[OpenAFS] Issue report: "invalid argument" with 1.5.68 but not with 1.4.11

Simon Wilkinson sxw@inf.ed.ac.uk
Mon, 21 Dec 2009 13:44:25 +0000


On 21 Dec 2009, at 03:30, Adam Megacz wrote:

>
> I just upgraded one of my machines to the 1.5.68 client, and I've been
> experiencing some mysterious behavior when using git.  The one
> reproducible oddity is this:

Firstly, bug reports should really be sent to openafs- 
bugs@openafs.org. This means both that they don't get lost in the  
mailing list, and that they're tracked and recorded - both so we can  
be sure to fix them, and so we can refer to them in future when we're  
trying to work out why something is the way it is! I've sent your  
original message into RT - it's #125954

Secondly, I'm unable to reproduce this with the current git master,  
running on Fedora 11 (with Fedora's 2.6.29.4 kernel). I also can't see  
anywhere obvious in the code that returns EINVAL - which makes it  
likely that there's some interaction between client and fileserver  
coming into play.

To debug this further, we really need some more information. Could you  
provide
   *) The kernel's message log with any messages output when the git  
clone failed
   *) A fstrace log of the git clone operation (see http://blob.inf.ed.ac.uk/sxw/2009/01/24/using-fstrace-to-debug-the-afs-cache-manager/ 
  )
   *) A packet trace of traffic between the client and the fileserver,  
showing the last 10 or so RPCs before the git clone fails

Thanks,

Simon.