Description of problem: NVMe connect command hung and Call traces Version-Release number of selected component (if applicable): 5.14.0-70.13.1.el9_0.x86_64 How reproducible: Easily Steps to Reproduce: 1) Configure the system for HPE Millbury testing. 2) RHEL 9 BFS system on SN1700Q adapter. 3) Present NVMe namespace 4) Reboot the server or Initiator Port toggle 5) Expected: NVMe namespace are discovered without any call trace 6) Actual: NVMe namespace not discovered and connect command got hung and resulted in Call trace. Actual results: Connect command hangs Expected results: Connect should not hang -------------- [root@MillburyRegRH9 ~]# nvme list-subsys nvme-subsys0 - NQN=nqn.2020-07.com.hpe:5553f61e-48a2-449b-a28c-759d4d938820 \ +- nvme0 fc traddr=nn-0x2ff70102ac0282dd:pn-0x22340102adf282dd host_traddr=nn-0x51402ec012c9cdef:pn-0x51402ec012c9cdee live +- nvme1 fc traddr=nn-0x2ff70102ac0282dd:pn-0x22340102adf282dd host_traddr=nn-0x51402ec012c9ab5b:pn-0x51402ec012c9ab5a live +- nvme2 fc traddr=nn-0x2ff70102ac0282dd:pn-0x23320102adf282dd host_traddr=nn-0x51402ec012c9ab5b:pn-0x51402ec012c9ab5a live +- nvme3 fc traddr=nn-0x2ff70102ac0282dd:pn-0x23320102adf282dd host_traddr=nn-0x51402ec012c9cdef:pn-0*x51402ec012c9cdee connecting* ----------------- Additional info:
Call trace: Aug 9 11:59:27 MillburyRegRH9 systemd[9932]: Finished Cleanup of User's Temporary Files and Directories. Aug 9 11:59:41 MillburyRegRH9 kernel: INFO: task nvme:8970 blocked for more than 245 seconds. Aug 9 11:59:41 MillburyRegRH9 kernel: Tainted: G OE --------- — 5.14.0-70.13.1.el9_0.x86_64 #1 Aug 9 11:59:41 MillburyRegRH9 kernel: "echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message. Aug 9 11:59:41 MillburyRegRH9 kernel: task:nvme state stack: 0 pid: 8970 ppid: 1 flags:0x00004000 Aug 9 11:59:41 MillburyRegRH9 kernel: Call Trace: Aug 9 11:59:41 MillburyRegRH9 kernel: __schedule+0x203/0x560 Aug 9 11:59:41 MillburyRegRH9 kernel: schedule+0x43/0xb0 Aug 9 11:59:41 MillburyRegRH9 kernel: schedule_timeout+0x115/0x150 Aug 9 11:59:41 MillburyRegRH9 kernel: ? enqueue_task+0x48/0x140 Aug 9 11:59:41 MillburyRegRH9 kernel: ? __prepare_to_swait+0x4b/0x70 Aug 9 11:59:41 MillburyRegRH9 kernel: wait_for_completion+0x89/0xe0 Aug 9 11:59:41 MillburyRegRH9 kernel: __flush_work.isra.0+0x160/0x220 Aug 9 11:59:41 MillburyRegRH9 kernel: ? flush_workqueue_prep_pwqs+0x110/0x110 Aug 9 11:59:41 MillburyRegRH9 kernel: nvme_fc_init_ctrl+0x4f3/0x510 [nvme_fc] Aug 9 11:59:41 MillburyRegRH9 kernel: nvme_fc_create_ctrl+0x1ac/0x250 [nvme_fc] Aug 9 11:59:41 MillburyRegRH9 kernel: nvmf_create_ctrl+0x12c/0x220 [nvme_fabrics] Aug 9 11:59:41 MillburyRegRH9 kernel: nvmf_dev_write+0x7d/0xd2 [nvme_fabrics] Aug 9 11:59:41 MillburyRegRH9 kernel: vfs_write+0xb9/0x270 Aug 9 11:59:41 MillburyRegRH9 kernel: ksys_write+0x5f/0xe0 Aug 9 11:59:41 MillburyRegRH9 kernel: do_syscall_64+0x38/0x90 Aug 9 11:59:41 MillburyRegRH9 kernel: entry_SYSCALL_64_after_hwframe+0x44/0xae Aug 9 11:59:41 MillburyRegRH9 kernel: RIP: 0033:0x7fe34c1ab127 Aug 9 11:59:41 MillburyRegRH9 kernel: RSP: 002b:00007ffed1eabc28 EFLAGS: 00000246 ORIG_RAX: 0000000000000001 Aug 9 11:59:41 MillburyRegRH9 kernel: RAX: ffffffffffffffda RBX: 0000000000000153 RCX: 00007fe34c1ab127 Aug 9 11:59:41 MillburyRegRH9 kernel: RDX: 0000000000000153 RSI: 00007ffed1eacd00 RDI: 0000000000000003 Aug 9 11:59:41 MillburyRegRH9 kernel: RBP: 000055649a4eea50 R08: 000000000000002b R09: 00007fe34c21d4e0 Aug 9 11:59:41 MillburyRegRH9 kernel: R10: 0000000000000000 R11: 0000000000000246 R12: 000055649855a160 Aug 9 11:59:41 MillburyRegRH9 kernel: R13: 000055649855a1bb R14: 0000000000000000 R15: 0000000000000001 Aug 9 11:59:41 MillburyRegRH9 kernel: INFO: task nvme:8979 blocked for more than 245 seconds. Aug 9 11:59:41 MillburyRegRH9 kernel: Tainted: G OE --------- — 5.14.0-70.13.1.el9_0.x86_64 #1 Aug 9 11:59:41 MillburyRegRH9 kernel: "echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message. Aug 9 11:59:41 MillburyRegRH9 kernel: task:nvme state stack: 0 pid: 8979 ppid: 1 flags:0x00000000 Aug 9 11:59:41 MillburyRegRH9 kernel: Call Trace: Aug 9 11:59:41 MillburyRegRH9 kernel: __schedule+0x203/0x560 Aug 9 11:59:41 MillburyRegRH9 kernel: schedule+0x43/0xb0 Aug 9 11:59:41 MillburyRegRH9 kernel: schedule_preempt_disabled+0x11/0x20 Aug 9 11:59:41 MillburyRegRH9 kernel: __mutex_lock.constprop.0+0x234/0x430 Aug 9 11:59:41 MillburyRegRH9 kernel: ? __fput+0xff/0x240 Aug 9 11:59:41 MillburyRegRH9 kernel: ? __check_object_size.part.0+0x11f/0x140 Aug 9 11:59:41 MillburyRegRH9 kernel: ? _copy_from_user+0x28/0x60 Aug 9 11:59:41 MillburyRegRH9 kernel: nvmf_dev_write+0x44/0xd2 [nvme_fabrics] Aug 9 11:59:41 MillburyRegRH9 kernel: vfs_write+0xb9/0x270 Aug 9 11:59:41 MillburyRegRH9 kernel: ksys_write+0x5f/0xe0 Aug 9 11:59:41 MillburyRegRH9 kernel: do_syscall_64+0x38/0x90 Aug 9 11:59:41 MillburyRegRH9 kernel: entry_SYSCALL_64_after_hwframe+0x44/0xae Aug 9 11:59:41 MillburyRegRH9 kernel: RIP: 0033:0x7f56e234f127 Aug 9 11:59:41 MillburyRegRH9 kernel: RSP: 002b:00007ffc79ecb9f8 EFLAGS: 00000246 ORIG_RAX: 0000000000000001 Aug 9 11:59:41 MillburyRegRH9 kernel: RAX: ffffffffffffffda RBX: 0000000000000153 RCX: 00007f56e234f127 Aug 9 11:59:41 MillburyRegRH9 kernel: RDX: 0000000000000153 RSI: 00007ffc79eccad0 RDI: 0000000000000003 Aug 9 11:59:41 MillburyRegRH9 kernel: RBP: 000056267f5afe90 R08: 000000000000002b R09: 00007f56e23c14e0 Aug 9 11:59:41 MillburyRegRH9 kernel: R10: 0000000000000000 R11: 0000000000000246 R12: 000056267e72b160 Aug 9 11:59:41 MillburyRegRH9 kernel: R13: 000056267e72b1bb R14: 0000000000000000 R15: 0000000000000001
I have share the crash files at https://secureshare.marvell.com/w/f-48f7cf08-3045-4a28-ae84-844104798998
Created attachment 1905901 [details] Analaysis
Sauruv, Is this problem reproducible with RHEL9.2?
Hi John, It is not reproducible on RH9.2. Even on 9.0, it's not easily reproducible. We can close this BZ for now and will reopen if we are able to figure out a consistent way of reproduction.