sec execute 63 sec latency 179.616 ms 11:12:47 (1780326767) shut down 1 5820 0.37 MB/sec execute 64 sec latency 139.036 ms facet: mds1 facet_host: oleg303-server facet_failover_host: oleg303-server 1 5820 0.37 MB/sec execute 65 sec latency 1031.437 ms 1 5820 0.36 MB/sec execute 66 sec latency 2031.609 ms 1 5820 0.36 MB/sec execute 67 sec latency 3031.955 ms 1 5820 0.35 MB/sec execute 68 sec latency 4032.196 ms 1 5820 0.35 MB/sec execute 69 sec latency 5163.828 ms 1 5820 0.34 MB/sec execute 70 sec latency 6163.991 ms 1 5820 0.34 MB/sec execute 71 sec latency 7164.382 ms 1 5820 0.33 MB/sec execute 72 sec latency 8165.644 ms 1 5820 0.33 MB/sec execute 73 sec latency 9165.909 ms 1 5820 0.32 MB/sec execute 74 sec latency 10172.892 ms Failover mds1 to oleg303-server mount facets: mds1 1 5820 0.32 MB/sec execute 75 sec latency 11173.083 ms 1 5820 0.32 MB/sec execute 76 sec latency 12173.313 ms 1 5820 0.31 MB/sec execute 77 sec latency 13173.509 ms 1 5820 0.31 MB/sec execute 78 sec latency 14173.663 ms 1 5820 0.30 MB/sec execute 79 sec latency 15173.924 ms Start mds1: mount -t lustre -o localrecov /dev/mapper/mds1_flakey /mnt/lustre-mds1 1 5820 0.30 MB/sec execute 80 sec latency 16174.334 ms 1 5820 0.30 MB/sec execute 81 sec latency 17174.518 ms 1 5820 0.29 MB/sec execute 82 sec latency 18174.702 ms 1 5820 0.29 MB/sec execute 83 sec latency 19175.103 ms oleg303-server: oleg303-server.virtnet: executing set_default_debug -1 all 1 5820 0.29 MB/sec execute 84 sec latency 20175.345 ms pdsh@oleg303-client: oleg303-server: ssh exited with exit code 1 Started lustre-MDT0000 11:13:07 (1780326787) targets are mounted 11:13:07 (1780326787) facet_failover done 1 5820 0.28 MB/sec execute 85 sec latency 21175.882 ms 1 5820 0.28 MB/sec execute 86 sec latency 22176.599 ms 1 5823 0.28 MB/sec execute 87 sec latency 23179.636 ms 1 5919 0.28 MB/sec execute 88 sec latency 65.522 ms oleg303-client.virtnet: executing wait_import_state_mount (FULL|IDLE) mdc.lustre-MDT0000-mdc-*.mds_server_uuid 1 5976 0.28 MB/sec execute 89 sec latency 76.993 ms mdc.lustre-MDT0000-mdc-*.mds_server_uuid in FULL state after 0 sec 1 6027 0.29 MB/sec execute 90 sec latency 143.782 ms 1 6130 0.29 MB/sec execute 91 sec latency 98.704 ms 1 6276 0.32 MB/sec execute 92 sec latency 69.880 ms 1 6456 0.32 MB/sec execute 93 sec latency 107.266 ms 1 6555 0.33 MB/sec execute 94 sec latency 70.718 ms UUID 1K-blocks Used Available Use% Mounted on lustre-MDT0000_UUID 1414116 3316 1267988 1% /mnt/lustre[MDT:0] lustre-MDT0001_UUID 1414116 3116 1268188 1% /mnt/lustre[MDT:1] lustre-OST0000_UUID 3833116 52116 3552176 2% /mnt/lustre[OST:0] lustre-OST0001_UUID 3833116 21320 3584772 1% /mnt/lustre[OST:1] filesystem_summary: 7666232 73436 7136948 2% /mnt/lustre 1 6787 0.33 MB/sec execute 95 sec latency 79.363 ms 1 6922 0.34 MB/sec execute 96 sec latency 71.876 ms 1 7128 0.37 MB/sec execute 97 sec latency 45.795 ms 1 7261 0.37 MB/sec execute 98 sec latency 47.984 ms 1 7388 0.37 MB/sec execute 99 sec latency 35.975 ms 1 cleanup 100 sec test_26 fail mds2 4 times Failing mds2 on oleg303-server 1 cleanup 101 sec Stopping /mnt/lustre-mds2 (opts:) on oleg303-server 0 cleanup 102 sec Operation Count AvgLat MaxLat ---------------------------------------- NTCreateX 1096 62.986 25522.966 Close 820 3.682 139.015 Rename 48 32.180 104.819 Unlink 199 8.997 35.777 Qpathinfo 1040 28.087 23436.713 Qfileinfo 185 1.137 7.300 Qfsinfo 184 0.570 8.704 Sfileinfo 82 21.068 98.694 Find 389 4.602 19.400 WriteX 558 4.990 41.278 ReadX 1836 0.202 43.589 LockX 4 8.022 18.094 UnlockX 4 6.051 9.936 Flush 62 20.926 107.252 Throughput 0.3728 MB/sec 1 clients 1 procs max_latency=25522.976 ms stopping dbench on /mnt/lustre at Mon Jun 1 11:13:24 EDT 2026 with return code 0 clean dbench files on /mnt/lustre /mnt/lustre /mnt/lustre removed 'client.txt' /mnt/lustre dbench successfully finished 11:13:32 (1780326812) shut down facet: mds2 facet_host: oleg303-server facet_failover_host: oleg303-server Failover mds2 to oleg303-server mount facets: mds2 Start mds2: mount -t lustre -o localrecov /dev/mapper/mds2_flakey /mnt/lustre-mds2 oleg303-server: oleg303-server.virtnet: executing set_default_debug -1 all pdsh@oleg303-client: oleg303-server: ssh exited with exit code 1 Started lustre-MDT0001 11:13:59 (1780326839) targets are mounted 11:13:59 (1780326839) facet_failover done striped dir -i0 -c2 -H crush /mnt/lustre2/d26.replay-dual/run_dbench oleg303-client.virtnet: executing wait_import_state_mount (FULL|IDLE) mdc.lustre-MDT0001-mdc-*.mds_server_uuid looking for dbench program /usr/bin/dbench found dbench client file /usr/share/dbench/client.txt mdc.lustre-MDT0001-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 Mon Jun 1 11:14:06 EDT 2026 waiting for dbench pid 68828 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 104.104 ms 1 61 1.27 MB/sec warmup 2 sec latency 128.777 ms 1 121 1.83 MB/sec warmup 3 sec latency 150.683 ms 1 157 1.63 MB/sec warmup 4 sec latency 141.947 ms 1 188 1.51 MB/sec warmup 5 sec latency 367.381 ms UUID 1K-blocks Used Available Use% Mounted on lustre-MDT0000_UUID 1414116 3700 1267604 1% /mnt/lustre[MDT:0] lustre-MDT0001_UUID 1414116 3580 1267724 1% /mnt/lustre[MDT:1] lustre-OST0000_UUID 3833116 13408 3591740 1% /mnt/lustre[OST:0] lustre-OST0001_UUID 3833116 39520 3561600 2% /mnt/lustre[OST:1] filesystem_summary: 7666232 52928 7153340 1% /mnt/lustre 1 208 1.35 MB/sec warmup 6 sec latency 267.204 ms 1 246 1.34 MB/sec warmup 7 sec latency 256.709 ms 1 287 1.34 MB/sec warmup 8 sec latency 131.277 ms 1 323 1.30 MB/sec warmup 9 sec latency 134.678 ms 1 352 1.25 MB/sec warmup 10 sec latency 300.908 ms 1 397 1.28 MB/sec warmup 11 sec latency 124.564 ms 1 445 1.30 MB/sec warmup 12 sec latency 146.460 ms 1 478 1.28 MB/sec warmup 13 sec latency 216.153 ms 1 513 1.26 MB/sec warmup 14 sec latency 251.138 ms 1 638 1.40 MB/sec warmup 15 sec latency 86.070 ms test_26 fail mds1 5 times 1 695 1.32 MB/sec warmup 16 sec latency 95.991 ms Failing mds1 on oleg303-server 1 740 1.25 MB/sec warmup 17 sec latency 81.674 ms Stopping /mnt/lustre-mds1 (opts:) on oleg303-server 1 799 1.19 MB/sec warmup 18 sec latency 76.691 ms 1 803 1.13 MB/sec warmup 19 sec latency 916.303 ms 11:14:26 (1780326866) shut down facet: mds1 facet_host: oleg303-server facet_failover_host: oleg303-server 1 803 0.00 MB/sec execute 1 sec latency 2916.945 ms 1 803 0.00 MB/sec execute 2 sec latency 3917.517 ms 1 803 0.00 MB/sec execute 3 sec latency 4917.805 ms 1 803 0.00 MB/sec execute 4 sec latency 5918.007 ms 1 803 0.00 MB/sec execute 5 sec latency 6918.217 ms 1 803 0.00 MB/sec execute 6 sec latency 7918.378 ms 1 803 0.00 MB/sec execute 7 sec latency 8921.283 ms 1 803 0.00 MB/sec execute 8 sec latency 9921.538 ms 1 803 0.00 MB/sec execute 9 sec latency 10921.767 ms 1 803 0.00 MB/sec execute 10 sec latency 11921.918 ms Failover mds1 to oleg303-server mount facets: mds1 1 803 0.00 MB/sec execute 11 sec latency 12922.130 ms 1 803 0.00 MB/sec execute 12 sec latency 13922.355 ms 1 803 0.00 MB/sec execute 13 sec latency 14923.013 ms 1 803 0.00 MB/sec execute 14 sec latency 15923.273 ms 1 803 0.00 MB/sec execute 15 sec latency 16925.134 ms 1 803 0.00 MB/sec execute 16 sec latency 17925.554 ms Start mds1: mount -t lustre -o localrecov /dev/mapper/mds1_flakey /mnt/lustre-mds1 1 803 0.00 MB/sec execute 17 sec latency 18927.181 ms 1 803 0.00 MB/sec execute 18 sec latency 19927.849 ms 1 803 0.00 MB/sec execute 19 sec latency 20928.207 ms oleg303-server: oleg303-server.virtnet: executing set_default_debug -1 all 1 803 0.00 MB/sec execute 20 sec latency 21928.736 ms pdsh@oleg303-client: oleg303-server: ssh exited with exit code 1 1 803 0.00 MB/sec execute 21 sec latency 22929.018 ms Started lustre-MDT0000 11:14:48 (1780326888) targets are mounted 11:14:48 (1780326888) facet_failover done 1 803 0.00 MB/sec execute 22 sec latency 23929.653 ms 1 803 0.00 MB/sec execute 23 sec latency 24929.868 ms 1 803 0.00 MB/sec execute 24 sec latency 25930.017 ms 1 873 0.01 MB/sec execute 25 sec latency 26187.284 ms oleg303-client.virtnet: executing wait_import_state_mount (FULL|IDLE) mdc.lustre-MDT0000-mdc-*.mds_server_uuid 1 888 0.01 MB/sec execute 26 sec latency 596.098 ms mdc.lustre-MDT0000-mdc-*.mds_server_uuid in FULL state after 0 sec 1 926 0.01 MB/sec execute 27 sec latency 88.777 ms 1 981 0.02 MB/sec execute 28 sec latency 85.753 ms 1 1076 0.05 MB/sec execute 29 sec latency 66.577 ms kill: sending signal to 64775 failed: No such process 1 1147 0.05 MB/sec execute 30 sec latency 56.888 ms 1 1221 0.05 MB/sec execute 31 sec latency 57.989 ms 1 1277 0.05 MB/sec execute 32 sec latency 61.537 ms 1 1435 0.14 MB/sec execute 33 sec latency 77.276 ms dbench killed by signal 15 stopping dbench on /mnt/lustre at Mon Jun 1 11:15:00 EDT 2026 with return code 0 68828 pts/0 S+ 0:00 dbench -c client.txt 1 -D /mnt/lustre2/d26.replay-dual/run_dbench -t 100 killed dbench main pid 68828 clean dbench files on /mnt/lustre /mnt/lustre /mnt/lustre removed 'client.txt' /mnt/lustre dbench successfully finished