SAN ストレージの更新中にシステムがクラッシュする

Issue

• 3PAR ファームウェアの更新中に、いくつかのシステムがクラッシュしたり、少なくとも SAN ストレージ (HPE 3PAR) へのパスが失われたりしました。
  この問題は、RHEL 7.6 がインストールされているシステムでのみ発生し、7.4 カーネルを搭載したシステムには影響しません。

• 以下のエラーでは、クラッシュ中に呼び出しトレースが記録されました。

  sd 4:0:1:3: [sdbo] FAILED Result: hostbyte=DID_NO_CONNECT driverbyte=DRIVER_OK
  sd 4:0:1:3: [sdbo] CDB: Write(10) 2a 00 00 38 11 1b 00 00 73 00
  blk_update_request: I/O error, dev sdbo, sector 3674395
  device-mapper: multipath: Failing path 68:16.
  device-mapper: multipath: Failing path 68:32.
  sd 4:0:1:4: rejecting I/O to offline device
  device-mapper: multipath: Failing path 71:112.
  qla2xxx [0000:0b:00.0]-8030:3: TM IOCB failed (9).
  qla2xxx [0000:0b:00.0]-800c:3: do_reset failed for cmd=ffff93c6953e2300.
  qla2xxx [0000:0b:00.0]-800f:3: DEVICE RESET FAILED: Task management failed nexus=3:0:1 cmd=ffff93c6953e2300.
  qla2xxx [0000:0b:00.0]-8009:3: DEVICE RESET ISSUED nexus=3:0:3 cmd=ffff93f4c63cda40.
  qla2xxx [0000:0b:00.0]-8030:3: TM IOCB failed (5).
  qla2xxx [0000:0b:00.0]-800c:3: do_reset failed for cmd=ffff93f4c63cda40.
  qla2xxx [0000:0b:00.0]-800f:3: DEVICE RESET FAILED: Task management failed nexus=3:0:3 cmd=ffff93f4c63cda40.
  qla2xxx [0000:0b:00.0]-8009:3: TARGET RESET ISSUED nexus=3:0:0 cmd=ffff93f4c67b3100.
  qla2xxx [0000:0b:00.0]-8030:3: TM IOCB failed (9).
  qla2xxx [0000:0b:00.0]-800c:3: do_reset failed for cmd=ffff93f4c67b3100.
  qla2xxx [0000:0b:00.0]-800f:3: TARGET RESET FAILED: Task management failed nexus=3:0:0 cmd=ffff93f4c67b3100.
  qla2xxx [0000:0b:00.0]-8012:3: BUS RESET ISSUED nexus=3:0:1.
  BUG: unable to handle kernel NULL pointer dereference at           (null)
  IP: [<ffffffffc02a974a>] qla2x00_eh_wait_on_command+0x1a/0xa0 [qla2xxx]
  PGD 0 
  Oops: 0000 [#1] SMP 
  Modules linked in: nfsv3 nfs_acl rpcsec_gss_krb5 [...] qla2xxx nvme_fc nvme_fabrics
  CPU: 22 PID: 4286 Comm: scsi_eh_3 Kdump: loaded Tainted: P          IOE  ------------   3.10.0-957.5.1.el7.x86_64 #1
  Hardware name: HP ProLiant DL380 G7, BIOS P67 08/16/2015
  task: ffff93fe087b4100 ti: ffff93fd4e894000 task.ti: ffff93fd4e894000
  RIP: 0010:[<ffffffffc02a974a>]  [<ffffffffc02a974a>] qla2x00_eh_wait_on_command+0x1a/0xa0 [qla2xxx]
  RSP: 0018:ffff93fd4e897ce8  EFLAGS: 00010286
  RAX: 0000000000000000 RBX: ffff93d56e8db9c0 RCX: ffff93cddf4a1740
  RDX: 0000000000000140 RSI: ffff93d56e8da680 RDI: ffff93d56e8db9c0
  RBP: ffff93fd4e897cf0 R08: ffff93d56e8d9c00 R09: 0000000000004000
  R10: ffff9400b1433ec0 R11: 0000000000000800 R12: 0000000000000000
  R13: ffff93cddf4a1740 R14: 0000000000000000 R15: ffff93d56e8db9c0
  FS:  0000000000000000(0000) GS:ffff93d6e7cc0000(0000) knlGS:0000000000000000
  CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
  CR2: 0000000000000000 CR3: 0000005931210000 CR4: 00000000000207e0
  Call Trace:
   [<ffffffffc02ad50a>] qla2x00_eh_wait_for_pending_commands+0xda/0x140 [qla2xxx]
   [<ffffffffc02b5a4a>] qla2xxx_eh_bus_reset+0x17a/0x1c0 [qla2xxx]
   [<ffffffffa0ad33c6>] scsi_try_bus_reset+0x46/0x100
   [<ffffffffa0ad52b1>] scsi_eh_ready_devs+0x771/0xc60
   [<ffffffffa0ad6a8c>] scsi_error_handler+0x56c/0x8b0
   [<ffffffffa0ad6520>] ? scsi_eh_get_sense+0x250/0x250
   [<ffffffffa06c1c71>] kthread+0xd1/0xe0
   [<ffffffffa06c1ba0>] ? insert_kthread_work+0x40/0x40
   [<ffffffffa0d74c37>] ret_from_fork_nospec_begin+0x21/0x21
   [<ffffffffa06c1ba0>] ? insert_kthread_work+0x40/0x40
  Code: e8 2c e2 69 e0 eb 97 66 2e 0f 1f 84 00 00 00 00 00 66 66 66 66 90 55 48 89 e5 53 48 8b 07 48 89 fb 48 8b 30 48 8b 86 f0 0b 00 00 <48> 8b 10 83 ba 90 00 00 00 01 75 4a 48 8b 40 10 a9 00 00 04 00 
  RIP  [<ffffffffc02a974a>] qla2x00_eh_wait_on_command+0x1a/0xa0 [qla2xxx]
   RSP <ffff93fd4e897ce8>
  CR2: 0000000000000000
  

• 以下は、3PAR ストレージアレイのファームウェアのアップグレード中に確認された別のクラッシュです。

  CPU: 0 PID: 421 Comm: scsi_eh_1 Kdump: loaded Tainted: G               ------------ T 3.10.0-957.el7.debugscsi.x86_64 #1
  Hardware name: HP ProLiant BL460c Gen9, BIOS I36 02/17/2017
  task: ffffa0626cf2a080 ti: ffffa0626ad18000 task.ti: ffffa0626ad18000
  RIP: 0010:[<0000000000000000>]  [<          (null)>]           (null)
  RSP: 0018:ffffa0626ad1bbf8  EFLAGS: 00010082
  qla2xxx [0000:09:00.1]-801c:2: Abort command issued nexus=2:1:2 --  0 2003.
  RAX: 0000000000000000 RBX: ffffa05f02bdc540 RCX: 000000010024000d
  RDX: 000000010024000e RSI: ffffe63c240af700 RDI: ffffa05f02bdda40
  RBP: ffffa0626ad1bc10 R08: ffffa05f02bdc540 R09: 000000010024000d
  R10: 0000000002bdd801 R11: ffffe63c240af700 R12: ffffa05f02bdda40
  R13: ffffa05e58fc8740 R14: ffffa05e5d84b380 R15: 00000000000005e1
  FS:  0000000000000000(0000) GS:ffffa05e5fc00000(0000) knlGS:0000000000000000
  CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
  CR2: 0000000000000000 CR3: 00000001df610000 CR4: 00000000001607f0
  Call Trace:
   [<ffffffffc0451da4>] ? qla2x00_sp_compl+0x54/0xb0 [qla2xxx]
   [<ffffffffc0451bb2>] __qla2x00_abort_all_cmds+0xc2/0x260 [qla2xxx]
   [<ffffffffc0455bd7>] qla2x00_abort_all_cmds+0x27/0x70 [qla2xxx]
   [<ffffffffc046be13>] qla2x00_abort_isp_cleanup+0x2a3/0x330 [qla2xxx]
   [<ffffffffc046bf9d>] qla2x00_abort_isp+0xfd/0x6d0 [qla2xxx]
   [<ffffffffc04557f5>] qla2xxx_eh_host_reset+0x285/0x2c0 [qla2xxx]
   [<ffffffffc045da1a>] ? qla2xxx_eh_bus_reset+0x14a/0x1c0 [qla2xxx]
   [<ffffffffbecd2e46>] scsi_try_host_reset+0x46/0x100
   [<ffffffffbecd4cf6>] scsi_eh_ready_devs+0x876/0xc60
   [<ffffffffbecd63cc>] scsi_error_handler+0x56c/0x8b0
   [<ffffffffbecd5e60>] ? scsi_eh_get_sense+0x250/0x250
   [<ffffffffbe8c1c31>] kthread+0xd1/0xe0
   [<ffffffffbe8c1b60>] ? insert_kthread_work+0x40/0x40
   [<ffffffffbef74c37>] ret_from_fork_nospec_begin+0x21/0x21
   [<ffffffffbe8c1b60>] ? insert_kthread_work+0x40/0x40
  Code:  Bad RIP value.
  RIP  [<          (null)>]           (null)
   RSP <ffffa0626ad1bbf8>
  CR2: 0000000000000000