[openib-general] Solaris IPoIB MTU with OpenSM

Tue Feb 15 13:36:46 PST 2005

On Tue, 2005-02-15 at 16:36, Nitin Hande wrote:
> I have a hunch for whats happening here,

Glad to hear this as I don't have a clue :-)

>  but before I jump into any
> conclusions, I am seeing some other issue between Solaris IPoIB driver
> and OpenSM. After joining the Broadcast group, the PathRecord Response
> coming from OpenSM signals an error with Invalid GUID. I wonder why,

I missed this as my decode is manual being 1.0.a style. I will look at
this and my traces and get back to you later on this.

> Here is the mad trace:
> 
> 
> Outgoing MAD:
>         BaseVersion: 0x1
>         MgmtClass: 0x3 - SubnAdm
>         ClassVersion: 0x2
>         R_Method: 0x2 - SubnAdmSet()
>         Status: 0x0 - NO_ERROR
>         ClassSpecific: 0x0
>         TransactionID: 0x97651d100000034
>         AttributeID: 0x38 - SA_MCMEMBERRECORD_ATTRID
> 
>      0  1  2  3  4  5  6  7  8  9  a  b  c  d  e  f 0123456789abcdef
>  0: 01 03 02 02 00 00 00 00 09 76 51 d1 00 00 00 34  .........vQ....4
> 10: 00 38 00 00 ff ff ff ff 00 00 00 00 00 00 00 00  .8..............
> 20: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00  ................
> 30: 00 00 00 00 00 01 b0 c7 ff 12 40 1b ff ff 00 00  .......... at .....
> 40: 00 00 00 00 ff ff ff ff fe 80 00 00 00 00 00 00  ................
> 50: 00 02 c9 01 09 76 51 d1 00 00 0b 1b 00 00 00 00  .....vQ.........
> 60: ff ff 00 00 00 00 00 00 21 00 00 00 00 00 00 00  ........!.......
> 70: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00  ................
> 80: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00  ................
> 90: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00  ................
> a0: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00  ................
> b0: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00  ................
> c0: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00  ................
> d0: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00  ................
> e0: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00  ................
> f0: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00  ................
> 
> Incoming MAD:
>         BaseVersion: 0x1
>         MgmtClass: 0x3 - SubnAdm
>         ClassVersion: 0x2
>         R_Method: 0x81 -
>         Status: 0x0 - NO_ERROR
>         ClassSpecific: 0x0
>         TransactionID: 0x97651d100000034
>         AttributeID: 0x38 - SA_MCMEMBERRECORD_ATTRID
> 
>     0  1  2  3  4  5  6  7  8  9  a  b  c  d  e  f  0123456789abcdef
>  0: 01 03 02 81 00 00 00 00 09 76 51 d1 00 00 00 34  .........vQ....4
> 10: 00 38 00 00 ff ff ff ff 00 00 00 00 00 00 00 00  .8..............
> 20: 00 00 00 00 00 00 00 00 00 00 00 00 00 07 00 00  ................
> 30: 00 00 00 00 00 01 b0 c7 ff 12 40 1b ff ff 00 00  .......... at .....
> 40: 00 00 00 00 ff ff ff ff fe 80 00 00 00 00 00 00  ................
> 50: 00 02 c9 01 09 76 51 d1 00 00 0b 1b c0 00 04 00  .....vQ.........
> 60: ff ff 03 12 00 00 00 00 21 00 00 00 00 00 00 00  ........!.......
> 70: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00  ................
> 80: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00  ................
> 90: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00  ................
> a0: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00  ................
> b0: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00  ................
> c0: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00  ................
> d0: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00  ................
> e0: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00  ................
> f0: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00  ................
> f0: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00  ................
> 
> OpenSM responds positively to MCMEMBERRECORD and then:
> 
> Outgoing MAD:
>         BaseVersion: 0x1
>         MgmtClass: 0x3 - SubnAdm
>         ClassVersion: 0x2
>         R_Method: 0x12 - SubnAdmGetTable()
>         Status: 0x0 - NO_ERROR
>         ClassSpecific: 0x0
>         TransactionID: 0x97651d10000003a
>         AttributeID: 0x35 - SA_PATHRECORD_ATTRID
> 
>     0  1  2  3  4  5  6  7  8  9  a  b  c  d  e  f  0123456789abcdef
>  0: 01 03 02 12 00 00 00 00 09 76 51 d1 00 00 00 3a  .........vQ....:
> 10: 00 35 00 00 ff ff ff ff 00 00 00 00 00 00 00 00  .5..............
> 20: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00  ................
> 30: 00 00 00 00 00 00 18 0c 00 00 00 00 00 00 00 00  ................
> 40: ff 12 40 1b ff ff 00 00 00 00 00 00 ff ff ff ff  .. at .............
> 50: fe 80 00 00 00 00 00 00 00 02 c9 01 09 76 51 d1  .............vQ.
> 60: 00 00 00 00 00 00 00 00 00 81 00 00 00 00 00 00  ................
> 70: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00  ................
> 80: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00  ................
> 90: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00  ................
> a0: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00  ................
> b0: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00  ................
> c0: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00  ................
> d0: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00  ................
> e0: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00  ................
> f0: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00  ................
> 
> Incoming MAD:
>         BaseVersion: 0x1
>         MgmtClass: 0x3 - SubnAdm
>         ClassVersion: 0x2
>         R_Method: 0x92 -
>         Status: 0x500 -
>         ClassSpecific: 0x0
>         TransactionID: 0x97651d10000003a
>         AttributeID: 0x35 - SA_PATHRECORD_ATTRID
> 
>     0  1  2  3  4  5  6  7  8  9  a  b  c  d  e  f  0123456789abcdef
>  0: 01 03 02 92 05 00 00 00 09 76 51 d1 00 00 00 3a  .........vQ....:
> 10: 00 35 00 00 ff ff ff ff 00 00 00 00 00 00 00 00  .5..............
> 20: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00  ................
> 30: 00 00 00 00 00 00 18 0c 00 00 00 00 00 00 00 00  ................
> 40: ff 12 40 1b ff ff 00 00 00 00 00 00 ff ff ff ff  .. at .............
> 50: fe 80 00 00 00 00 00 00 00 02 c9 01 09 76 51 d1  .............vQ.
> 60: 00 00 00 00 00 00 00 00 00 81 00 00 00 00 00 00  ................
> 70: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00  ................
> 80: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00  ................
> 90: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00  ................
> a0: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00  ................
> b0: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00  ................
> c0: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00  ................
> d0: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00  ................
> e0: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00  ................
> f0: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00  ................
> [I have manually formatted the output, hope it comes clean]

It came thru just fine.

> Hal, Any idea ???. 

Not yet but stay tuned.

> BTW, I havent applied your Sunday nights patch to the
> OpenSM while I try to reproduce the 252 byte MTU size problem. 

That's to set the selectors to exactly in the MCMemberRecord response.

> thanks
likewise

-- Hal

> Nitin
> 
> 
> 
> 
> 
> 
> On Tue, 2005-02-15 at 11:12, Hal Rosenstock wrote:
> > Hi,
> > 
> > Unfortunately, the Solaris 10 IPoIB MTU with OpenSM is back to the
> > maximum size of 252 again :-( I'm not sure whether this was ever really
> > fixed although I do now see the packets indicating an exact MTU of 4
> > (2048 bytes). I'm not sure what Solaris doesn't like about the OpenSM
> > response to the MCMemberRecord.
> > 
> > -- Hal
> > 
> > _______________________________________________
> > openib-general mailing list
> > openib-general at openib.org
> > http://openib.org/mailman/listinfo/openib-general
> > 
> > To unsubscribe, please visit http://openib.org/mailman/listinfo/openib-general
> 