Re: error installing stand/i386/loader_4th in HEAD

On Thu, 9 Sep 2021 10:37:30 -0600
Warner Losh <imp@bsdimp.com> wrote:

> On Thu, Sep 9, 2021, 10:03 AM Gary Jennejohn <gljennjohn@gmail.com> wrote:
> 
> > On Thu, 9 Sep 2021 09:07:08 -0600
> > Warner Losh <imp@bsdimp.com> wrote:
> >  
> > > On Thu, Sep 9, 2021, 7:11 AM Gary Jennejohn <gljennjohn@gmail.com>  
> > wrote:  
> > >  
> > > > I got a very strange error while trying to make installworld on HEAD:
> > > >
> > > > stand/i386/loader_4th (install)
> > > > /tmp/install.rIhHqnPC/sh: cc not found
> > > > *** Error code 127
> > > >
> > > > This from a tree which I updated at 08:37 UTC.  According to uname -a
> > > > the tree is at commit main-n249276-fc29a8b9d95.
> > > >
> > > > I booted the new kernel from this tree.
> > > >
> > > > I've never seen this error before.  Why in the world would install need
> > > > cc?
> > > >  
> > >
> > > The usual cause of this error is time skew. There were some build issues
> > > that I've fixed that I was sure I merged to stable/13, but I'll check
> > > when I get home.
> > >  
> >
> > Thanks.  I'll check the times set under /usr/src and /usr/obj.
> >  
> 
> If you have a reliable way to reproduce this,  I'd be quite interested if
> it isn't a typical "oops" on dates
> 

Well, the dates under /usr/obj are all more recent than under
/usr/src/stand/i386/loader_4th.  Seems like installworld should
not fail the installation based on the dates alone.

Since I'm having probelms with ssh I'm going to update the tree
and do a buildworld.  I'll see what happens after that.

-- 
Gary Jennejohn