Kernel Oops with nfs4 mount
Affects | Status | Importance | Assigned to | Milestone | |
---|---|---|---|---|---|
nfs-utils (Ubuntu) |
New
|
Undecided
|
Unassigned |
Bug Description
When accessing nfs4 mounts I get kernel 2 Oops.
On the server I am running the latest feisty version:
ii nfs-common 1.0.12-4 NFS support files common to client and serve
ii nfs-kernel-ser 1.0.12-4 support for NFS kernel server
On the client I am running the latest gutsy version:
ii nfs-common 1:1.1.1~
ii nfs-kernel-server 1:1.1.1~
The Oops does not happen after the mount, but after some random read / write time using the share.
Oct 3 17:15:42 rumo kernel: [112389.656000] Oops: 0000 [#1]
Oct 3 17:15:42 rumo kernel: [112389.656000] SMP
Oct 3 17:15:42 rumo kernel: [112389.656000] Modules linked in: isofs udf af_packet binfmt_misc rfcomm l2cap bluetooth nfsd exportfs thinkpad_acpi ppdev autofs4 fglrx(P) ipv6 acpi_cpufreq cpufreq_ondemand cpufreq_
Oct 3 17:15:42 rumo kernel: [112389.656000] CPU: 0
Oct 3 17:15:42 rumo kernel: [112389.656000] EIP: 0060:[<f8d30cf5>] Tainted: P VLI
Oct 3 17:15:42 rumo kernel: [112389.656000] EFLAGS: 00010246 (2.6.22-12-generic #1)
Oct 3 17:15:42 rumo kernel: [112389.656000] EIP is at nfs_update_
Oct 3 17:15:42 rumo kernel: [112389.656000] eax: 00000000 ebx: 000081a4 ecx: 000081a4 edx: 00008000
Oct 3 17:15:42 rumo kernel: [112389.656000] esi: 0006e007 edi: d5cdfe40 ebp: d00b5890 esp: ca9d1dd8
Oct 3 17:15:42 rumo kernel: [112389.656000] ds: 007b es: 007b fs: 00d8 gs: 0033 ss: 0068
Oct 3 17:15:42 rumo kernel: [112389.656000] Process umount.nfs4 (pid: 28956, ti=ca9d0000 task=ddf2c000 task.ti=ca9d0000)
Oct 3 17:15:42 rumo kernel: [112389.656000] Stack: d00b5890 c02f373b 00000001 cf380800 00000004 00000000 ddf2c000 cf64d6b0
Oct 3 17:15:42 rumo kernel: [112389.656000] d00b5768 ca9d1e1c d00b5890 01ab968e c01320c5 cf380780 00000000 d00b5890
Oct 3 17:15:42 rumo kernel: [112389.656000] cf64d6b0 cf380780 d00b5890 f8d326d4 cf64d600 00000000 f8d404fb cf64d600
Oct 3 17:15:42 rumo kernel: [112389.656000] Call Trace:
Oct 3 17:15:42 rumo kernel: [112389.656000] [out_of_
Oct 3 17:15:42 rumo kernel: [112389.656000] [sigprocmask+
Oct 3 17:15:42 rumo kernel: [112389.656000] [<f8d326d4>] nfs_post_
Oct 3 17:15:42 rumo kernel: [112389.656000] [<f8d404fb>] nfs4_proc_
Oct 3 17:15:42 rumo kernel: [112389.656000] [<f8d4c867>] nfs_do_
Oct 3 17:15:42 rumo kernel: [112389.656000] [<f8d2f2e9>] nfs_dentry_
Oct 3 17:15:42 rumo kernel: [112389.656000] [shrink_
Oct 3 17:15:42 rumo kernel: [112389.656000] [__wake_
Oct 3 17:15:42 rumo kernel: [112389.656000] [shrink_
Oct 3 17:15:42 rumo kernel: [112389.656000] [generic_
Oct 3 17:15:42 rumo kernel: [112389.656000] [kill_anon_
Oct 3 17:15:42 rumo kernel: [112389.656000] [<f8d3328c>] nfs_kill_
Oct 3 17:15:42 rumo kernel: [112389.656000] [deactivate_
Oct 3 17:15:42 rumo kernel: [112389.656000] [expire_
Oct 3 17:15:42 rumo kernel: [112389.656000] [shrink_
Oct 3 17:15:43 rumo kernel: [112389.656000] [sys_umount+
Oct 3 17:15:43 rumo kernel: [112389.656000] [dput+135/256] dput+0x87/0x100
Oct 3 17:15:43 rumo kernel: [112389.656000] [__fput+299/416] __fput+0x12b/0x1a0
Oct 3 17:15:43 rumo kernel: [112389.656000] [mntput_
Oct 3 17:15:43 rumo kernel: [112389.656000] [sys_oldumount+
Oct 3 17:15:43 rumo kernel: [112389.656000] [sysenter_
Oct 3 17:15:43 rumo kernel: [112389.656000] =======
Oct 3 17:15:43 rumo kernel: [112389.656000] Code: 0f ba ab 8c 00 00 00 02 89 e8 e8 e7 fc ff ff b8 8c ff ff ff 83 c4 3c 5b 5e 5f 5d c3 8b 85 98 00 00 00 8b b8 64 01 00 00 8b 40 38 <3b> 68 0c 0f 84 57 04 00 00 8b 54 24 1c 8b 4c 24 20 8b 82 8c 00
Oct 3 17:15:43 rumo kernel: [112389.656000] EIP: [<f8d30cf5>] nfs_update_
Oct 3 17:15:43 rumo automount[28954]: >> mount.nfs4: /mnt/nfs4/smeik is already mounted or busy
Oct 3 17:15:43 rumo automount[28954]: mount(generic): failed to mount smeik:/ (type nfs4) on /mnt/nfs4/smeik
Oct 3 17:15:43 rumo automount[28954]: failed to recover from partial expiry of /mnt/nfs4/smeik
Oct 3 17:16:16 rumo kernel: [112423.544000] BUG: unable to handle kernel NULL pointer dereference at virtual address 0000000c
Oct 3 17:16:16 rumo kernel: [112423.544000] printing eip:
Oct 3 17:16:16 rumo kernel: [112423.544000] f8d30cf5
Oct 3 17:16:16 rumo kernel: [112423.544000] *pde = 00000000
Oct 3 17:16:16 rumo kernel: [112423.544000] Oops: 0000 [#2]
Oct 3 17:16:16 rumo kernel: [112423.544000] SMP
Oct 3 17:16:16 rumo kernel: [112423.544000] Modules linked in: isofs udf af_packet binfmt_misc rfcomm l2cap bluetooth nfsd exportfs thinkpad_acpi ppdev autofs4 fglrx(P) ipv6 acpi_cpufreq cpufreq_ondemand cpufreq_
Oct 3 17:16:16 rumo kernel: [112423.544000] CPU: 1
Oct 3 17:16:16 rumo kernel: [112423.544000] EIP: 0060:[<f8d30cf5>] Tainted: P VLI
Oct 3 17:16:16 rumo kernel: [112423.544000] EFLAGS: 00010246 (2.6.22-12-generic #1)
Oct 3 17:16:16 rumo kernel: [112423.544000] EIP is at nfs_update_
Oct 3 17:16:16 rumo kernel: [112423.544000] eax: 00000000 ebx: 000081a4 ecx: 000081a4 edx: 00008000
Oct 3 17:16:16 rumo kernel: [112423.544000] esi: 0006e003 edi: d5cdfe40 ebp: d00b5618 esp: c9bb5f24
Oct 3 17:16:16 rumo kernel: [112423.544000] ds: 007b es: 007b fs: 00d8 gs: 0000 ss: 0068
Oct 3 17:16:16 rumo kernel: [112423.544000] Process 192.168.1.4-del (pid: 30508, ti=c9bb4000 task=ddf2d4c0 task.ti=c9bb4000)
Oct 3 17:16:16 rumo kernel: [112423.544000] Stack: d00b5618 c02f373b 00000001 e386af80 00000004 00000000 ddf2d4c0 d550bcb0
Oct 3 17:16:16 rumo kernel: [112423.544000] d00b54f0 c9bb5f68 d00b5618 01abb7a6 c01320c5 e386af00 00000000 d00b5618
Oct 3 17:16:16 rumo kernel: [112423.544000] d550bcb0 e386af00 d00b5618 f8d326d4 d550bc00 00000000 f8d404fb d550bc00
Oct 3 17:16:16 rumo kernel: [112423.544000] Call Trace:
Oct 3 17:16:16 rumo kernel: [112423.544000] [out_of_
Oct 3 17:16:16 rumo kernel: [112423.544000] [sigprocmask+
Oct 3 17:16:16 rumo kernel: [112423.544000] [<f8d326d4>] nfs_post_
Oct 3 17:16:16 rumo kernel: [112423.544000] [<f8d404fb>] nfs4_proc_
Oct 3 17:16:16 rumo kernel: [112423.544000] [<f8d4c867>] nfs_do_
Oct 3 17:16:16 rumo kernel: [112423.544000] [<f8d4d0f7>] nfs_do_
Oct 3 17:16:16 rumo kernel: [112423.544000] [<f8d4d080>] nfs_do_
Oct 3 17:16:16 rumo kernel: [112423.544000] [kthread+66/112] kthread+0x42/0x70
Oct 3 17:16:16 rumo kernel: [112423.544000] [kthread+0/112] kthread+0x0/0x70
Oct 3 17:16:16 rumo kernel: [112423.544000] [kernel_
Oct 3 17:16:16 rumo kernel: [112423.544000] =======
Oct 3 17:16:16 rumo kernel: [112423.544000] Code: 0f ba ab 8c 00 00 00 02 89 e8 e8 e7 fc ff ff b8 8c ff ff ff 83 c4 3c 5b 5e 5f 5d c3 8b 85 98 00 00 00 8b b8 64 01 00 00 8b 40 38 <3b> 68 0c 0f 84 57 04 00 00 8b 54 24 1c 8b 4c 24 20 8b 82 8c 00
Oct 3 17:16:16 rumo kernel: [112423.544000] EIP: [<f8d30cf5>] nfs_update_
Oct 3 17:16:52 rumo automount[306]: >> mount.nfs4: can't get address for .hidden
Oct 3 17:16:52 rumo automount[306]: mount(generic): failed to mount .hidden:/ (type nfs4) on /mnt/nfs4/.hidden
Oct 3 17:16:52 rumo automount[306]: failed to mount /mnt/nfs4/.hidden
same result with nfs3 and sec=krb5. mount worsk, but any access to the mount crashes the caller with a similar call trace to above.
I also replicated the nfs4 crash.
In nfs3 the keytab is optional, and the I can get immediate segfaults without the keytab in place. size+0x1/ 0x20 [auth_rpcgss] SS:ESP 0068:f02c9c40
My log output always seems to end with a line about auth_rpcgssd.
kernel: [ 2636.734267] EIP: [<f8dab901>] g_token_
Kerberos is working fine