"Martin 'MC' Brown" <[EMAIL PROTECTED]> wrote:
> > Now with Build 24, the conversion if vold from a /etc/init.d/volfs
> > based
> > application to a "svcadm" managed application limits the results of
> > the
> > bug and it is no longer a show stopper bug as the SchilliX Boot CD
> > is no
> > longer ejected. The reason is that the CD is no longer ejected with
> > Build 24
> > is that smf knows that a dependency of vold is not running and thus
> > does
> > not start vold.
>
> I don't have an issue with Build 22, and I don't remember the issue
> in Build 18 either; it's always worked fine for me throughout all the
> builds, and on both SPARC and x86 platforms too.
>
> I also don't have any problems with inetd not kicking smserver into
> touch, which is probably why my vold is working fine... :)
Well, if you look at newer OpenSolaris version you will see that svc.startd
corrupts it's own memory and libumem writes tons of debug messages to the
console.
It may be that by accident the Sun compiles _Sun_ _Solaris_ does not have the
problems but this week, I started to build a SchilliX-0.2 ISO from the
Sun supplied BFU archies and it turns out that they also have the problem.
I am currently compiling Build 24 and plan to hack inetd today in order to
try to fix the problems or at least find a work around.
Jörg
--
EMail:[EMAIL PROTECTED] (home) Jörg Schilling D-13353 Berlin
[EMAIL PROTECTED] (uni)
[EMAIL PROTECTED] (work) Blog: http://schily.blogspot.com/
URL: http://cdrecord.berlios.de/old/private/ ftp://ftp.berlios.de/pub/schily
_______________________________________________
opensolaris-discuss mailing list
[email protected]
