[Users] odd subset manager behavior

[Narayan Desai]

Here are the full logs from that execution of the subnet manager. It
looks like it was functioning properly, intermittently, right? (I'm
assuming since the intervals are irregular, and the fabric was
continuing to function, it seems like the SM was probably more or less
working)

Ira, what level of debug should I be setting? I'm running opensm with -D 0x40.
thanks
 -nld


May 10 14:14:19 856909 [512F9700] 0x80 -> OpenSM 3.3.15
May 10 14:14:19 872752 [512F9700] 0x80 -> Entering DISCOVERING state
May 10 14:14:20 844905 [4B0C9700] 0x80 -> Entering MASTER state
May 10 14:14:21 604958 [4B0C9700] 0x80 -> SUBNET UP
May 10 18:19:41 103790 [4B0C9700] 0x80 -> Errors during initialization
May 10 23:19:18 147739 [4B0C9700] 0x80 -> Errors during initialization
May 11 00:09:29 119760 [4B0C9700] 0x80 -> Errors during initialization
May 11 01:39:44 171770 [4B0C9700] 0x80 -> Errors during initialization
May 11 01:40:35 187802 [4B0C9700] 0x80 -> Errors during initialization
May 11 04:20:43 199806 [4B0C9700] 0x80 -> Errors during initialization
May 11 06:17:38 399745 [4B0C9700] 0x80 -> Errors during initialization
May 11 07:20:36 183738 [4B0C9700] 0x80 -> Errors during initialization
May 11 07:42:16 207784 [4B0C9700] 0x80 -> Errors during initialization
May 11 08:07:02 199763 [4B0C9700] 0x80 -> Errors during initialization
May 11 12:00:18 243758 [4B0C9700] 0x80 -> Errors during initialization
May 11 14:05:18 295779 [4B0C9700] 0x80 -> Errors during initialization
May 11 15:38:14 299780 [4B0C9700] 0x80 -> Errors during initialization
May 11 16:03:30 503760 [4B0C9700] 0x80 -> Errors during initialization
May 11 16:16:01 295739 [4B0C9700] 0x80 -> Errors during initialization
May 11 17:37:38 303782 [4B0C9700] 0x80 -> Errors during initialization
May 11 17:38:42 327748 [4B0C9700] 0x80 -> Errors during initialization
May 11 22:01:49 339753 [4B0C9700] 0x80 -> Errors during initialization
May 11 22:02:19 379695 [4B0C9700] 0x80 -> Errors during initialization
May 12 03:04:48 403771 [4B0C9700] 0x80 -> Errors during initialization
May 12 03:31:22 403752 [4B0C9700] 0x80 -> Errors during initialization
May 12 03:44:41 415740 [4B0C9700] 0x80 -> Errors during initialization
May 12 10:35:59 475849 [4B0C9700] 0x80 -> Errors during initialization
May 12 10:41:39 467814 [4B0C9700] 0x80 -> Errors during initialization
May 12 11:27:02 471770 [4B0C9700] 0x80 -> Errors during initialization
May 12 13:34:57 467829 [4B0C9700] 0x80 -> Errors during initialization
May 12 14:33:43 487784 [4B0C9700] 0x80 -> Errors during initialization
May 12 14:35:56 507728 [4B0C9700] 0x80 -> Errors during initialization
May 12 14:39:56 499739 [4B0C9700] 0x80 -> Errors during initialization
May 12 15:37:38 687764 [4B0C9700] 0x80 -> Errors during initialization
May 12 18:06:41 531744 [4B0C9700] 0x80 -> Errors during initialization
May 12 18:32:50 551773 [4B0C9700] 0x80 -> Errors during initialization
May 12 18:54:21 511818 [4B0C9700] 0x80 -> Errors during initialization
May 12 19:16:14 527799 [4B0C9700] 0x80 -> Errors during initialization
May 13 01:34:58 583765 [4B0C9700] 0x80 -> Errors during initialization
May 13 02:25:13 615760 [4B0C9700] 0x80 -> Errors during initialization
May 13 05:16:22 611734 [4B0C9700] 0x80 -> Errors during initialization
May 13 05:22:09 603862 [4B0C9700] 0x80 -> Errors during initialization
May 13 05:56:45 851842 [4B0C9700] 0x80 -> Errors during initialization
May 13 06:15:47 851775 [4B0C9700] 0x80 -> Errors during initialization
May 13 13:30:24 290706 [512F9700] 0x80 -> Exiting SM

On Mon, May 13, 2013 at 2:02 PM, Hal Rosenstock
<hal.rosenstock at gmail.com> wrote:
>
>
> On Mon, May 13, 2013 at 2:50 PM, Narayan Desai <narayan.desai at gmail.com>
> wrote:
>>
>> Our subnet manager started producing some weird error messages:
>> May 13 05:16:22 611734 [4B0C9700] 0x80 -> Errors during initialization
>> May 13 05:22:09 603862 [4B0C9700] 0x80 -> Errors during initialization
>> May 13 05:56:45 851842 [4B0C9700] 0x80 -> Errors during initialization
>> May 13 06:15:47 851775 [4B0C9700] 0x80 -> Errors during initialization
>>
>> The subnet manager was actually up and running prior to this, and had
>> successfully configured the network:
>> May 10 14:14:19 856909 [512F9700] 0x80 -> OpenSM 3.3.15
>> May 10 14:14:19 872752 [512F9700] 0x80 -> Entering DISCOVERING state
>> May 10 14:14:20 844905 [4B0C9700] 0x80 -> Entering MASTER state
>> May 10 14:14:21 604958 [4B0C9700] 0x80 -> SUBNET UP
>> May 10 18:19:41 103790 [4B0C9700] 0x80 -> Errors during initialization
>> May 10 23:19:18 147739 [4B0C9700] 0x80 -> Errors during initialization
>> May 11 00:09:29 119760 [4B0C9700] 0x80 -> Errors during initialization
>>
>> Any clue what is causing this?
>
>
> It means some critical set to configure the subnet failed.
>
> Are there error messages are in the opensm log ?
>
> -- Hal
>
>>
>> thanks.
>>  -nld
>> _______________________________________________
>> Users mailing list
>> Users at lists.openfabrics.org
>> http://lists.openfabrics.org/cgi-bin/mailman/listinfo/users
>
>