chapwong@chevron.com
2007-Jan-16 22:37 UTC
[Lustre-devel] [Bug 11556] messages in log ROUTER_NOTIFY
Please don''t reply to lustre-devel. Instead, comment in Bugzilla by
using the following link:
https://bugzilla.lustre.org/show_bug.cgi?id=11556
What |Removed |Added
----------------------------------------------------------------------------
Group| |Enterprise Support_CHEVRON
We have a 32 cluster with a 10GE interconnect, that is for MPI only and a 1GE
for i/o. Lustre is mount form the 1GE. This afternoon, the node hung, and we
found this in the log.
Jan 16 14:44:29 oct001 kernel: Lustre: 20:0:(linux-debug.c:96:libcfs_run_upcall
()) Invoked LNET upcall /usr/lib/lustre/lnet_upcall
ROUTER_NOTIFY,146.36.106.50@tcp,down,1168980191
Jan 16 14:44:29 oct001 kernel: Lustre: 20:0:(linux-debug.c:96:libcfs_run_upcall
()) Invoked LNET upcall /usr/lib/lustre/lnet_upcall
ROUTER_NOTIFY,146.36.106.51@tcp,down,1168980191
Jan 16 14:44:29 oct001 kernel: Lustre: 20:0:(linux-debug.c:96:libcfs_run_upcall
()) Invoked LNET upcall /usr/lib/lustre/lnet_upcall
ROUTER_NOTIFY,146.36.106.52@tcp,down,1168980191
Jan 16 14:44:29 oct001 kernel: Lustre: 20:0:(linux-debug.c:96:libcfs_run_upcall
()) Invoked LNET upcall /usr/lib/lustre/lnet_upcall
ROUTER_NOTIFY,146.36.106.53@tcp,down,1168980191
Jan 16 14:44:29 oct001 kernel: Lustre: 20:0:(linux-debug.c:96:libcfs_run_upcall
()) Invoked LNET upcall /usr/lib/lustre/lnet_upcall
ROUTER_NOTIFY,146.36.106.54@tcp,down,1168980191
Jan 16 15:08:54 oct001 syslogd 1.4.1: restart.
The ip addresses are our mds and oss. We are running 1.4.8 with 1 mds and 4 OSS.
Why we see ROUTER_NOTIFY messages?