[openib-general] [Bug 122] New: mad layer problem

[bugzilla-daemon at openib.org]

http://openib.org/bugzilla/show_bug.cgi?id=122

           Summary: mad layer problem
           Product: OpenFabrics Linux
           Version: gen2
          Platform: All
        OS/Version: Other
            Status: NEW
          Severity: blocker
          Priority: P2
         Component: IB Core
        AssignedTo: sean.hefty at intel.com
        ReportedBy: eli at mellanox.co.il
                CC: bugzilla at openib.org


We were running polygraph http://freshmeat.net/projects/polygraph/ over ipoib
and at some time ipoib connectivity was lost. When looking at the state of the
machines (two machines conected through a switch) I noticed that on on one of
the machines, I could not run any program that uses mads. Specifically I tried
sminfo and then opensm, both got stuck. I assume what happend is that at some
time the kernel refreshed its arp cache and at that time there was already a
problem in sending mads so the kernel could not resolve the address so ipoib
connectivity got lost.




------- You are receiving this mail because: -------
You are on the CC list for the bug, or are watching someone who is.