On Fri, 2005-09-09 at 18:42 -0700, Gerrit Huizenga wrote:
> On Fri, 09 Sep 2005 18:21:53 PDT, Matthew Helsley wrote:
> > On Fri, 2005-09-09 at 11:34 -0300, Ronan Resende wrote:
> > > For those with the same problem, I fix it in my machine. I would like
> > > the opinion of the maintainers of this project and if it is correct,
> > > add the changes as a patch.
> > > 
> > > For the first problem, when the rules doesn't work for sockets, the
> > > error is in /usr/src/linux/kernel/ckrm/rbce/rbce_core.c
> > > (ckrm_e18_2612_single.patch line 8383):
> > > change:                                       if (*ptr != '\\') {
> > > by:                                               if (*ptr != ':') {
> > > 
> > > Only change '\\' by ':'. Its is because the old interface used 0.0.0.0
> > > \port for socket and they change for 0.0.0.0:port. But this piece
> > > remained.
> > > 
> > > The second problem: Unable to handle kernel paging request at virtual
> > > address.
> > > The error is in /usr/src/linux/kernel/ckrm/ckrm_listenaq.c
> > > (patch2.3.Accept_queue_controller of 2005-06-02):
> > > change:        while( *p != "/" && p != c->name)
> > > by:                while( *p != '/' && p != c->name)
> > > 
> > > The only change is " by '.
> > > 
> > > It is working for me now.
> > > 
> > > Ronan
> > 
> > Ronan,
> > 
> >     Thank you for finding these bugs.
> 
> Agreed - does someone have a patch gen'ed up?  I'm building a forward port
> of e18 now and will hopefully get testing started shortly.  quilt didn't
> quite do the right thing originally with patch; I'm not sure Matt if your
> version did the right thing.  Some of the sched.h patches would misapply
> changes to task_struct to the newer sched_domain struct based on context,

Are you referring to the RBCE-removal patch? Or e18 applied to 2.6.13?
Either way, I definitely was not messing with sched_domains.

> which was wrong.  I think most of the other things were pretty straight
> forward and I used your suggestion to overload the netlink port for
> audit for now, although that isn't a very good long term choice.
> 
> gerrit

Yeah, it's not a long-term choice.

Cheers,
        -Matt Helsley



-------------------------------------------------------
SF.Net email is Sponsored by the Better Software Conference & EXPO
September 19-22, 2005 * San Francisco, CA * Development Lifecycle Practices
Agile & Plan-Driven Development * Managing Projects & Teams * Testing & QA
Security * Process Improvement & Measurement * http://www.sqe.com/bsce5sf
_______________________________________________
ckrm-tech mailing list
https://lists.sourceforge.net/lists/listinfo/ckrm-tech

Reply via email to