GitLab

Fix tied STDIN and STDOUT with perl 5.8 in libtblog.

Fix a phototype warning in  os_load.in

Creating the groups for main group is enough for testing.

wildly obfuscated rc.d mechanism.

the current port, but my fix is harmless.

is run before samba tries to start. This is harmless, so it will not matter
if its still in place when ports are updated.

editing.  I know, codegen bad, but makes gui development much less
painful.

listing the tools available, so that it's easier to find them all.

check later.

fixed.

<= len(max other one), except for a 'new' flag.

if we work futher on it, we have a good starting point. Make the
existing templates validate with the DTD.

default values out of the template XML file, and gives the user an
interface to fill them in. Passes the new NS file to the beginexp
page for them.

Still pretty basic. Should support some kind of data typing for the
variables, and error checking. But a good start, and enough for what I
want to do with it now.

variables.

PlanetLab software.

real XML. <Foo> is still supported.

I got close to getting it to work by adding this:

	sub FILENO  { my $this = shift; fileno($$this) }
	sub CLOSE   { my $this = shift; close($$this) }

	sub OPEN {
	    my $this = shift;

	    close($$this) if defined(fileno($$this));
	    @_ == 1 ? open($$this, $_[0]) : open($$this, $_[0], $_[1]);
	}

But subprocesses were not seeing the right stdout/stderr after doing
something like:

	open(STDERR, ">> $logname");
	open(STDOUT, ">> $logname");

So, I will let Kevin work on it; I've spent too much time on it
already!

tagged from the 6.0 release. Also lots of changes to the install
scripts to handle inatlling boss/ops/fs on 5.4. All changes are
conditionalized so that things will continue to work on 4.10, but
hopefully we will move away from that before the year is out.

can report it once the connection is successful.

Now decides whether or not to monitor an FD in socket(), rather than
connect(), since we now have to start collecting data before connect()
is called.

Move code to report on connection state changes into inform*()
functions.

Warning: Tested, but not exhaustively.

ports. This *should* continue to work on the 4.10 images as well, and
I even tested that.

Currently the names and path of the 5.4 packages are hardwired into
place. We probably need to add site-vars for this, but going to wait
since there is plenty of work still to do.