Skip to Content.
Sympa Menu

sm-discuss - Re: [SM-Discuss] Proposal: gaze newer last_update

sm-discuss AT lists.ibiblio.org

Subject: Public SourceMage Discussion List

List archive

Chronological Thread  
  • From: Seth Alan Woolley <seth AT positivism.org>
  • To: sm-discuss AT lists.ibiblio.org
  • Subject: Re: [SM-Discuss] Proposal: gaze newer last_update
  • Date: Fri, 21 Jan 2005 13:25:23 -0800

On Fri, Jan 21, 2005 at 08:48:52PM +0100, Arwed von Merkatz wrote:
> On Fri, Jan 21, 2005 at 10:52:15AM -0800, Seth Alan Woolley wrote:
> > On Thu, Jan 20, 2005 at 01:59:49PM -0800, Andrew wrote:
> > > On Thu, Jan 20, 2005 at 11:29:32AM -0800, Seth Alan Woolley wrote:
> > > > If you're going to do work on this, I think that we should make a
> > > > distinction between last compile and last update.
> > > >
> > > > With afrayedknot's new tablet support we can know the difference now
> > > > and
> > > > head -n 1 the tablet history entry to get a real newer date.
> > > Well, theres no history per-se in the tablet, however things are stored
> > > based on the time their installation completed. So you could easily
> > > compare that timestamp with whenever the last system-update occurred.
> >
> > I think I need to clarify:
> >
> > spell imaginary-spell is added with version 1 in the distant past
> >
> > spell imaginary-spell is updated with a security patch on day 1 to devel
> >
> > on day 7 test grimoire is re-downloaded on computer-foo.
> >
> > on day 7 imaginary-spell is recompiled via a sorcery rebuild
> > (or maybe an update from version 0)
> >
> > spell imaginary-spell is updated on day 14 to test with security patch
> >
> > ***problem cause*** The UPDATED value is listed as "day 1"
> >
> > on day 20, computer-foo is updated.
> >
> > with UPDATED listed as day 1 on imaginary-spell and compile date is listed
> > as day 7, imaginary-spell's security update is not triggered for
> > recompile
> >
> > I'm just hoping for some way to avoid that. This also happens when
> > scribe update is run a time significantly before an update, not just
> > when we have lagged updates to the grimoire.
>
> Didn't you think about a PATCHLEVEL variable some time ago that was used
> for updates instead of the date based UPDATED? I think that would solve
> this problem easily.

Yeah, I did. I also thought there might be a solution that didn't
require any additional guru overheard, but the patchlevel variable might
be what we should use, instead.

Are we prepared to support PATCHLEVEL instead of UPDATED? IF we use
PATCHLEVEL, UPDATED (or a new PATCHRECOMPILE variable) should probably be
changed to match the patchlevel that is a suggested recompile instead of
being the date.

patchlevel 1 - new features to spell but not a change in compile
patchlevel 2 patchrecompile 2 - should recompile
patchlevel 3 patchrecompile still 2 - should recompile only if
patchlevel was less than 2

We should have an implied default patchlevel of 0 so the variable(s)
is(are)n't needed.

Sound good?

Seth
--
Seth Alan Woolley [seth at positivism.org], SPAM/UCE is unauthorized
Key id EF10E21A = 36AD 8A92 8499 8439 E6A8 3724 D437 AF5D EF10 E21A
http://smgl.positivism.org:11371/pks/lookup?op=get&search=0xEF10E21A
Security Team Leader Source Mage GNU/Linux http://www.sourcemage.org

Attachment: pgpC42_Q_LDmi.pgp
Description: PGP signature




Archive powered by MHonArc 2.6.24.

Top of Page