On Tue, 29 Mar 2005 23:14:13 +0200 christophe varoqui <christophe.varoqui@xxxxxxx> wrote > That was the idea, yes. > In early design, multipath was an all-or-nothing thing. > > As it evolved to trim its scope to specific multipaths, I > thought it was > not necessary to keep it from running in parralel on different > multipaths. > > I eyeballed the code and didn't see a justification to keep it running > alone. > > Did I miss something important there ? Not sure. I agree that servicing different maps in parallel is a good thing to do ... but what keeps multiple invocations of multipath from running concurrently on the same multipath mapped device? I think nothing does and in this case, I think there is nothing that guarantees a time ordering of path related actions. Not clear if this is a problem though.
