[openib-general] [PATCH] validate MADs issued from userspace for spec compliance C13-18.1.1

Hal Rosenstock halr at voltaire.com
Wed Jul 12 10:16:14 PDT 2006


On Wed, 2006-07-12 at 12:41, Sean Hefty wrote:
> Hal Rosenstock wrote:
> >>>and running multiple copies of opensm on different systems.
> >>
> >>Not sure what that would fail. The other SMs should be standbys. I can't
> >>think of what would fail in osmtest off the top of my head but haven't
> >>tried this yet but am now about to.
> 
> I was starting / stopping openSM on different systems soon before running the 
> tests.  

Not sure I quite understand the sequencing.

> Whenever I started it on one of my systems, it would always become the 
> master and reassign LIDs.

Reassigning LIDs is disruptive. However, this should be made to work.

You can avoid this by synchronizing the /var/cache/osm/guid2lid file.

> > I think I see what might be going on. To confirm, can you increase the
> > transaction timeout on the osmtest side and see if that helps ?
> > Something like:
> > 
> > osmtest -f a -t 1000
> 
> Thanks - I'll try to rerun the tests and see if I get similar behavior.
> 
> The failure I was seeing was while running osmtest -f c.

If -c fails, there's no use in running a -a. Your failure is a different
scenario than I was envisioning.

>   If I'm remembering 
> correctly, the second query of that test wasn't getting a response.

Can you run with -V and send me the output ? I want to recreate this so
I understand what is going on.

-- Hal

> - Sean





More information about the general mailing list