oops with umount -lf of krb5 NFS mount.

Wed Jul 19 12:15:52 EDT 2006

On Mon, Jul 17, 2006 at 04:28:40PM -0700, Vince Busam wrote:
> We've been able to reproduce an oops in 2.6.16.11 when forcing an unmount 
> of a dead krb5 (nfsv3) mount.

Hi Vince,

We have been seeing some issues during unmounts (esp. during shutdown),
so we have attempted to relicate the issue you're seeing on our
systems.  Unfortunately we haven't been able to reproduce it.

Could you give us more information about your environment, including
versions of libs you're using, hardware/network, etc. etc.?

Thanks,
Bryce

> turn the nfs server off
> cd /path/to/nfsmount/ (this will hang)
> (in a different terminal) umount -lf /path/to/nfsmount/
> (in previous terminal) ctrl+c, cd /
> 
> It doesn't always work, which is frustrating... but it happens fairly 
> often.  Is this a known issue?
> 
> Thanks,
> Vince
> 
> Here's the kernel output (we'll reproduce this with an untainted kernel):
> 
> Jul 17 16:44:04 return kernel: net/sunrpc/rpc_pipe.c: rpc_lookup_parent 
> failed to find path /nfs/clnt62/krb5
> Jul 17 16:44:04 return kernel: Unable to handle kernel NULL pointer 
> dereference at virtual address 00000068
> Jul 17 16:44:04 return kernel: printing eip:
> Jul 17 16:44:04 return kernel: c02b6243
> Jul 17 16:44:04 return kernel: *pde = 00000000
> Jul 17 16:44:04 return kernel: Oops: 0002 [#1]
> Jul 17 16:44:04 return kernel: SMP
> Jul 17 16:44:04 return kernel: Modules linked in: isofs udf nls_iso8859_1 
> nls_cp437 vfat fat des binfmt_misc rfcomm l2cap bluetooth acpi_cpufreq 
> cpufreq_userspace cpufreq_ondemand cpufreq_powersave nfsd exportfs autofs4 
> video button battery container ac nfs lockd af_packet floppy rtc hw_random 
> snd_hda_intel snd_hda_codec snd_pcm_oss snd_mixer_oss snd_pcm snd_timer snd 
> soundcore snd_page_alloc shpchp pci_hotplug dm_mod nvidia agpgart psmouse 
> mousedev parport_pc lp parport ide_disk rpcsec_gss_krb5 auth_rpcgss sunrpc 
> md_mod ext3 jbd mbcache thermal processor fan usb_storage usbhid tg3 
> ehci_hcd uhci_hcd usbcore sd_mod ide_cd cdrom ide_generic ata_piix libata 
> scsi_mod piix generic ide_core unix
> Jul 17 16:44:04 return kernel: CPU: 1
> Jul 17 16:44:04 return kernel: EIP: 0060:[_spin_lock+3/32] Tainted: P VLI
> Jul 17 16:44:04 return kernel: EFLAGS: 00010296 (2.6.16.11-gg1 #1)
> Jul 17 16:44:04 return kernel: EIP is at _spin_lock+0x3/0x20
> Jul 17 16:44:04 return kernel: eax: 00000068 ebx: c2b28900 ecx: dae1bda0 
> edx: 00000000
> Jul 17 16:44:04 return kernel: esi: 00000000 edi: f8957724 ebp: dae1bd64 
> esp: dae1bd64
> Jul 17 16:44:04 return kernel: ds: 007b es: 007b ss: 0068
> Jul 17 16:44:04 return kernel: Process bash (pid: 23302, 
> threadinfo=dae1a000 task=c4d20550)
> Jul 17 16:44:04 return kernel: Stack: <0>dae1bd80 f894eb90 c2b28900 
> 00000068 c2b28900 c2b28a7c f8957724 dae1bdb4
> Jul 17 16:44:04 return kernel: f8b8b65e c2b28900 dae1bda0 f894ebe0 ffffffe0 
> f744dae0 c2b28974 dae1bda0
> Jul 17 16:44:04 return kernel: dae1bda0 f744df00 f744db18 f744d324 dae1be14 
> f8b8bf5f c2b28900 f744d324
> Jul 17 16:44:04 return kernel: Call Trace:
> Jul 17 16:44:04 return kernel: [show_stack_log_lvl+170/224] 
> show_stack_log_lvl+0xaa/0xe0
> Jul 17 16:44:04 return kernel: [show_registers+411/528] 
> show_registers+0x19b/0x210
> Jul 17 16:44:04 return kernel: [die+265/416] die+0x109/0x1a0
> Jul 17 16:44:04 return kernel: [do_page_fault+954/1457] 
> do_page_fault+0x3ba/0x5b1
> Jul 17 16:44:04 return kernel: [error_code+79/84] error_code+0x4f/0x54
> Jul 17 16:44:04 return kernel: [pg0+945359760/1069810688] 
> gss_pipe_release+0x20/0x70 [auth_rpcgss]
> Jul 17 16:44:04 return kernel: [pg0+947705438/1069810688] 
> rpc_close_pipes+0x13e/0x140 [sunrpc]
> Jul 17 16:44:04 return kernel: [pg0+947707743/1069810688] 
> rpc_depopulate+0xaf/0x120 [sunrpc]
> Jul 17 16:44:04 return kernel: [pg0+947709123/1069810688] 
> rpc_rmdir+0x73/0xb0 [sunrpc]
> Jul 17 16:44:04 return kernel: [pg0+947639885/1069810688] 
> rpc_destroy_client+0x9d/0xf0 [sunrpc]
> Jul 17 16:44:04 return kernel: [pg0+947639856/1069810688] 
> rpc_destroy_client+0x80/0xf0 [sunrpc]
> Jul 17 16:44:04 return kernel: [pg0+947639629/1069810688] 
> rpc_shutdown_client+0xdd/0x140 [sunrpc]
> Jul 17 16:44:04 return kernel: [pg0+949308424/1069810688] 
> nfs_kill_super+0x78/0xd0 [nfs]
> Jul 17 16:44:04 return kernel: [deactivate_super+122/176] 
> deactivate_super+0x7a/0xb0
> Jul 17 16:44:04 return kernel: [mntput_no_expire+80/128] 
> mntput_no_expire+0x50/0x80
> Jul 17 16:44:04 return kernel: [do_exit+747/1024] do_exit+0x2eb/0x400
> Jul 17 16:44:04 return kernel: [do_group_exit+55/160] 
> do_group_exit+0x37/0xa0
> Jul 17 16:44:04 return kernel: [sys_exit_group+21/32] 
> sys_exit_group+0x15/0x20
> Jul 17 16:44:04 return kernel: [sysenter_past_esp+84/117] 
> sysenter_past_esp+0x54/0x75
> Jul 17 16:44:04 return kernel: Code: 21 e2 81 42 14 00 01 00 00 f0 81 28 00 
> 00 00 01 74 05 e8 fd d9 ff ff 5d c3 eb 0d 90 90 90 90 90 90 90 90 90 90 90 
> 90 90 55 89 e5 <f0> fe 08 79 09 f3 90 80 38 00 7e f9 eb f2 5d c3 8d b6 00 
> 00 00
> Jul 17 16:44:04 return kernel: <1>Fixing recursive fault but reboot is 
> needed!
> Jul 17 16:44:52 return kernel: device eth0 left promiscuous mode
> _______________________________________________
> NFSv4 mailing list
> NFSv4 at linux-nfs.org
> http://linux-nfs.org/cgi-bin/mailman/listinfo/nfsv4