After update of kernel to 5.14.21-150500.55.59.1 (or ...55.62.2), we get the
following kernel BUG in the journalctl log when trying to mount an NFSv4
Kerberos authenticated disk:
May 16 11:08:22 eoc.dlr.de kernel: BUG: kernel NULL pointer dereference,
address: 0000000000000004
May 16 11:08:22 eoc.dlr.de kernel: #PF: supervisor read access in kernel mode
May 16 11:08:22 eoc.dlr.de kernel: #PF: error_code(0x0000) - not-present page
May 16 11:08:22 eoc.dlr.de kernel: PGD 0 P4D 0
May 16 11:08:22 eoc.dlr.de kernel: Oops: 0000 [#1] PREEMPT SMP PTI
May 16 11:08:22 eoc.dlr.de kernel: CPU: 5 PID: 8166 Comm: mount.nfs Tainted: G
X N 5.14.21-150500.55.62-default #1 SLE15-SP5
c363f6b34a951d50530113de3511a86499c762cd
May 16 11:08:22 eoc.dlr.de kernel: Hardware name: Dell Inc.
May 16 11:08:22 eoc.dlr.de kernel: RIP: 0010:gss_get_mic_kerberos+0x2d/0x300
[rpcsec_gss_krb5]
May 16 11:08:22 eoc.dlr.de kernel: Code: 00 00 41 57 41 56 41 55 41 54 49 89 d5
55 53 49 89 f4 48 83 ec 38 48 8b 5f 08 65 48 8b 04 25 28 00 00 00 48 89 44 24
30 31 c0 <8b> 43 04 83 f8 06 0f 84 fa 00 00 >
May 16 11:08:22 eoc.dlr.de kernel: RSP: 0018:ffff9bbc8a99f870 EFLAGS: 00010246
May 16 11:08:22 eoc.dlr.de kernel: RAX: 0000000000000000 RBX: 0000000000000000
RCX: ffff90acbd132ad0
May 16 11:08:22 eoc.dlr.de kernel: RDX: ffff9bbc8a99f900 RSI: ffff9bbc8a99f920
RDI: ffff90ac854905e0
May 16 11:08:22 eoc.dlr.de kernel: RBP: ffff9bbc8a99f9a0 R08: ffff90acbd132008
R09: ffff90acbd132ad0
May 16 11:08:22 eoc.dlr.de kernel: R10: ffff9bbc8a99f9c0 R11: 000000000011ee2a
R12: ffff9bbc8a99f920
May 16 11:08:22 eoc.dlr.de kernel: R13: ffff9bbc8a99f900 R14: ffff90adb2751240
R15: ffff90acbd132020
May 16 11:08:22 eoc.dlr.de kernel: FS: 00007fc244241840(0000)
GS:ffff90b38eb40000(0000) knlGS:0000000000000000
May 16 11:08:22 eoc.dlr.de kernel: CS: 0010 DS: 0000 ES: 0000 CR0:
0000000080050033
May 16 11:08:22 eoc.dlr.de kernel: CR2: 0000000000000004 CR3: 0000000138690006
CR4: 00000000001706e0
May 16 11:08:22 eoc.dlr.de kernel: Call Trace:
May 16 11:08:22 eoc.dlr.de kernel: <TASK>
May 16 11:08:22 eoc.dlr.de kernel: gss_marshal+0x181/0x320 [auth_rpcgss
0f368b4c04963921938df1845be6981ef9b3b687]
May 16 11:08:22 eoc.dlr.de kernel: ? __rpc_sleep_on_priority_timeout+0xf0/0xf0
[sunrpc 21bd2d2d2f7dd857ffba00711d9cf31d8a1e34b3]
May 16 11:08:22 eoc.dlr.de kernel: ? call_encode+0x151/0x310 [sunrpc
21bd2d2d2f7dd857ffba00711d9cf31d8a1e34b3]
May 16 11:08:22 eoc.dlr.de kernel: ? gss_create_cred+0xe0/0xe0 [auth_rpcgss
0f368b4c04963921938df1845be6981ef9b3b687]
May 16 11:08:22 eoc.dlr.de kernel: call_encode+0x151/0x310 [sunrpc
21bd2d2d2f7dd857ffba00711d9cf31d8a1e34b3]
May 16 11:08:22 eoc.dlr.de kernel: ? rpc_malloc+0x4d/0x90 [sunrpc
21bd2d2d2f7dd857ffba00711d9cf31d8a1e34b3]
May 16 11:08:22 eoc.dlr.de kernel: ? call_decode+0x210/0x210 [sunrpc
21bd2d2d2f7dd857ffba00711d9cf31d8a1e34b3]
May 16 11:08:22 eoc.dlr.de kernel: __rpc_execute+0xb7/0x4d0 [sunrpc
21bd2d2d2f7dd857ffba00711d9cf31d8a1e34b3]
May 16 11:08:22 eoc.dlr.de kernel: rpc_execute+0x6b/0xd0 [sunrpc
21bd2d2d2f7dd857ffba00711d9cf31d8a1e34b3]
May 16 11:08:22 eoc.dlr.de kernel: rpc_run_task+0x111/0x130 [sunrpc
21bd2d2d2f7dd857ffba00711d9cf31d8a1e34b3]
May 16 11:08:22 eoc.dlr.de kernel: nfs4_call_sync_custom+0xb/0x30 [nfsv4
d59cef2e53679b0000b74b5adfd2d9a4aadf9e3e]
May 16 11:08:22 eoc.dlr.de kernel: nfs4_proc_setclientid+0x18b/0x450 [nfsv4
d59cef2e53679b0000b74b5adfd2d9a4aadf9e3e]
May 16 11:08:22 eoc.dlr.de kernel: ? try_to_wake_up+0x177/0x550
May 16 11:08:22 eoc.dlr.de kernel: nfs40_discover_server_trunking+0x8e/0x110
[nfsv4 d59cef2e53679b0000b74b5adfd2d9a4aadf9e3e]
May 16 11:08:22 eoc.dlr.de kernel: nfs4_discover_server_trunking+0x72/0x2a0
[nfsv4 d59cef2e53679b0000b74b5adfd2d9a4aadf9e3e]
May 16 11:08:22 eoc.dlr.de kernel: nfs4_init_client+0x6e/0x130 [nfsv4
d59cef2e53679b0000b74b5adfd2d9a4aadf9e3e]
May 16 11:08:22 eoc.dlr.de kernel: nfs4_set_client+0xef/0x140 [nfsv4
d59cef2e53679b0000b74b5adfd2d9a4aadf9e3e]
May 16 11:08:22 eoc.dlr.de kernel: nfs4_create_server+0x128/0x2b0 [nfsv4
d59cef2e53679b0000b74b5adfd2d9a4aadf9e3e]
May 16 11:08:22 eoc.dlr.de kernel: nfs4_try_get_tree+0x33/0xd0 [nfsv4
d59cef2e53679b0000b74b5adfd2d9a4aadf9e3e]
May 16 11:08:22 eoc.dlr.de kernel: vfs_get_tree+0x25/0xd0
May 16 11:08:22 eoc.dlr.de kernel: path_mount+0x6f2/0x9c0
May 16 11:08:22 eoc.dlr.de kernel: do_mount+0x79/0x90
...
A kernel revert to .55.52.1 makes it go away, so probably it is due to some
NFSv4/Kerberos patch appearing between 5.14.21-150500.55.52.1 and ....55.59.1
Thanks for any advice or fix!