[lustre-discuss] oss servers crashing

[Kurt Strosahl]

Good Morning,

   Yesterday one of our lustre file servers rebooted several times.  the crash dump showed:

[14333982.153989] Pid: 381367, comm: ll_ost_io01_076 3.10.0-957.10.1.el7_lustre.x86_64 #1 SMP Tue Apr 30 22:18:15 UTC 2019
[14333982.153989] Kernel panic - not syncing: LBUG
[14333982.153990] Call Trace:
[14333982.153993] CPU: 4 PID: 380760 Comm: ll_ost_io01_072 Kdump: loaded Tainted: P           OE  ------------   3.10.0-957.10.1.el7_lustre.x86_64 #1
[14333982.153994] Hardware name: Supermicro Super Server/X11DPL-i, BIOS 3.1 05/21/2019
[14333982.153995] Call Trace:
[14333982.154002]  [<ffffffffbaf62e41>] dump_stack+0x19/0x1b
[14333982.154006]  [<ffffffffbaf5c550>] panic+0xe8/0x21f
[14333982.154018]  [<ffffffffc0ab87cc>] libcfs_call_trace+0x8c/0xc0 [libcfs]
[14333982.154026]  [<ffffffffc0ab88cb>] lbug_with_loc+0x9b/0xa0 [libcfs]
[14333982.154036]  [<ffffffffc0ab887c>] lbug_with_loc+0x4c/0xa0 [libcfs]
[14333982.154096]  [<ffffffffc12dfae0>] tgt_grant_incoming.isra.6+0x570/0x570 [ptlrpc]
[14333982.154174]  [<ffffffffc12dfae0>] tgt_grant_prepare_read+0x0/0x3b0 [ptlrpc]
[14333982.154232]  [<ffffffffc12dfbeb>] tgt_grant_prepare_read+0x10b/0x3b0 [ptlrpc]
[14333982.154297]  [<ffffffffc12dfbeb>] tgt_grant_prepare_read+0x10b/0x3b0 [ptlrpc]
[14333982.154306]  [<ffffffffc15e1ad0>] ofd_preprw+0x450/0x1160 [ofd]

lustre versions:
lustre-resource-agents-2.12.1-1.el7.x86_64
lustre-2.12.1-1.el7.x86_64
kernel-devel-3.10.0-957.10.1.el7_lustre.x86_64
lustre-osd-zfs-mount-2.12.1-1.el7.x86_64
kernel-headers-3.10.0-957.10.1.el7_lustre.x86_64
kernel-3.10.0-957.10.1.el7_lustre.x86_64
lustre-zfs-dkms-2.12.1-1.el7.noarch

Could this be: https://jira.whamcloud.com/browse/LU-12120

w/r,

Kurt J. Strosahl
System Administrator: Lustre, HPC
Scientific Computing Group, Thomas Jefferson National Accelerator Facility
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.lustre.org/pipermail/lustre-discuss-lustre.org/attachments/20200715/361f94c2/attachment.html>