Section 9.41) – HP StorageWorks Scalable File Share User Manual

Page 292

Troubleshooting

9–68

9.41 Determining if the Network ID of a server on a Quadrics or

Myrinet interconnect has been changed

If you relocate a server to a different port on a Quadrics switch, or replace a Myrinet interconnect adapter

on a server, the Network ID of the server is changed. The procedures for dealing with these changes are

described in Section 8 .1.17 and Section 8.1.19. If you do not perform the documented steps, errors similar

to the following will occur on the server where the Network ID changed.

This example is taken from a server connected to a Quadrics switch; the

LustreError

information will be

different on a server connected to a Myrinet interconnect:

LustreError: 6088:0:(lib-move.c:1509:lib_api_put()) skipped 35 similar
messages (ending 50.000 seconds ago)
Lustre: 5860:0:(import.c:276:import_select_connection()) OSC_south2.my.dmn_south-
ost1_south-mds1: continuing with current connection
Lustre: 5860:0:(import.c:276:import_select_connection()) skipped 35 similar
messages (ending 100.000 seconds ago)
Lustre: 5860:0:(import.c:308:import_select_connection()) OSC_south2.my.dmn_south-
ost1_south-mds1: Using connection NID_105_UUID
Lustre: 5860:0:(import.c:308:import_select_connection()) skipped 35 similar
messages (ending 50.000 seconds ago)
LustreError: 6088:0:(qswnal_cb.c:1076:kqswnal_sendmsg()) Can't route to 0x69:
router error -101
LustreError: 6088:0:(qswnal_cb.c:1076:kqswnal_sendmsg()) skipped 35 similar
messages (ending 50.000 seconds ago)
LustreError: 6088:0:(lib-move.c:1509:lib_api_put()) Error sending PUT to 0x69:
19
LustreError: 6088:0:(lib-move.c:1509:lib_api_put()) skipped 35 similar
messages (ending 50.000 seconds ago)
Lustre: 5860:0:(import.c:276:import_select_connection()) OSC_south2.my.dmn_south-
ost1_south-mds1: continuing with current connection
Lustre: 5860:0:(import.c:276:import_select_connection()) skipped 35 similar
messages (ending 50.000 seconds ago)
Lustre: 5860:0:(import.c:308:import_select_connection()) OSC_south2.my.dmn_south-
ost1_south-mds1: Using connection NID_105_UUID
Lustre: 5860:0:(import.c:308:import_select_connection()) skipped 35 similar
messages (ending 50.000 seconds ago)
LustreError: 6088:0:(qswnal_cb.c:1076:kqswnal_sendmsg()) Can't route to 0x69:
router error -101

On a server connected to a Quadrics interconnect, you can determine whether the Network ID has changed

as follows:

Enter the

show server

command for the server, as shown in the following example:

sfs> show server south2

Enter the following command on the server:

# cat /proc/qsnet/ep/rail0/state

Compare the output from the two commands. If the commands report different Network ID values,

you must perform the procedures documented in Section 8.1.17 (including reconfiguring the file

systems).

On a server connected to a Myrinet interconnect, you can determine whether the Network ID has changed