[ofa-general] Infiniband and Opensuse 10.3 stock rpms

[Yevgeny Kliteynik]

Hi Diego,

Diego Guella wrote:
> Hi,
>  
> I've just installed Opensuse 10.3 on a Dell PowerEdge 2850 server, and 
> updated it as of today (Aug 6, 2008).
> I noticed that opensuse has some rpms in its repository, and I was 
> wondering if I could get Infiniband up without installing OFED.
> But, I get this error from opensm:
> -----
> Aug 06 10:15:52 476103 [2340AB00] -> OpenSM Rev:openib-3.0.14
> Aug 06 10:15:52 476206 [2340AB00] -> OpenSM Rev:openib-3.0.14
> Aug 06 10:15:52 485719 [2340AB00] -> osm_vendor_bind: Binding to port 
> 0x2c9020021c9f1
> Aug 06 10:15:52 488535 [2340AB00] -> osm_vendor_open_port: ERR 542C: 
> umad_open_port() failed
> Aug 06 10:15:52 488565 [2340AB00] -> osm_vendor_bind: ERR 5424: Unable 
> to open port 0x2c9020021c9f1
> Aug 06 10:15:52 488578 [2340AB00] -> osm_sm_mad_ctrl_bind: ERR 3118: 
> Vendor specific bind failed
> Aug 06 10:15:52 488593 [2340AB00] -> osm_sm_bind: ERR 2E10: SM MAD 
> Controller bind failed (IB_ERROR)
> Aug 06 10:15:52 488614 [2340AB00] -> osm_sa_mad_ctrl_unbind: ERR 1A11: 
> No previous bind
> Aug 06 10:15:52 488977 [2340AB00] -> Exiting SM
> -----
>  
>  
> What's wrong?

Perhaps you already have opensm on this machine?
ibstat shows that port state is 'Active'.
Port cannot reach this state without opensm - it
should be in 'Init' state w/o opensm.
You can just grep for 'opensm' process, or check
where opensm is running with 'sminfo' tool.

-- Yevgeny


> Here are some more information on this system:
>  
> The Infiniband-related rpms that I have installed are:
> infiniband-diags, kernel-default, libibmad, libibumad, libibverbs, 
> libsdp, ofed-kmp-default, opensm
> -----
> Server19:~ # lsmod
> Module                  Size  Used by
> ib_umad                32792  0
> iptable_filter         19840  0
> ip_tables              37848  1 iptable_filter
> ip6table_filter        19584  0
> ip6_tables             31944  1 ip6table_filter
> x_tables               37000  2 ip_tables,ip6_tables
> microcode              31256  0
> firmware_class         27520  1 microcode
> apparmor               58544  0
> loop                   36356  0
> dm_mod                 77152  0
> ib_ipoib               89160  0
> ib_cm                  51480  1 ib_ipoib
> ib_sa                  57688  2 ib_ipoib,ib_cm
> ipv6                  372600  35 ib_ipoib
> rtc_cmos               25016  0
> rtc_core               38156  1 rtc_cmos
> floppy                 79624  0
> rtc_lib                19968  1 rtc_core
> sr_mod                 33444  0
> iTCO_wdt               28624  0
> serio_raw              24068  0
> cdrom                  52392  1 sr_mod
> iTCO_vendor_support    20740  1 iTCO_wdt
> usbhid                 58160  0
> hid                    43776  1 usbhid
> ff_memless             22536  1 usbhid
> ib_mthca              141540  0
> ib_mad                 54436  4 ib_umad,ib_cm,ib_sa,ib_mthca
> ib_core                76032  6 ib_umad,ib_ipoib,ib_cm,ib_sa,ib_mthca,ib_mad
> e1000                 203200  0
> shpchp                 50716  0
> pci_hotplug            49396  1 shpchp
> e752x_edac             28036  0
> edac_mc                43584  1 e752x_edac
> button                 26528  0
> sg                     53304  0
> ehci_hcd               50956  0
> uhci_hcd               42144  0
> usbcore               156456  4 usbhid,ehci_hcd,uhci_hcd
> sd_mod                 45824  3
> edd                    26760  0
> ext3                  156688  1
> mbcache                26248  1 ext3
> jbd                    89192  1 ext3
> fan                    22792  0
> mptspi                 36112  2
> mptscsih               39680  1 mptspi
> mptbase                73952  2 mptspi,mptscsih
> scsi_transport_spi     43776  1 mptspi
> ata_piix               37636  0
> libata                166800  1 ata_piix
> scsi_mod              176536  7 
> sr_mod,sg,sd_mod,mptspi,mptscsih,scsi_transport_spi,libata
> thermal                36112  0
> processor              59720  1 thermal
> -----
>  
> -----
> Server19:~ # ibstat
> CA 'mthca0'
>         CA type: MT25208
>         Number of ports: 2
>         Firmware version: 5.1.400
>         Hardware version: a0
>         Node GUID: 0x0002c9020021c9f0
>         System image GUID: 0x0002c9020021c9f3
>         Port 1:
>                 State: Active
>                 Physical state: LinkUp
>                 Rate: 20
>                 Base lid: 4
>                 LMC: 0
>                 SM lid: 1
>                 Capability mask: 0x02510a68
>                 Port GUID: 0x0002c9020021c9f1
>         Port 2:
>                 State: Down
>                 Physical state: Polling
>                 Rate: 10
>                 Base lid: 0
>                 LMC: 0
>                 SM lid: 0
>                 Capability mask: 0x02510a68
>                 Port GUID: 0x0002c9020021c9f2
> -----
>  
>  
> Any advice is greatly appreciated.
> Thanks,
> Diego
>  
> 
> 
> ------------------------------------------------------------------------
> 
> _______________________________________________
> 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