== conf-sanity test 66: replace nids ===================== 08:04:34 (1789387474) client=34683432 MDS=34683432 OSS=34683432 start mds service on oleg247-server Start mds1: mount -t lustre -o localrecov /dev/mapper/mds1_flakey /mnt/lustre-mds1 oleg247-server: oleg247-server.virtnet: executing set_default_debug -1 all pdsh@oleg247-client: oleg247-server: ssh exited with exit code 1 Started lustre-MDT0000 oleg247-server: params: OBD_IOC_LLOG_PRINT failed: No such file or directory pdsh@oleg247-client: oleg247-server: ssh exited with exit code 2 pdsh@oleg247-client: oleg247-server: ssh exited with exit code 123 allow_register=unlimited start mds service on oleg247-server Start mds2: mount -t lustre -o localrecov /dev/mapper/mds2_flakey /mnt/lustre-mds2 oleg247-server: oleg247-server.virtnet: executing set_default_debug -1 all pdsh@oleg247-client: oleg247-server: ssh exited with exit code 1 Started lustre-MDT0001 oleg247-client.virtnet: executing wait_import_state_mount FULL mdc.lustre-MDT0000-mdc-*.mds_server_uuid 1475 0 oleg247-client.virtnet: executing wait_import_state_mount FULL mdc.lustre-MDT0001-mdc-*.mds_server_uuid 1475 0 start ost1 service on oleg247-server Start ost1: mount -t lustre -o localrecov /dev/mapper/ost1_flakey /mnt/lustre-ost1 seq.cli-lustre-OST0000-super.width=65536 oleg247-server: oleg247-server.virtnet: executing set_default_debug -1 all pdsh@oleg247-client: oleg247-server: ssh exited with exit code 1 Started lustre-OST0000 oleg247-client.virtnet: executing wait_import_state_mount (FULL|IDLE) osc.lustre-OST0000-osc-[-0-9a-f]*.ost_server_uuid 1475 0 mount lustre on /mnt/lustre..... Starting client: oleg247-client.virtnet: -o user_xattr,flock 192.168.202.147@tcp:/lustre /mnt/lustre Setting lustre-OST0000.osc.active from 1 to 0 Waiting 90s for '0' Updated after 10s: want '0' got '0' replace_nids should fail if MDS, OSTs and clients are UP oleg247-server: error: replace_nids: Operation now in progress pdsh@oleg247-client: oleg247-server: ssh exited with exit code 115 umount lustre on /mnt/lustre..... Stopping client oleg247-client.virtnet /mnt/lustre (opts:) replace_nids should fail if MDS and OSTs are UP oleg247-server: error: replace_nids: Operation now in progress pdsh@oleg247-client: oleg247-server: ssh exited with exit code 115 stop ost1 service on oleg247-server Stopping /mnt/lustre-ost1 (opts:-f) on oleg247-server replace_nids should fail if MDS is UP oleg247-server: error: replace_nids: Operation now in progress pdsh@oleg247-client: oleg247-server: ssh exited with exit code 115 stop mds service on oleg247-server Stopping /mnt/lustre-mds1 (opts:-f) on oleg247-server stop mds service on oleg247-server Stopping /mnt/lustre-mds2 (opts:-f) on oleg247-server start mds service on oleg247-server Start mds1: mount -t lustre -o localrecov -o nosvc /dev/mapper/mds1_flakey /mnt/lustre-mds1 oleg247-server: oleg247-server.virtnet: executing set_default_debug -1 all Start /dev/mapper/mds1_flakey without service Started lustre-MDT0000 pdsh@oleg247-client: oleg247-server: ssh exited with exit code 1 command should accept two parameters pdsh@oleg247-client: oleg247-server: ssh exited with exit code 4 replace primary NIDs for device (clients/servers must be unmounted) usage: replace_nids [,NID2,NID3:NID4,NID5:NID6] correct device name should be passed oleg247-server: error: replace_nids: No such device or address pdsh@oleg247-client: oleg247-server: ssh exited with exit code 6 wrong nids list should not destroy the system pdsh@oleg247-client: oleg247-server: ssh exited with exit code 4 replace primary NIDs for device (clients/servers must be unmounted) usage: replace_nids [,NID2,NID3:NID4,NID5:NID6] pdsh@oleg247-client: oleg247-server: ssh exited with exit code 4 replace primary NIDs for device (clients/servers must be unmounted) usage: replace_nids [,NID2,NID3:NID4,NID5:NID6] replace OST nid command should accept two parameters pdsh@oleg247-client: oleg247-server: ssh exited with exit code 4 replace primary NIDs for device (clients/servers must be unmounted) usage: replace_nids [,NID2,NID3:NID4,NID5:NID6] wrong nids list should not destroy the system pdsh@oleg247-client: oleg247-server: ssh exited with exit code 4 replace primary NIDs for device (clients/servers must be unmounted) usage: replace_nids [,NID2,NID3:NID4,NID5:NID6] set NIDs with failover replace MDS nid stop mds service on oleg247-server Stopping /mnt/lustre-mds1 (opts:-f) on oleg247-server stop mds service on oleg247-server start mds service on oleg247-server Start mds1: mount -t lustre -o localrecov /dev/mapper/mds1_flakey /mnt/lustre-mds1 oleg247-server: oleg247-server.virtnet: executing set_default_debug -1 all pdsh@oleg247-client: oleg247-server: ssh exited with exit code 1 Started lustre-MDT0000 pdsh@oleg247-client: oleg247-server: ssh exited with exit code 1 start mds service on oleg247-server Start mds2: mount -t lustre -o localrecov /dev/mapper/mds2_flakey /mnt/lustre-mds2 oleg247-server: oleg247-server.virtnet: executing set_default_debug -1 all pdsh@oleg247-client: oleg247-server: ssh exited with exit code 1 Started lustre-MDT0001 oleg247-client.virtnet: executing wait_import_state_mount FULL mdc.lustre-MDT0000-mdc-*.mds_server_uuid 1475 0 oleg247-client.virtnet: executing wait_import_state_mount FULL mdc.lustre-MDT0001-mdc-*.mds_server_uuid 1475 0 Setting lustre-OST0000.osc.active from 0 to 1 Waiting 90s for '1' Waiting 80s for '1' Updated after 28s: want '1' got '1' start ost1 service on oleg247-server Start ost1: mount -t lustre -o localrecov /dev/mapper/ost1_flakey /mnt/lustre-ost1 seq.cli-lustre-OST0000-super.width=65536