[ofa-general] IB network problem - "Unknown remote side for node"

[Hal Rosenstock]

On Tue, 2008-06-17 at 11:07 -0400, Michael Di Domenico wrote:
> Can anyone tell me what i likely causing this?  The SM seems to be in
> a loop entering/exiting,

Are you saying the SM exits all on its own ? If so, is there some script
that restarts it ?

>  and the "Unknown remote side" comes up with a different list of port
> each time it cycles.

That's weird. How different are the ports ?

Unknown remote side is due to a port which is not physically DOWN but
perhaps unresponsive (or slow response).

>   I though it was bad cables, but since it keeps changing, that seems
> unlikely.
> 
> Thanks
> - Michael
> 
> 
> Jun 17 04:05:51 681669 [AAF0D060] -> OpenSM Rev:openib-2.0.5 OpenIB
> svn 9905

This looks close to OFED 1.1. Any chance of updating to a more recent
OpenSM version ?

> Jun 17 04:05:51 681723 [AAF0D060] -> OpenSM Rev:openib-2.0.5 OpenIB
> svn 9905
> Jun 17 04:05:51 686837 [AAF0D060] -> osm_vendor_bind: Binding to port
> 0x2c9030000792e
> Jun 17 04:05:51 689709 [AAF0D060] -> osm_vendor_bind: Binding to port
> 0x2c9030000792e
> Jun 17 04:05:52 785232 [46409940] -> osm_drop_mgr_process: ERR 0108:
> Unknown remote side for node 0x000b8cffff00510d port 4. Adding to
> light sweep sampling list
> Jun 17 04:05:52 785273 [46409940] -> Directed Path Dump of 4 hop path:
>                 Path = [0][2][3][11][13]
> Jun 17 04:05:52 785281 [46409940] -> osm_drop_mgr_process: ERR 0108:
> Unknown remote side for node 0x000b8cffff00510d port 6. Adding to
> light sweep sampling list

Can you check the ports indicated and the switch to see if they are
responsive ?

-- Hal

> Jun 17 04:05:52 785290 [46409940] -> Directed Path Dump of 4 hop path:
>                 Path = [0][2][3][11][13]
> Jun 17 04:05:52 785296 [46409940] -> osm_drop_mgr_process: ERR 0108:
> Unknown remote side for node 0x000b8cffff00510d port 8. Adding to
> light sweep sampling list
> Jun 17 04:05:52 785305 [46409940] -> Directed Path Dump of 4 hop path:
>                 Path = [0][2][3][11][13]
> Jun 17 04:05:52 785312 [46409940] -> osm_drop_mgr_process: ERR 0108:
> Unknown remote side for node 0x000b8cffff00510d port 10. Adding to
> light sweep sampling list
> Jun 17 04:05:52 785319 [46409940] -> Directed Path Dump of 4 hop path:
>                 Path = [0][2][3][11][13]
> Jun 17 04:05:52 785325 [46409940] -> osm_drop_mgr_process: ERR 0108:
> Unknown remote side for node 0x000b8cffff00510d port 12. Adding to
> light sweep sampling list
> Jun 17 04:05:52 785334 [46409940] -> Directed Path Dump of 4 hop path:
>                 Path = [0][2][3][11][13]
> Jun 17 04:05:52 785352 [46409940] -> osm_drop_mgr_process: ERR 0108:
> Unknown remote side for node 0x000b8cffff005118 port 8. Adding to
> light sweep sampling list
> Jun 17 04:05:52 785359 [46409940] -> Directed Path Dump of 5 hop path:
>                 Path = [0][2][3][11][13][4]
> Jun 17 04:05:52 785424 [46409940] -> osm_drop_mgr_process: ERR 0108:
> Unknown remote side for node 0x000b8cffff00507a port 8. Adding to
> light sweep sampling list
> Jun 17 04:05:52 785431 [46409940] -> Directed Path Dump of 5 hop path:
>                 Path = [0][2][3][11][13][6]
> Jun 17 04:05:52 785444 [46409940] -> osm_drop_mgr_process: ERR 0108:
> Unknown remote side for node 0x000b8cffff005094 port 8. Adding to
> light sweep sampling list
> Jun 17 04:05:52 785454 [46409940] -> Directed Path Dump of 2 hop path:
>                 Path = [0][2][3]
> Jun 17 04:05:52 785467 [46409940] -> osm_drop_mgr_process: ERR 0108:
> Unknown remote side for node 0x000b8cffff005095 port 10. Adding to
> light sweep sampling list
> Jun 17 04:05:52 785475 [46409940] -> Directed Path Dump of 5 hop path:
>                 Path = [0][2][3][11][8][2]
> Jun 17 04:05:52 785501 [46409940] -> osm_drop_mgr_process: ERR 0108:
> Unknown remote side for node 0x000b8cffff0050a0 port 8. Adding to
> light sweep sampling list
> Jun 17 04:05:52 785509 [46409940] -> Directed Path Dump of 5 hop path:
>                 Path = [0][2][3][11][13][8]
> Jun 17 04:05:52 785536 [46409940] -> osm_drop_mgr_process: ERR 0108:
> Unknown remote side for node 0x000b8cffff0050af port 8. Adding to
> light sweep sampling list
> Jun 17 04:05:52 785545 [46409940] -> Directed Path Dump of 5 hop path:
>                 Path = [0][2][3][11][13][C]
> Jun 17 04:05:52 785553 [46409940] -> osm_drop_mgr_process: ERR 0108:
> Unknown remote side for node 0x000b8cffff0050b0 port 8. Adding to
> light sweep sampling list
> Jun 17 04:05:52 785561 [46409940] -> Directed Path Dump of 5 hop path:
>                 Path = [0][2][3][11][13][A]
> Jun 17 04:05:52 785602 [46409940] -> osm_drop_mgr_process: ERR 0108:
> Unknown remote side for node 0x000b8cffff0050c1 port 4. Adding to
> light sweep sampling list
> Jun 17 04:05:52 785631 [46409940] -> Directed Path Dump of 4 hop path:
>                 Path = [0][2][3][11][D]
> Jun 17 04:05:52 785645 [46409940] -> osm_drop_mgr_process: ERR 0108:
> Unknown remote side for node 0x000b8cffff0050c4 port 4. Adding to
> light sweep sampling list
> Jun 17 04:05:52 785651 [46409940] -> Directed Path Dump of 3 hop path:
>                 Path = [0][2][3][8]
> Jun 17 04:05:52 785662 [46409940] -> osm_drop_mgr_process: ERR 0108:
> Unknown remote side for node 0x000b8cffff0050c6 port 4. Adding to
> light sweep sampling list
> Jun 17 04:05:52 785671 [46409940] -> Directed Path Dump of 5 hop path:
>                 Path = [0][2][3][11][D][4]
> Jun 17 04:05:52 785686 [46409940] -> osm_drop_mgr_process: ERR 0108:
> Unknown remote side for node 0x000b8cffff0050d2 port 2. Adding to
> light sweep sampling list
> Jun 17 04:05:52 785694 [46409940] -> Directed Path Dump of 4 hop path:
>                 Path = [0][2][3][11][8]
> Jun 17 04:05:52 785702 [46409940] -> osm_drop_mgr_process: ERR 0108:
> Unknown remote side for node 0x000b8cffff0050d2 port 12. Adding to
> light sweep sampling list
> Jun 17 04:05:52 785710 [46409940] -> Directed Path Dump of 4 hop path:
>                 Path = [0][2][3][11][8]
> Jun 17 04:05:52 785741 [46409940] -> osm_drop_mgr_process: ERR 0108:
> Unknown remote side for node 0x000b8cffff0050fa port 10. Adding to
> light sweep sampling list
> Jun 17 04:05:52 785748 [46409940] -> Directed Path Dump of 5 hop path:
>                 Path = [0][2][3][11][8][C]
> Jun 17 04:05:52 785774 [46409940] -> Entering MASTER state
> Jun 17 04:05:52 786126 [46409940] -> osm_report_notice: Reporting
> Generic Notice type:3 num:66 from LID:0x0000
> GID:0x000000000000e80f,0x0002c9030000792e
> Jun 17 04:05:52 786254 [46409940] -> osm_report_notice: Reporting
> Generic Notice type:3 num:66 from LID:0x0000
> GID:0x000000000000e80f,0x0002c9030000792e
> Jun 17 04:05:56 691253 [AAF0D060] -> Exiting SM
> 
> _______________________________________________
> general mailing list
> general at lists.openfabrics.org
> http://lists.openfabrics.org/cgi-bin/mailman/listinfo/general
> 
> To unsubscribe, please visit http://openib.org/mailman/listinfo/openib-general