== sanity-lnet test 205: Check health and resends for multi-rail local failures ========================================================== 10:15:06 (1749478506) Cleaning up LNet LNET unconfigure error 22: Invalid argument unloading modules via unload_modules_local on: 'oleg104-server' oleg104-server: oleg104-server.virtnet: executing unload_modules_local oleg104-server: LNET unconfigure error 22: Invalid argument modules unloaded. Loading modules from /home/green/git/lustre-release/lustre detected 4 online CPUs by sysfs MODOPTS_LIBCFS= Force libcfs to create 2 CPU partitions ../libcfs/libcfs/libcfs options: 'cpu_npartitions=2' /home/green/git/lustre-release/lustre/../lnet/utils/lnetctl lnet configure -a oleg104-server: Writer error: failed to resolve Netlink family id oleg104-server: opening /dev/lnet failed: No such file or directory oleg104-server: hint: the kernel modules may not be loaded oleg104-server: IOC_LIBCFS_GET_NI error 2: No such file or directory pdsh@oleg104-client: oleg104-server: ssh exited with exit code 1 oleg104-server: oleg104-server.virtnet: executing load_lnet oleg104-server: Loading modules from /home/green/git/lustre-release/lustre oleg104-server: detected 4 online CPUs by sysfs oleg104-server: MODOPTS_LIBCFS= oleg104-server: Force libcfs to create 2 CPU partitions /home/green/git/lustre-release/lustre/../lnet/utils/lnetctl discover 192.168.201.104@tcp discover: - primary nid: 192.168.201.104@tcp Multi-Rail: true peer_ni: - nid: 192.168.201.104@tcp oleg104-server: oleg104-server.virtnet: executing lnet_if_list oleg104-server: oleg104-server.virtnet: executing /home/green/git/lustre-release/lustre/../lnet/utils/lnetctl net add --net tcp1 --if eth0 /home/green/git/lustre-release/lustre/../lnet/utils/lnetctl lnet configure /home/green/git/lustre-release/lustre/../lnet/utils/lnetctl net add --net tcp1 --if eth0 default via 192.168.201.254 dev eth0 192.168.201.0/24 dev eth0 proto kernel scope link src 192.168.201.4 default via 192.168.201.254 dev eth0 192.168.201.0/24 dev eth0 proto kernel scope link src 192.168.201.4 net: - net type: lo local NI(s): - nid: 0@lo status: up - net type: tcp local NI(s): - nid: 192.168.201.4@tcp status: up interfaces: 0: eth0 - net type: tcp1 local NI(s): - nid: 192.168.201.4@tcp1 status: up interfaces: 0: eth0 - primary nid: 192.168.201.104@tcp - nid: 192.168.201.104@tcp health stats: health value: 1000 - nid: 192.168.201.104@tcp1 health stats: health value: 1000 debug=+net Simulate local_interrupt /home/green/git/lustre-release/lustre/utils/lctl net_drop add -s 192.168.201.4@tcp -d 192.168.201.104@tcp -m GET -r 1 -e local_interrupt /home/green/git/lustre-release/lustre/utils/lctl net_drop add -s 192.168.201.4@tcp -d 192.168.201.104@tcp1 -m GET -r 1 -e local_interrupt /home/green/git/lustre-release/lustre/utils/lctl net_drop add -s 192.168.201.4@tcp -d 192.168.201.4@tcp -m GET -r 1 -e local_interrupt /home/green/git/lustre-release/lustre/../lnet/utils/lnetctl fault drop add -s 192.168.201.4@tcp -d 192.168.201.4@tcp1 -m GET -r 1 -e local_interrupt /home/green/git/lustre-release/lustre/../lnet/utils/lnetctl fault drop add -s 192.168.201.4@tcp1 -d 192.168.201.104@tcp -m GET -r 1 -e local_interrupt /home/green/git/lustre-release/lustre/utils/lctl net_drop add -s 192.168.201.4@tcp1 -d 192.168.201.104@tcp1 -m GET -r 1 -e local_interrupt /home/green/git/lustre-release/lustre/../lnet/utils/lnetctl fault drop add -s 192.168.201.4@tcp1 -d 192.168.201.4@tcp -m GET -r 1 -e local_interrupt /home/green/git/lustre-release/lustre/utils/lctl net_drop add -s 192.168.201.4@tcp1 -d 192.168.201.4@tcp1 -m GET -r 1 -e local_interrupt /home/green/git/lustre-release/lustre/../lnet/utils/lnetctl ping 192.168.201.104@tcp manage: - ping: errno: -5 descr: ! 'failed to ping 192.168.201.104@tcp: Input/output error' /home/green/git/lustre-release/lustre/../lnet/utils/lnetctl peer set --health 1000 --all /home/green/git/lustre-release/lustre/../lnet/utils/lnetctl net set --health 1000 --all /home/green/git/lustre-release/lustre/../lnet/utils/lnetctl fault drop del -a Check that 2 resends took place Check that local NI health has been changed Simulate local_dropped /home/green/git/lustre-release/lustre/../lnet/utils/lnetctl fault drop add -s 192.168.201.4@tcp -d 192.168.201.104@tcp -m GET -r 1 -e local_dropped /home/green/git/lustre-release/lustre/utils/lctl net_drop add -s 192.168.201.4@tcp -d 192.168.201.104@tcp1 -m GET -r 1 -e local_dropped /home/green/git/lustre-release/lustre/utils/lctl net_drop add -s 192.168.201.4@tcp -d 192.168.201.4@tcp -m GET -r 1 -e local_dropped /home/green/git/lustre-release/lustre/../lnet/utils/lnetctl fault drop add -s 192.168.201.4@tcp -d 192.168.201.4@tcp1 -m GET -r 1 -e local_dropped /home/green/git/lustre-release/lustre/../lnet/utils/lnetctl fault drop add -s 192.168.201.4@tcp1 -d 192.168.201.104@tcp -m GET -r 1 -e local_dropped /home/green/git/lustre-release/lustre/../lnet/utils/lnetctl fault drop add -s 192.168.201.4@tcp1 -d 192.168.201.104@tcp1 -m GET -r 1 -e local_dropped /home/green/git/lustre-release/lustre/utils/lctl net_drop add -s 192.168.201.4@tcp1 -d 192.168.201.4@tcp -m GET -r 1 -e local_dropped /home/green/git/lustre-release/lustre/../lnet/utils/lnetctl fault drop add -s 192.168.201.4@tcp1 -d 192.168.201.4@tcp1 -m GET -r 1 -e local_dropped /home/green/git/lustre-release/lustre/../lnet/utils/lnetctl ping 192.168.201.104@tcp manage: - ping: errno: -5 descr: ! 'failed to ping 192.168.201.104@tcp: Input/output error' /home/green/git/lustre-release/lustre/../lnet/utils/lnetctl peer set --health 1000 --all /home/green/git/lustre-release/lustre/../lnet/utils/lnetctl net set --health 1000 --all /home/green/git/lustre-release/lustre/../lnet/utils/lnetctl fault drop del -a Check that 2 resends took place Check that local NI health has been changed Simulate local_aborted /home/green/git/lustre-release/lustre/../lnet/utils/lnetctl fault drop add -s 192.168.201.4@tcp -d 192.168.201.104@tcp -m GET -r 1 -e local_aborted /home/green/git/lustre-release/lustre/../lnet/utils/lnetctl fault drop add -s 192.168.201.4@tcp -d 192.168.201.104@tcp1 -m GET -r 1 -e local_aborted /home/green/git/lustre-release/lustre/../lnet/utils/lnetctl fault drop add -s 192.168.201.4@tcp -d 192.168.201.4@tcp -m GET -r 1 -e local_aborted /home/green/git/lustre-release/lustre/utils/lctl net_drop add -s 192.168.201.4@tcp -d 192.168.201.4@tcp1 -m GET -r 1 -e local_aborted /home/green/git/lustre-release/lustre/../lnet/utils/lnetctl fault drop add -s 192.168.201.4@tcp1 -d 192.168.201.104@tcp -m GET -r 1 -e local_aborted /home/green/git/lustre-release/lustre/../lnet/utils/lnetctl fault drop add -s 192.168.201.4@tcp1 -d 192.168.201.104@tcp1 -m GET -r 1 -e local_aborted /home/green/git/lustre-release/lustre/utils/lctl net_drop add -s 192.168.201.4@tcp1 -d 192.168.201.4@tcp -m GET -r 1 -e local_aborted /home/green/git/lustre-release/lustre/utils/lctl net_drop add -s 192.168.201.4@tcp1 -d 192.168.201.4@tcp1 -m GET -r 1 -e local_aborted /home/green/git/lustre-release/lustre/../lnet/utils/lnetctl ping 192.168.201.104@tcp manage: - ping: errno: -5 descr: ! 'failed to ping 192.168.201.104@tcp: Input/output error' /home/green/git/lustre-release/lustre/../lnet/utils/lnetctl peer set --health 1000 --all /home/green/git/lustre-release/lustre/../lnet/utils/lnetctl net set --health 1000 --all /home/green/git/lustre-release/lustre/../lnet/utils/lnetctl fault drop del -a Check that 2 resends took place Check that local NI health has been changed Simulate local_no_route /home/green/git/lustre-release/lustre/../lnet/utils/lnetctl fault drop add -s 192.168.201.4@tcp -d 192.168.201.104@tcp -m GET -r 1 -e local_no_route /home/green/git/lustre-release/lustre/../lnet/utils/lnetctl fault drop add -s 192.168.201.4@tcp -d 192.168.201.104@tcp1 -m GET -r 1 -e local_no_route /home/green/git/lustre-release/lustre/utils/lctl net_drop add -s 192.168.201.4@tcp -d 192.168.201.4@tcp -m GET -r 1 -e local_no_route /home/green/git/lustre-release/lustre/../lnet/utils/lnetctl fault drop add -s 192.168.201.4@tcp -d 192.168.201.4@tcp1 -m GET -r 1 -e local_no_route /home/green/git/lustre-release/lustre/../lnet/utils/lnetctl fault drop add -s 192.168.201.4@tcp1 -d 192.168.201.104@tcp -m GET -r 1 -e local_no_route /home/green/git/lustre-release/lustre/../lnet/utils/lnetctl fault drop add -s 192.168.201.4@tcp1 -d 192.168.201.104@tcp1 -m GET -r 1 -e local_no_route /home/green/git/lustre-release/lustre/../lnet/utils/lnetctl fault drop add -s 192.168.201.4@tcp1 -d 192.168.201.4@tcp -m GET -r 1 -e local_no_route /home/green/git/lustre-release/lustre/utils/lctl net_drop add -s 192.168.201.4@tcp1 -d 192.168.201.4@tcp1 -m GET -r 1 -e local_no_route /home/green/git/lustre-release/lustre/../lnet/utils/lnetctl ping 192.168.201.104@tcp manage: - ping: errno: -5 descr: ! 'failed to ping 192.168.201.104@tcp: Input/output error' /home/green/git/lustre-release/lustre/../lnet/utils/lnetctl peer set --health 1000 --all /home/green/git/lustre-release/lustre/../lnet/utils/lnetctl net set --health 1000 --all /home/green/git/lustre-release/lustre/utils/lctl net_drop del -a Check that 2 resends took place Check that local NI health has been changed Simulate local_timeout /home/green/git/lustre-release/lustre/../lnet/utils/lnetctl fault drop add -s 192.168.201.4@tcp -d 192.168.201.104@tcp -m GET -r 1 -e local_timeout /home/green/git/lustre-release/lustre/utils/lctl net_drop add -s 192.168.201.4@tcp -d 192.168.201.104@tcp1 -m GET -r 1 -e local_timeout /home/green/git/lustre-release/lustre/../lnet/utils/lnetctl fault drop add -s 192.168.201.4@tcp -d 192.168.201.4@tcp -m GET -r 1 -e local_timeout /home/green/git/lustre-release/lustre/../lnet/utils/lnetctl fault drop add -s 192.168.201.4@tcp -d 192.168.201.4@tcp1 -m GET -r 1 -e local_timeout /home/green/git/lustre-release/lustre/../lnet/utils/lnetctl fault drop add -s 192.168.201.4@tcp1 -d 192.168.201.104@tcp -m GET -r 1 -e local_timeout /home/green/git/lustre-release/lustre/utils/lctl net_drop add -s 192.168.201.4@tcp1 -d 192.168.201.104@tcp1 -m GET -r 1 -e local_timeout /home/green/git/lustre-release/lustre/utils/lctl net_drop add -s 192.168.201.4@tcp1 -d 192.168.201.4@tcp -m GET -r 1 -e local_timeout /home/green/git/lustre-release/lustre/../lnet/utils/lnetctl fault drop add -s 192.168.201.4@tcp1 -d 192.168.201.4@tcp1 -m GET -r 1 -e local_timeout /home/green/git/lustre-release/lustre/../lnet/utils/lnetctl ping 192.168.201.104@tcp manage: - ping: errno: -5 descr: ! 'failed to ping 192.168.201.104@tcp: Input/output error' /home/green/git/lustre-release/lustre/../lnet/utils/lnetctl peer set --health 1000 --all /home/green/git/lustre-release/lustre/../lnet/utils/lnetctl net set --health 1000 --all /home/green/git/lustre-release/lustre/utils/lctl net_drop del -a Check that 2 resends took place Check that local NI health has been changed Simulate local_error /home/green/git/lustre-release/lustre/utils/lctl net_drop add -s 192.168.201.4@tcp -d 192.168.201.104@tcp -m GET -r 1 -e local_error /home/green/git/lustre-release/lustre/utils/lctl net_drop add -s 192.168.201.4@tcp -d 192.168.201.104@tcp1 -m GET -r 1 -e local_error /home/green/git/lustre-release/lustre/utils/lctl net_drop add -s 192.168.201.4@tcp -d 192.168.201.4@tcp -m GET -r 1 -e local_error /home/green/git/lustre-release/lustre/../lnet/utils/lnetctl fault drop add -s 192.168.201.4@tcp -d 192.168.201.4@tcp1 -m GET -r 1 -e local_error /home/green/git/lustre-release/lustre/utils/lctl net_drop add -s 192.168.201.4@tcp1 -d 192.168.201.104@tcp -m GET -r 1 -e local_error /home/green/git/lustre-release/lustre/../lnet/utils/lnetctl fault drop add -s 192.168.201.4@tcp1 -d 192.168.201.104@tcp1 -m GET -r 1 -e local_error /home/green/git/lustre-release/lustre/../lnet/utils/lnetctl fault drop add -s 192.168.201.4@tcp1 -d 192.168.201.4@tcp -m GET -r 1 -e local_error /home/green/git/lustre-release/lustre/utils/lctl net_drop add -s 192.168.201.4@tcp1 -d 192.168.201.4@tcp1 -m GET -r 1 -e local_error /home/green/git/lustre-release/lustre/../lnet/utils/lnetctl ping 192.168.201.104@tcp manage: - ping: errno: -5 descr: ! 'failed to ping 192.168.201.104@tcp: Input/output error' /home/green/git/lustre-release/lustre/../lnet/utils/lnetctl peer set --health 1000 --all /home/green/git/lustre-release/lustre/../lnet/utils/lnetctl net set --health 1000 --all /home/green/git/lustre-release/lustre/../lnet/utils/lnetctl fault drop del -a Check that no resends took place Check that local NI health has been changed oleg104-server: oleg104-server.virtnet: executing /home/green/git/lustre-release/lustre/../lnet/utils/lnetctl net del --net tcp1 --if eth0 Writer error: failed to resolve Netlink family id unloading modules via unload_modules_local on: 'oleg104-server' oleg104-server: oleg104-server.virtnet: executing unload_modules_local modules unloaded. oleg104-server: oleg104-server.virtnet: executing unload_modules_local oleg104-server: LNET unconfigure error 22: Invalid argument pdsh@oleg104-client: oleg104-client: ssh exited with exit code 2 pdsh@oleg104-client: oleg104-server: ssh exited with exit code 2 pdsh@oleg104-client: oleg104-client: ssh exited with exit code 2 pdsh@oleg104-client: oleg104-server: ssh exited with exit code 2