Re: [PATCH OSSTEST v2 5/5] ms-flights-summary: Produce an HTML report of all active flights

[Ian Campbell <ian.campbell@citrix.com>]

On Fri, 2015-07-31 at 16:50 +0100, Ian Jackson wrote:
> Ian Campbell writes ("[PATCH OSSTEST v2 5/5] ms-flights-summary: Produce 
> an HTML report of all active flights"):
> > This could surely use better Perl and produce better output, however
> > I'm sending it now because it would be useful for further development
> > if some or all of the preceding patches could go into production and
> > this serves as an example of why I think I want them.
> 
> I think it's pretty good actually.  I have some minor stylistic
> comments.  I haven't inspected the output, but as you say we can
> improve it later.

Thanks, I actually posted a v3 shortly after which rewrote a lot of this.

I'll try and apply those of your comments which are relevant to the v3
though.

> > diff --git a/ms-planner b/ms-planner
> > index f38f05b..35d430b 100755
> > --- a/ms-planner
> > +++ b/ms-planner
> > @@ -289,6 +289,7 @@ END
> >  	    $info= "rogue task $arow->{subtask}";
> >  	}
> >  	$plan->{Allocations}{$reskey}= {
> > +	    # Can we find a Job here?
> >              Task => $arow->{owntaskid},
> >  	    Info => $info,
> 
> I don't understand this comment.

This didn't really belong in this patch. It was in relation to this
paragraph from patch #1's commit message:

    cmd_reset does not include a ->Job for jobs which are "(preparing)",
    corresponding to a job which is going to use a shared host which is
    currently being installed by another job. I was unable to figure out a
    way to include these.

The upshot is that N-1 of the N jobs which are going to share a host sit in
the "Anonymous Jobs" list until the first job has finished with the actual
installation of the host, instead of being associated with the proper job.

I wasn't able to figure out how to get the original Job to include in those
records in the allocations.

Ian.