[PATCH] CTDB log message improvements

[Martin Schwenke]

On Tue, 5 Jul 2016 14:57:36 +1000, Martin Schwenke <martin at meltin.net>
wrote:

> On Tue, 5 Jul 2016 14:24:21 +1000, Amitay Isaacs <amitay at gmail.com>
> wrote:
> 
> > On Tue, Jul 5, 2016 at 2:14 PM, Richard Sharpe <realrichardsharpe at gmail.com>
> > wrote:
> >   
> > > On Mon, Jul 4, 2016 at 12:00 AM, Amitay Isaacs <amitay at gmail.com> wrote:    
>  [...]  
> > >
> > > I am confused by the wording in this hunk:
> > >
> > > @@ -1606,12 +1606,14 @@ int32_t ctdb_control_set_recmaster(struct
> > > ctdb_context *
> > > ctdb, uint32_t opcode, T
> > >
> > >   if (ctdb->pnn != new_recmaster && ctdb->recovery_master == ctdb->pnn) {
> > >   DEBUG(DEBUG_NOTICE,
> > > -      ("This node (%u) is no longer the recovery master\n", ctdb    
>  [...]  
> > > +      ("Local node %u is now the recovery master\n",
> > > +       new_recmaster));
> > >   }
> > >
> > >   if (ctdb->pnn == new_recmaster && ctdb->recovery_master != new_recmaster
> > > ) {
> > >   DEBUG(DEBUG_NOTICE,
> > > -      ("This node (%u) is now the recovery master\n", ctdb->pnn)
> > > );
> > > +      ("Remote node %u is now the recovery master\n",
> > > +       ctdb->pnn));
> > >   }
> > >
> > >   ctdb->recovery_master = new_recmaster;
> > >
> > > What local node are you referring to? The one the log is printed on?
> > >
> > > Wouldn't it be better to say: "This node (%u) is now the recovery master"?
> > >    
> > 
> > Oops.  I switched the two log messages around.
> > 
> > "Local" and "Remote" were late additions just to distinguish between the
> > two cases.  The original message sounds much better.
> > 
> > Here are the updated patches.  
> 
> Yeah, that's better...  :-)
> 
> Reviewed-by: Martin Schwenke <martin at meltin.net>
> 
> If Richard doesn't chime in soon then I'll push...

Pushed, along with "CTDB client code fixes"...

peace & happiness,
martin