Re: [Tracker] What invokes writeback actions and what process carries out these actions

[Philip Van Hoof <philip codeminded be>]

On Sun, 2011-08-21 at 14:20 +0300, Adrien Bustany wrote:
> Le Sun, 21 Aug 2011 11:29:51 +1200,
> James Hurford <terrasea gmail com> a Ãcrit :
>
> > Hi
> >
> > Don't worry I answered my own question, updates are what trigger it
> > and only from updates that are not from miners.  Tracker store is
> > responsible for carrying out this action.
> >
> > If I got this wrong I would be grateful if you would let me know
> >
> > Thanks
>
> Actually, updates from miner would also trigger writeback, but the
> miner calls a method called IgnoreNextUpdate (iirc) to avoid a
> mining/writeback loop. That mechanism will probably be replaced in the
> future though, since it's racy.

No that's not entirely true.

The Writeback signal only happens in case the update happens in de
default GRAPH (when no graph is passed).

In both master and 0.10 branch doesn't writeback issue a call to
IgnoreNextUpdate either anymore (because it's indeed racy). This isn't
future but the current situation already :-)

A mining/writeback loop is avoided internally (because the FS miner
wraps the writeback call internally, it knows when a writeback happens
and when it's finished, allowing it to ignore writes to the file during
that time).

Regards,

Philip


-- 


Philip Van Hoof
freelance software developer
Codeminded BVBA - http://codeminded.be