Well ... I am assuming that the order of importance is ...
1. PCP works
2. non-root execution is goodness
3. lintian compliance
So I think there is no option other than finding a way to smack lintian.
I battled for 6 hours trying to figure out why qa/255 was failing, and I
would be very surprised to find an alternative solution that does not
involve making the mode for $PCP_LOG_DIR/pmcd more permissive.
If it is not 1777, then I suspect it could possibly be 755 and _owned_
by the user pcp ... I changed it by hand post-install to be thus and
pmcd starts (with a bunch of root owned log files therein) and qa/255
passes. Would that be any less sucky for lintian and any of the other
packaging pixies?
I cannot test on fedora/suse/centos as I am still stuck there with the
python rpm packaging failures I mentioned on 26 Nov.
On Thu, 2012-12-13 at 18:17 -0500, Nathan Scott wrote:
>
> ----- Original Message -----
> > Changes committed to git://oss.sgi.com/kenj/pcp.git dev
> > ...
> > commit 41552a2fe0407f2cfeff34c59aa0b1a67f8ca652
> > Author: Ken McDonell <kenj@xxxxxxxxxxxxxxxx>
> > Date: Fri Dec 14 06:07:14 2012 +1100
> >
> > Make $PCP_LOG_DIR/pmcd mode 1777
> >
> > Since the non-root pmcd changes, we can have log files for PMDAs
> > created
> > by either "root" or the user "pcp". To allow a PMDA launched
> > from pmcd's
> > SIGHUP handling (PMDA install or restarting a failed PMDA) to be
> > able to
> > create their own log files, the mode of $PCP_LOG_DIR/pmcd has to
> > be less
> > restrictive ... mode 1777 follows the $PCP_TMP_DIR model.
>
> This will make lintian complain once more. I think we can override it,
> probably, but once we do we should consider whether making private temp
> dirs on Debian is worthwhile once more (since lintian was the reason we
> reverted that, and if we have to override it anyway...).
>
> cheers.
>
> --
> Nathan
|