[openib-general] OpenSM errors question..

Hal Rosenstock halr at voltaire.com
Wed Nov 2 19:58:51 PST 2005


On Wed, 2005-11-02 at 22:24, Troy Benjegerdes wrote:
> What does the following mean? (the ERR 1B11, in particular)
>
> Nov 02 16:18:33 656702 [41001960] -> osm_report_notice: Reporting
> Generic Notice type:4 num:144 from LID:0x001B
> GID:0xfe80000000000000,0x0002c90200402789
> Nov 02 16:18:33 674607 [41802960] -> osm_ucast_mgr_process: Min Hop
> Tables configured on all switches.
> Nov 02 16:18:34 197522 [41802960] -> osm_ucast_mgr_process: Min Hop
> Tables configured on all switches.
> Nov 02 16:19:59 917207 [41802960] -> osm_report_notice: Reporting
> Generic Notice type:3 num:66 from LID:0x001B
> GID:0xfe80000000000000,0x0002c90200402789
> Nov 02 16:19:59 917610 [41001960] -> osm_report_notice: Reporting
> Generic Notice type:3 num:66 from LID:0x001B
> GID:0xfe80000000000000,0x0002c90200402789
> Nov 02 16:19:59 926829 [41001960] -> osm_mcmr_rcv_join_mgrp: ERR 1B11:
> method =SubnAdmSet,scope_state = 0x1, component mask =
> 0x0000000000010083, expected comp mask = 0x00000000000130c7, MGID:
> 0xff12601bffff0000 : 0x0000000000000016
> Nov 02 16:20:01 670893 [41802960] -> osm_mcmr_rcv_join_mgrp: ERR 1B11:
> method =SubnAdmSet,scope_state = 0x1, component mask =
> 0x0000000000010083, expected comp mask = 0x00000000000130c7, MGID:
> 0xff12601bffff0000 : 0x0000000000000002
>
>
> (I got this on an isolated subnet with 3 machines.. two opterons with
> mellanox cards and an IBM with the eHCA card)

That means a join is being attempted to a multicast group which is not
yet created. These are typically groups that you can ignore. They are
benign.

The two above are both IPv6 multicast groups. The first one ends in 0x16
and the second one 0x2. I think those are IGMP and all routers multicast
groups.

-- Hal





More information about the general mailing list