make index strangeness after today's cvsup of ports tree

[Matthew Seaman]

On Tue, Jul 15, 2003 at 10:10:54AM -0700, Kent Stewart wrote:
> On Tuesday 15 July 2003 12:59 am, Matthew Seaman wrote:
> > On Mon, Jul 14, 2003 at 06:51:51PM -0700, Kent Stewart wrote:
> > > On Monday 14 July 2003 01:59 pm, Stacey Roberts wrote:
> > > > On Mon, 2003-07-14 at 21:36, Matthew Seaman wrote:
> > > > >     happy-idiot-talk:...ports/x11/xstroke:% make -V PORTSDIR
> > > > >     cd: can't cd to /devel/libgnugetopt
> > > > >     cd: can't cd to /devel/libgnugetopt
> > > > >     cd: can't cd to /devel/libgnugetopt
> > > > >     cd: can't cd to /devel/libgnugetopt
> > > > >     cd: can't cd to /devel/libgnugetopt
> > > > >     cd: can't cd to /devel/libgnugetopt
> > > > >     cd: can't cd to /devel/libgnugetopt
> > > > >     cd: can't cd to /devel/libgnugetopt
> > > > >     cd: can't cd to /devel/libgnugetopt
> > > > >     cd: can't cd to /devel/libgnugetopt
> > > > >     cd: can't cd to /devel/libgnugetopt
> > > > >     cd: can't cd to /devel/libgnugetopt
> > > > >     cd: can't cd to /devel/libgnugetopt
> > > > >     cd: can't cd to /devel/libgnugetopt
> > > > >     ...
> > > >
> > > > Thanks for that explanation.., I imagine this is another one of
> > > > those things that eventually sorts itself out..,
> > >
> > > The other thing that is happening is the following.from a dmesg
> > >
> > > maxproc limit exceeded by uid 0, please see tuning(7) and
> > > login.conf(5). maxproc limit exceeded by uid 0, please see
> > > tuning(7) and login.conf(5). maxproc limit exceeded by uid 0,
> > > please see tuning(7) and login.conf(5).
> > >
> > > You get 3 or 4 of these while all of the above is going on.
> >
> > Yes.  make(1) is invoking itself recursively because of this bug. 
> > The aposite term is "fork bomb".  Eventually you'll run out of system
> > resources and the command will fail.  When Stacey said "sorts itself
> > out", I believe he was talking about getting fixes committed to the
> > ports tree.  He didn't mean that there would be loads of error
> > messages printed out, and then the command would succeed anyhow.
> >
> 
> The are a number of problems with building an INDEX[-5] at this point. 
> The bitch message from Bento seems to be turned off at this moment. 
> With out its constant reminder, things don't seem to get sorted out as 
> fast :).

Hmmm... A lot to do with the recent import of gcc-3.3.1 as the system
compiler in 5-CURRENT, I think.  Seems that gcc just gets stricter and
stricter on the C++ syntax as time goes by.  Still, it shouldn't be as
much effort to fix as during that great long ports freeze before
5.0-RELEASE came out.
 
> The "make index" runs with your latest patch but a "make index" on a 
> 5-current systems shows some of the problems that exist with 
> /usr/ports/print and the epson printers. On a 4-stable system, you 
> don't see the message but you can view an INDEX and the entry for 
> pips2200 and the entries for a number of nearby ports looks strange.

Yes --- a different patch was committed to the ports tree about 7
hours ago now.  However you cut it, that problem is fixed.
 
> On a 5-current system, there are many other messages for other ports.  
> These just stand out above the rest.

4.8-STABLE does a 'make index' pretty smoothly.  No error messages at
all, usually.

	Cheers,

	Matthew
 

-- 
Dr Matthew J Seaman MA, D.Phil.                       26 The Paddocks
                                                      Savill Way
PGP: http://www.infracaninophile.co.uk/pgpkey         Marlow
Tel: +44 1628 476614                                  Bucks., SL7 1TH UK
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 187 bytes
Desc: not available
Url : http://lists.freebsd.org/pipermail/freebsd-questions/attachments/20030715/7725b955/attachment.bin