Kernel Panic writing to LSI SAS 9200-8e attached disks

I have a HP ML10 Gen 2 server with 16gb ram and a LS! SAS 9200-8e HBA attached to a drive cabinet with 2 Seagate ST3000DM001 drives in a RAID 1 with one partition formatted as EXT4. I am running LEAP 42.1 as a XEN host. LEAP is patched to current levels.

System boots up fine but I get a kernel panic when I start writing to the partition that is attached to the HBA. I have flashed the HBA with the latest bios from LSI and the same with the BIOS on the ML10.

Any help would be appreciated.

Joe

Here is the OOPS backtrace from the journal:

                    Aug 17 16:12:43 lu-tze kernel: BUG: unable to handle kernel NULL pointer dereference at           (null)

Aug 17 16:12:43 lu-tze kernel: IP: < (null)>] (null)
Aug 17 16:12:43 lu-tze kernel: PGD 2abb73067 PUD 29ac28067 PMD 0
Aug 17 16:12:43 lu-tze kernel: Oops: 0010 #1] SMP
Aug 17 16:12:43 lu-tze kernel: Modules linked in: edd joydev st tun ebtable_filter ebtables ip6table_filter ip6_tables iptable_filter ip_tables x_tables af_packet bridge stp llc iscsi_ibft iscsi_boot_sysfs xfs libcrc32c blktap hid_generic blktap2 usbhid pciback btrfs ipmi_ssif raid456 async_raid6_recov async_memcpy async_pq async_xor xor async_tx raid6_pq usbbk xen_scsibk coretemp crct10dif_pclmul crc32_pclmul crc32c_intel 8250_fintek tg3 ptp pps_core iTCO_wdt aesni_intel iTCO_vendor_support pcspkr aes_x86_64 lrw gf128mul glue_helper ablk_helper cryptd serio_raw 8250 serial_core blkbk blkback_pagemap domctl libphy netbk xenbus_be ipmi_si ipmi_msghandler ie31200_edac hpilo thermal hpwdt edac_core lpc_ich shpchp mfd_core gntdev acpi_power_meter button processor thermal_sys hwmon evtchn raid1 md_mod sr_mod cdrom mgag200
Aug 17 16:12:43 lu-tze kernel: syscopyarea sysfillrect sysimgblt i2c_algo_bit drm_kms_helper xhci_pci xhci_hcd uhci_hcd ttm drm ehci_pci ehci_hcd i2c_core mpt2sas usbcore raid_class usb_common scsi_transport_sas dm_mod sg
Aug 17 16:12:43 lu-tze kernel: CPU: 1 PID: 50 Comm: kswapd0 Not tainted 4.1.27-27-xen #1
Aug 17 16:12:43 lu-tze kernel: Hardware name: HP ProLiant ML10 v2, BIOS J10 02/02/2015
Aug 17 16:12:43 lu-tze kernel: task: ffff8803cbc625d0 ti: ffff8803cbc64000 task.ti: ffff8803cbc64000
Aug 17 16:12:43 lu-tze kernel: RIP: e030:<0000000000000000>] < (null)>] (null)
Aug 17 16:12:43 lu-tze kernel: RSP: e02b:ffff8803cbc67b50 EFLAGS: 00010206
Aug 17 16:12:43 lu-tze kernel: RAX: 4000000001000000 RBX: ffff8803da3614c0 RCX: 4000000001000200
Aug 17 16:12:43 lu-tze kernel: RDX: ffff8803da363480 RSI: 0000000000000000 RDI: ffff8803da3614c0
Aug 17 16:12:43 lu-tze kernel: RBP: ffff8803d688a000 R08: 4000000001000000 R09: 0000000000000000
Aug 17 16:12:43 lu-tze kernel: R10: 0000000000007ff0 R11: 0000000000000a3d R12: ffff8803da3614c0
Aug 17 16:12:43 lu-tze kernel: R13: 0000000000000000 R14: ffff8803cd556520 R15: ffff8803cd556510
Aug 17 16:12:43 lu-tze kernel: FS: 00007fb22e762800(0000) GS:ffff8803d4680000(0000) knlGS:ffff8803d4680000
Aug 17 16:12:43 lu-tze kernel: CS: e033 DS: 0000 ES: 0000 CR0: 0000000080050033
Aug 17 16:12:43 lu-tze kernel: CR2: 0000000000000000 CR3: 00000002a6365000 CR4: 0000000000042660
Aug 17 16:12:43 lu-tze kernel: Stack:
Aug 17 16:12:43 lu-tze kernel: ffffffff8013c10d 000000005230e000 000000005230f000 0000000000000000
Aug 17 16:12:43 lu-tze kernel: 0000000000000080 ffff8803cd556520 ffff8803da3614c0 ffff8803d688a000
Aug 17 16:12:43 lu-tze kernel: 0000000000000000 4000000001000000 ffff8803cd556520 ffff8803cd556510
Aug 17 16:12:43 lu-tze kernel: Call Trace:
Aug 17 16:12:43 lu-tze kernel: Inexact backtrace:
Aug 17 16:12:43 lu-tze kernel: <ffffffff8013c10d>] ? free_pages_prepare+0x1dd/0x2d0
Aug 17 16:12:43 lu-tze kernel: <ffffffff8013d056>] ? free_hot_cold_page+0x26/0x1a0
Aug 17 16:12:43 lu-tze kernel: <ffffffffa027e306>] ? ttm_pages_put+0x36/0x60 [ttm]
Aug 17 16:12:43 lu-tze kernel: <ffffffffa027e911>] ? ttm_page_pool_free+0x121/0x1b0 [ttm]
Aug 17 16:12:43 lu-tze kernel: <ffffffffa027ea2b>] ? ttm_pool_shrink_scan+0x8b/0xc0 [ttm]
Aug 17 16:12:43 lu-tze kernel: <ffffffff801469cf>] ? shrink_slab.part.37+0x1ff/0x3d0
Aug 17 16:12:43 lu-tze kernel: <ffffffff8014ac57>] ? shrink_zone+0x277/0x290
Aug 17 16:12:43 lu-tze kernel: <ffffffff8014ba38>] ? kswapd+0x508/0x9e0
Aug 17 16:12:43 lu-tze kernel: <ffffffff8014b530>] ? mem_cgroup_shrink_node_zone+0x180/0x180
Aug 17 16:12:43 lu-tze kernel: <ffffffff80066761>] ? kthread+0xc1/0xe0
Aug 17 16:12:43 lu-tze kernel: <ffffffff800666a0>] ? do_raw_spin_unlock+0x20/0x20
Aug 17 16:12:43 lu-tze kernel: <ffffffff805f274d>] ? ret_from_fork+0x4d/0x77
Aug 17 16:12:43 lu-tze kernel: <ffffffff800666a0>] ? do_raw_spin_unlock+0x20/0x20
Aug 17 16:12:43 lu-tze kernel: Code: Bad RIP value.
Aug 17 16:12:43 lu-tze kernel: RIP < (null)>] (null)
Aug 17 16:12:43 lu-tze kernel: RSP <ffff8803cbc67b50>
Aug 17 16:12:43 lu-tze kernel: CR2: 0000000000000000
Aug 17 16:12:43 lu-tze kernel: — end trace b8d7f12ffa1171e0 ]—

A little more info. I have determined that the problem that generated the OOPS above is only happening in the XEN kernel. I have been running the machine in desktop mode fpr several days and the system is stable. I am going to run the guests on KVM instead and see if I can keep the machine up and running.

Joe