Problem with Automated Umounts on SLES10SP2, Invalid memory.

[tecs_dlsi]

Hi,

We are currently experiencing problems with SLES10.2, running Automount over NFS.

We have recently suffered failures of mounts, causing the kernel to go into invalid memory. We believe this to be happening when a mount is inactive for a length of time, and Automount is umounting these mount points.

We are suffering the same problem with SLES10.2 as soon as we try to mount an NFS export. The first time we try to use 'mount -t nfs ...' we get something like this:

Oct 29 15:46:12 sun2 kernel: Unable to handle kernel paging request at fffffffffffffffa RIP:
Oct 29 15:46:12 sun2 kernel: <ffffffff88355b74>{:nfs:nfs_proc_get_root+78}
Oct 29 15:46:12 sun2 kernel: PGD 103027 PUD 1030067 PMD 0
Oct 29 15:46:12 sun2 kernel: Oops: 0000 [1] SMP
Oct 29 15:46:12 sun2 kernel: last sysfs file: /devices/pci0000:00/0000:00:00.0/irq
Oct 29 15:46:12 sun2 kernel: CPU 3
Oct 29 15:46:12 sun2 kernel: Modules linked in: nfs lockd nfs_acl sunrpc ipv6 dock button battery ac apparmor loop sr_mod cdrom dm_mod usbhid u
sb_storage ehci_hcd uhci_hcd usbcore i2c_i801 shpchp i2c_core pci_hotplug hw_random e1000 mptctl ext3 jbd generic edd fan thermal processor sg
mptsas mptscsih mptbase scsi_transport_sas ata_piix libata piix sd_mod scsi_mod ide_disk ide_core
Oct 29 15:46:12 sun2 kernel: Pid: 5092, comm: mount Not tainted 2.6.16.60-0.31-smp #1
Oct 29 15:46:12 sun2 kernel: RIP: 0010:[<ffffffff88355b74>] <ffffffff88355b74>{:nfs:nfs_proc_get_root+78}
Oct 29 15:46:12 sun2 kernel: RSP: 0000:ffff810200e13a38 EFLAGS: 00010246
Oct 29 15:46:12 sun2 kernel: RAX: ffffffff88381d20 RBX: ffff810200e13b78 RCX: 0000000000000000
Oct 29 15:46:12 sun2 kernel: RDX: 00000001000123a2 RSI: ffff810217419e98 RDI: ffffffffffffffea
Oct 29 15:46:12 sun2 kernel: RBP: ffff810200e13aa8 R08: 00000000ffffffff R09: 0000000000000002
Oct 29 15:46:12 sun2 kernel: R10: 0000000000000000 R11: 0000000000000000 R12: ffff810217419e98
Oct 29 15:46:12 sun2 kernel: R13: ffff810217419e00 R14: 0000000000000001 R15: ffff81021483a400
Oct 29 15:46:12 sun2 kernel: FS: 00002b5145c006d0(0000) GS:ffff810217c918c0(0000) knlGS:0000000000000000
Oct 29 15:46:12 sun2 kernel: CS: 0010 DS: 0000 ES: 0000 CR0: 000000008005003b
Oct 29 15:46:12 sun2 kernel: CR2: fffffffffffffffa CR3: 0000000202546000 CR4: 00000000000006e0
Oct 29 15:46:12 sun2 kernel: Process mount (pid: 5092, threadinfo ffff810200e12000, task ffff810216615820)
Oct 29 15:46:12 sun2 kernel: Stack: 0000000000000286 ffff810200e13ac8 0000000000000001 ffffffff8830a623
Oct 29 15:46:12 sun2 kernel: 0000000000000000 ffff810200e13ad8 ffff810200e13ac8 ffff81000102fa18
Oct 29 15:46:12 sun2 kernel: ffff810217419c00 ffff810217419e00
Oct 29 15:46:12 sun2 kernel: Call Trace: <ffffffff8830a623>{:sunrpc:rpc_wait_bit_interrupti ble+0}
Oct 29 15:46:12 sun2 kernel: <ffffffff883527a1>{:nfs:nfs_sb_init+158} <ffffffff8013ce8a>{lock_timer_base+27}
Oct 29 15:46:12 sun2 kernel: <ffffffff8012d6bc>{__wake_up+56} <ffffffff8830a7ab>{:sunrpc:rpc_release_task+293}
Oct 29 15:46:12 sun2 kernel: <ffffffff883055ba>{:sunrpc:rpc_call_sync+161} <ffffffff88305c46>{:sunrpc:rpc_new_client+619}
Oct 29 15:46:12 sun2 kernel: <ffffffff88353fcc>{:nfs:nfs_get_sb+1391} <ffffffff8018c543>{do_kern_mount+166}
Oct 29 15:46:12 sun2 kernel: <ffffffff801a1c96>{do_mount+1790} <ffffffff801a0887>{mntput_no_expire+29}
Oct 29 15:46:12 sun2 kernel: <ffffffff80195815>{link_path_walk+218} <ffffffff802ed00b>{do_page_fault+1018}
Oct 29 15:46:13 sun2 kernel: <ffffffff801666dd>{__alloc_pages+113} <ffffffff801661d8>{__get_free_pages+27}
Oct 29 15:46:13 sun2 kernel: <ffffffff801a1fbb>{sys_mount+139} <ffffffff8010ae42>{system_call+126}
Oct 29 15:46:13 sun2 kernel:
Oct 29 15:46:13 sun2 kernel: Code: 48 8b 57 10 48 89 e6 48 89 6c 24 10 4c 89 64 24 08 48 c7 44
Oct 29 15:46:13 sun2 kernel: RIP <ffffffff88355b74>{:nfs:nfs_proc_get_root+78} RSP <ffff810200e13a38>
Oct 29 15:46:13 sun2 kernel: CR2: fffffffffffffffa

No problem with SLES10.1. Any suggestions?