Details
-
Bug
-
Resolution: Fixed
-
Minor
-
Lustre 2.4.0
-
3
-
5539
Description
Hit this assertion below on the MDS. Was running "/etc/init.d/lustre stop" on the MDS while the OSSs had just been rebooted and going through recovery.
LustreError: 32636:0:(ldlm_lockd.c:684:ldlm_handle_ast_error()) ### client (nid 172.20.2.191@o2ib500) returned 0 from blocking AST ns: MGS lock: ffff880fc3ef0080/0x2fff3cc7fa507817 lrc: 1/0,0 mode: --/CR res: 128038972191596/0 rrc: 504 type: PLN flags: 0x200000a0 nid: 172.20.2.191@o2ib500 remote: 0x561a2e8961e1bd9c expref: 7 pid: 32636 timeout 4295211050 LustreError: 32636:0:(ldlm_lockd.c:684:ldlm_handle_ast_error()) ### client (nid 172.20.2.191@o2ib500) returned 0 from blocking AST ns: MGS lock: ffff880fc89f4580/0x2fff3cc7fa507810 lrc: 1/0,0 mode: --/CR res: 128038972191596/0 rrc: 5 type: PLN flags: 0x200000a0 nid: 172.20.2.191@o2ib500 remote: 0x561a2e8961e1b871 expref: 4 pid: 32636 timeout 4295211050 LustreError: 32636:0:(ldlm_lock.c:853:ldlm_lock_decref_and_cancel()) ASSERTION( lock != ((void *)0) ) failed: LustreError: 32636:0:(ldlm_lock.c:853:ldlm_lock_decref_and_cancel()) LBUG
crash> bt PID: 32636 TASK: ffff880fd486a080 CPU: 2 COMMAND: "ll_mgs_0006" #0 [ffff880fc2929918] machine_kexec at ffffffff8103216b #1 [ffff880fc2929978] crash_kexec at ffffffff810b8d12 #2 [ffff880fc2929a48] panic at ffffffff814eea99 #3 [ffff880fc2929ac8] lbug_with_loc at ffffffffa05b8fcb [libcfs] #4 [ffff880fc2929ae8] ldlm_lock_decref_and_cancel at ffffffffa08b64ac [ptlrpc] #5 [ffff880fc2929b08] mgs_completion_ast_config at ffffffffa0e171b2 [mgs] #6 [ffff880fc2929b58] ldlm_cli_enqueue_local at ffffffffa08d15c6 [ptlrpc] #7 [ffff880fc2929be8] mgs_revoke_lock at ffffffffa0e16db3 [mgs] #8 [ffff880fc2929c88] mgs_handle_target_reg at ffffffffa0e17a98 [mgs] #9 [ffff880fc2929d08] mgs_handle at ffffffffa0e19d6b [mgs] #10 [ffff880fc2929d98] ptlrpc_server_handle_request at ffffffffa090a8bc [ptlrpc] #11 [ffff880fc2929e98] ptlrpc_main at ffffffffa090beac [ptlrpc] #12 [ffff880fc2929f48] kernel_thread at ffffffff8100c14a