On Tue, Jul 27, 2010 at 02:10:42PM -0700, Shawn Walker wrote:

> On 07/27/10 11:55 AM, John Levon wrote:
> >On Tue, Jul 27, 2010 at 06:10:32PM +0100, Peter Dennis wrote:
> >
> >>At first blush is looks like a race between the repos starting
> >>up (specifically the one for the on-nightly) and when onu
> >>runs pkg -R / refresh --full (within the function
> >>configure_publishers()).
> >
> >This has been reported elsewhere too. I think, if there is a race here,
> >it's a bug in pkg.depotd - for SMF purposes it's only supposed to return
> >success when the daemon is up and properly running.
> 
> It's a defect in how the service is launched:
> 
>   https://defect.opensolaris.org/bz/show_bug.cgi?id=15761

But onu doesn't start the depot servers via SMF. My point was that it
sounds like the "pkg.depotd" command is returning before the servers are
fully up, which is not best practice for SMF daemons.

regards
john
_______________________________________________
on-ips-dev mailing list
[email protected]
http://mail.opensolaris.org/mailman/listinfo/on-ips-dev

Reply via email to