Emanuele,
I've now fixed this a number of times locally by hand editing the
/etc/logrotate.d/lcfg-se file, and every time I update the system, it's back
again.
Any chance of someone looking into a proper fix?
Martin.
--
-------------------------------------------------------
Martin Bly | +44 1235 446981 | [log in to unmask]
Systems Admin, Tier 1/A Service, RAL PPD CSG
-------------------------------------------------------
> -----Original Message-----
> From: Bly, MJ (Martin)
> Sent: Tuesday, November 04, 2003 3:33 PM
> To: 'LHC Computer Grid - Rollout'
> Subject: RE: SE log rotate issue - /etc/logrotate.d/lcfg-se
>
>
> Hi All,
>
> Since we updated to 1_1_0, this problem is back again. I
> believe there is a malconfiguration somewhere in our SE
> config but don't know where. I'm tempted to apply the old
> fix (edit it by hand) but does anyone know why the problem
> has recurred and what the proper fix is please?
>
> Martin.
> --
> -------------------------------------------------------
> Martin Bly | +44 1235 446981 | [log in to unmask]
> Systems Admin, Tier 1/A Service, RAL PPD CSG
> -------------------------------------------------------
>
> > -----Original Message-----
> > From: Bly, MJ (Martin)
> > Sent: Wednesday, September 10, 2003 12:05 PM
> > To: 'LHC Computer Grid - Rollout'
> > Subject: SE log rotate issue - /etc/logrotate.d/lcfg-se
> >
> >
> > Dear All,
> >
> > Our SE has been giving a log rotation error message that
> > appear to be caused by the misconfiguration of one of the
> > logrotate scripts:
> >
> > /etc/logrotate.d/lcfg-se looks like this:
> >
> > <%ifdef: ng_extralogs%><%for: _log=<%ng_extralogs%>%><%\%>
> > /var/obj/log/<%_COMP%>.<%_log%> <%\%>
> > <%end:%> <%end:%><%\%>
> > /var/obj/log/<%_COMP%> {
> > <%ifdef: ng_logrotate%><%for: _line=<%ng_logrotate%>%><%\%>
> > <%ng_logrotate_<%_line%>%>
> > <%end:%><%end:%><%\%>
> > postrotate
> > /etc/obj/<%_COMP%> logrotate -q
> > endscript
> > }
> >
> > rather than similar to this /etc/logrotate.d/lcfg-seinfo :
> >
> > /var/obj/log/seinfo {
> > postrotate
> > /etc/obj/seinfo logrotate -q
> > endscript
> > }
> >
> > It would be nice out find out how it got like that but at
> > this range (weeks) that's unlikely. Can I just replace it
> > with one in the same style as the seinfo version or does it
> > need to be different?
> >
> > Martin.
> > --
> > -------------------------------------------------------
> > Martin Bly | +44 1235 446981 | [log in to unmask]
> > Systems Admin, Tier 1/A Service, RAL PPD CSG
> > -------------------------------------------------------
> >
>
|