[openib-general] multicast join errors

[Hal Rosenstock]

Hi Amith,

On Sat, 2006-01-28 at 16:18, amith rajith mamidala wrote:
> Hi Hal,
> 
> There is only one application running on a node. I am running opensm on
> a different node. I am also listing the other processes I observed on
> doing a "ps":
> 
> root      3564    11  0 Jan26 ?        00:00:00 [ib_cm/0]
> root      3565    11  0 Jan26 ?        00:00:00 [ib_cm/1]
> root      1294    11  0 Jan26 ?        00:00:00 [ib_mad1]
> root      1295    11  0 Jan26 ?        00:00:00 [ib_mad2]
> root      1298    11  0 Jan26 ?        00:00:00 [ib_mad1]
> root      1299    11  0 Jan26 ?        00:00:00 [ib_mad2]
> 
> 
> Thanks,
> Amith
> 
> On 28 Jan 2006, Hal Rosenstock wrote:
> 
> > Hi Amith,
> >
> > On Sat, 2006-01-28 at 12:46, amith rajith mamidala wrote:
> > > Hi,
> > >
> > > I was able to create multicast groups after Hal's fix. But, when I do join
> > > subsequently from the same program I am getting a port_alloc error:
> > >
> > > Jan 28 12:22:12 119632 [AB2223C0] -> osm_vendor_bind: Binding to port
> > > 0x6270510000005.
> > > -I- Created the Multicast Group:
> > >         MGID....................0xff13a01cfe800000 : 0x0000000000000000
> > >         PortGid.................0xfe80000000000000 : 0x0006270510000005
> > >         qkey....................0x0
> > >         Mlid....................0xC002
> > >         ScopeState..............0x21
> > >         Rate....................0x83
> > >         Mtu.....................0x84
> > > Jan 28 12:22:12 140486 [AB2223C0] -> osm_vendor_bind: Binding to port
> > > 0x6270510000005.
> > >
> > > ibwarn: [4057] port_alloc: umad port id 0 is already allocated for mthca0
> > > 1
> > > Jan 28 12:22:12 143240 [AB2223C0] -> osm_vendor_open_port: ERR 542C:
> > > umad_open_port() failed
> > > Jan 28 12:22:12 143253 [AB2223C0] -> osm_vendor_bind: ERR 5424: Unable to
> > > Open Port 0x6270510000005.
> > > Jan 28 12:22:12 143262 [AB2223C0] -> osmv_bind_sa: ERR 5506: Failed to
> > > bind to vendor GSI
> > > Jan 28 12:22:12 143267 [AB2223C0] -> ibmcgrp_bind: ERR 00137: Unable to
> > > bind to SA
> > >
> > > I am trying to trace the source of this error,
> >
> > Is this the only IB application running or are there others (and if so,
> > what else is running) ?


I'm able to do the following on both ports 1 and 2:

ibmcgrp -c -g 0xff13a01cfe800000:0000000000000000 --port_num=1
-I- Creating Multicast Group
-I- MGID 0xff13a01cfe800000:0000000000000000
-I- Port Num:1
Jan 30 09:23:54 980478 [B7F06720] -> osm_vendor_bind: Binding to port 0x8f10403960559.
-I- Created the Multicast Group:
        MGID....................0xff13a01cfe800000 : 0x0000000000000000
        PortGid.................0xfe80000000000000 : 0x0008f10403960559
        qkey....................0x0
        Mlid....................0xC008
        ScopeState..............0x21
        Rate....................0x82
        Mtu.....................0x84
IBMCGRP: PASS

ibmcgrp -c -g 0xff13a01cfe800000:0000000000000000 --port_num=2
-I- Creating Multicast Group
-I- MGID 0xff13a01cfe800000:0000000000000000
-I- Port Num:2
Jan 30 09:24:02 804602 [B7F50720] -> osm_vendor_bind: Binding to port 0x8f10403960559.
-I- Created the Multicast Group:
        MGID....................0xff13a01cfe800000 : 0x0000000000000000
        PortGid.................0xfe80000000000000 : 0x0008f10403960559
        qkey....................0x0
        Mlid....................0xC008
        ScopeState..............0x21
        Rate....................0x82
        Mtu.....................0x84

The only difference I see is in the rate but that wouldn't cause the
error you are seeing. 

Can you describe your scenario better so I can recreate it to see what
is going on ?

Thanks.

-- Hal

> > -- Hal
> >
> > > Thanks,
> > > Amith
> > >
> > > _______________________________________________
> > > 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
> >
>