== sanity-lnet test 258: TCP reset cancels an outstanding discovery Ping ========================================================== 11:55:24 (1788969324) Cleaning up LNet LNET unconfigure error 22: Invalid argument unloading modules via unload_modules_local on: 'oleg159-server' oleg159-server: oleg159-server.virtnet: executing unload_modules_local oleg159-server: oleg159-server: LNET unconfigure error 22: Invalid argument modules unloaded. /home/green/git/lustre-release/lustre/tests/test-framework.sh: line 1048: echo: write error: Device or resource busy 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 oleg159-server: Writer error: failed to resolve Netlink family id oleg159-server: opening /dev/lnet failed: No such file or directory oleg159-server: hint: the kernel modules may not be loaded oleg159-server: IOC_LIBCFS_GET_NI error 2: No such file or directory pdsh@oleg159-client: oleg159-server: ssh exited with exit code 1 oleg159-server: oleg159-server.virtnet: executing load_lnet oleg159-server: /home/green/git/lustre-release/lustre/tests/test-framework.sh: line 1048: echo: write error: Device or resource busy oleg159-server: Loading modules from /home/green/git/lustre-release/lustre oleg159-server: detected 4 online CPUs by sysfs oleg159-server: MODOPTS_LIBCFS= oleg159-server: Force libcfs to create 2 CPU partitions /home/green/git/lustre-release/lustre/../lnet/utils/lnetctl discover 192.168.201.159@tcp discover: - primary nid: 192.168.201.159@tcp Multi-Rail: true peer_ni: - nid: 192.168.201.159@tcp net: - net type: lo local NI(s): - nid: 0@lo status: up - net type: tcp local NI(s): - nid: 192.168.201.59@tcp status: up interfaces: 0: ens2 - primary nid: 192.168.201.159@tcp - nid: 192.168.201.159@tcp health stats: health value: 1000 debug=+net 258: client oleg159-client.virtnet (192.168.201.59@tcp), server oleg159-server /home/green/git/lustre-release/lustre/../lnet/utils/lnetctl ping 192.168.201.159@tcp ping: - primary nid: 192.168.201.159@tcp Multi-Rail: true peer_ni: - nid: 192.168.201.159@tcp 258: initial ping complete; cleared client debug buffer 258: remove stale remote GET drop rule pdsh@oleg159-client: oleg159-server: ssh exited with exit code 2 failed to delete drop rule: Object not found 258: install remote GET drop rule for forced discovery 258: start forced discovery (log /tmp/sanity-lnet-258-discover.log) 258: forced discovery pid 12028; deadline in 180s 258: client sent the forced discovery GET 258: client peer state after GET drop (PING_SENT=1) peer: - primary nid: 192.168.201.159@tcp Multi-Rail: true peer state: 16977 peer ni: - nid: 192.168.201.159@tcp udsp info: net priority: -1 nid priority: -1 state: NA max_ni_tx_credits: 8 available_tx_credits: 8 min_tx_credits: 7 tx_q_num_of_buf: 0 available_rtr_credits: 8 min_rtr_credits: 8 refcount: 2 statistics: send_count: 5 recv_count: 4 drop_count: 0 put_egress_samples: 1 put_egress_min_nsec: 325083 put_egress_max_nsec: 325083 put_egress_avg_nsec: 325083 get_rtt_samples: 2 get_rtt_min_nsec: 8228223 get_rtt_max_nsec: 44704105 get_rtt_avg_nsec: 26466164 put_ack_samples: 1 put_ack_min_nsec: 17240112 put_ack_max_nsec: 17240112 put_ack_avg_nsec: 17240112 sent_stats: put: 2 get: 3 reply: 0 ack: 0 hello: 0 received_stats: put: 1 get: 0 reply: 2 ack: 1 hello: 0 dropped_stats: put: 0 get: 0 reply: 0 ack: 0 hello: 0 health stats: health value: 1000 dropped: 0 timeout: 0 error: 0 network timeout: 0 ping_count: 0 next_ping: 0 258: forced discovery remains active 258: server SockLND sockets before ss -K 0 0 192.168.201.159:988 192.168.201.59:1022 timer:(keepalive,9.780ms,0) ino:70409 sk:1 <-> 0 0 192.168.201.159:988 192.168.201.59:1023 timer:(keepalive,9.766ms,0) ino:70405 sk:1001 <-> 0 0 192.168.201.159:988 192.168.201.59:1021 timer:(keepalive,9.792ms,0) ino:71498 sk:1002 <-> 258: ss -K rc 0; output follows 258: invoking ss -K 258: server SockLND sockets after ss -K 0 0 192.168.201.159:988 192.168.201.59:1022 timer:(keepalive,3.488ms,0) ino:70409 sk:1 <-> 0 0 192.168.201.159:988 192.168.201.59:1023 timer:(keepalive,3.474ms,0) ino:70405 sk:1001 <-> 0 0 192.168.201.159:988 192.168.201.59:1021 timer:(keepalive,3.500ms,0) ino:71498 sk:1002 <-> 258: every selected server socket identity survived 258: stop forced discovery pid 12028 258: remove remote GET drop rule for 192.168.201.59@tcp Writer error: failed to resolve Netlink family id unloading modules via unload_modules_local on: 'oleg159-server' oleg159-server: oleg159-server.virtnet: executing unload_modules_local oleg159-server: modules unloaded. oleg159-server: oleg159-server.virtnet: executing unload_modules_local oleg159-server: oleg159-server: LNET unconfigure error 22: Invalid argument SKIP: sanity-lnet test_258 ss -K did not destroy selected SockLND socket for 192.168.201.159@tcp Cleaning up LNet LNET unconfigure error 22: Invalid argument unloading modules via unload_modules_local on: 'oleg159-server' oleg159-server: oleg159-server.virtnet: executing unload_modules_local oleg159-server: oleg159-server: LNET unconfigure error 22: Invalid argument modules unloaded. Checking servers environments Checking clients oleg159-client.virtnet environments /home/green/git/lustre-release/lustre/tests/test-framework.sh: line 1048: echo: write error: Device or resource busy 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' ptlrpc/ptlrpc options: 'lbug_on_grant_miscount=1' quota/lquota options: 'hash_lqs_cur_bits=3' mdt/mdt options: 'mdt_enable_flr_ec=1' ln: failed to create symbolic link '/sbin/.libs': Read-only file system loading modules on: 'oleg159-server' oleg159-server: oleg159-server.virtnet: executing load_modules_local oleg159-server: /home/green/git/lustre-release/lustre/tests/test-framework.sh: line 1048: echo: write error: Device or resource busy oleg159-server: Loading modules from /home/green/git/lustre-release/lustre oleg159-server: detected 4 online CPUs by sysfs oleg159-server: MODOPTS_LIBCFS= oleg159-server: Force libcfs to create 2 CPU partitions oleg159-server: ptlrpc/ptlrpc options: 'lbug_on_grant_miscount=1' oleg159-server: quota/lquota options: 'hash_lqs_cur_bits=3' oleg159-server: mdt/mdt options: 'mdt_enable_flr_ec=1' Setup mgs, mdt, osts pdsh@oleg159-client: oleg159-server: ssh exited with exit code 1 Start mds1: mount -t lustre -o localrecov /dev/mapper/mds1_flakey /mnt/lustre-mds1 oleg159-server: oleg159-server.virtnet: executing set_default_debug neterror net nettrace malloc all pdsh@oleg159-client: oleg159-server: ssh exited with exit code 1 Started lustre-MDT0000 pdsh@oleg159-client: oleg159-server: ssh exited with exit code 1 Start ost1: mount -t lustre -o localrecov /dev/mapper/ost1_flakey /mnt/lustre-ost1 seq.cli-lustre-OST0000-super.width=65536 oleg159-server: oleg159-server.virtnet: executing set_default_debug neterror net nettrace malloc all pdsh@oleg159-client: oleg159-server: ssh exited with exit code 1 Started lustre-OST0000 pdsh@oleg159-client: oleg159-server: ssh exited with exit code 1 Start ost2: mount -t lustre -o localrecov /dev/mapper/ost2_flakey /mnt/lustre-ost2 seq.cli-lustre-OST0001-super.width=65536 oleg159-server: oleg159-server.virtnet: executing set_default_debug neterror net nettrace malloc all pdsh@oleg159-client: oleg159-server: ssh exited with exit code 1 Started lustre-OST0001 Starting client: oleg159-client.virtnet: -o user_xattr,flock 192.168.201.159@tcp:/lustre /mnt/lustre Starting client oleg159-client.virtnet: -o user_xattr,flock 192.168.201.159@tcp:/lustre /mnt/lustre Started clients oleg159-client.virtnet: 192.168.201.159@tcp:/lustre on /mnt/lustre type lustre (rw,flock,checksum,encrypt,lazystatfs,lruresize,nolock,statfs_project,nosync_on_close,nouser_fid2path,user_xattr,verbose) Using TIMEOUT=20 osc.lustre-OST0000-osc-ffff9f6285582800.idle_timeout=debug osc.lustre-OST0001-osc-ffff9f6285582800.idle_timeout=debug disable quota as required